*Article* **Optimizing Nitrogen and Seed Rate Combination for Improving Grain Yield and Nitrogen Uptake Efficiency in Winter Wheat**

**Hemat Mahmood 1,2, Jian Cai 1,\*, Qin Zhou 1, Xiao Wang 1, Allan Samo 1, Mei Huang 1, Tingbo Dai 1, Mohammad Shah Jahan <sup>3</sup> and Dong Jiang <sup>1</sup>**


**Abstract:** Nitrogen (N) supply and seed rate (SR) are two essential factors that affect the accumulation and partitioning of N and dry matter (DM) and, therefore, grain yield (GY) and N use efficiency (NUE). The objective of this experiment was to optimize N application and SR to regulate wheat growth and increase both GY and NUE. The results revealed that net photosynthetic rate (Pn), stomatal conductance (Gs), chlorophyll content, and activities of metabolic enzymes (NR and GS) significantly increased with increasing of N levels while decreasing SR. Plant tillers, GY, DM before anthesis, and N translocation, N agronomic efficiency (NAE), N recovery efficiency (NRE), and N uptake efficiency (NUPE) were highest in a combined treatment of N235 and SR180. However, N levels beyond 235 kg ha−<sup>1</sup> significantly decreased NAE, NRE, and NUPE. By increasing SR from 135 to 180 kg ha−<sup>1</sup> an increase of 12.9 % and 9.1% GY and NUPE, respectively, was observed. Based on this result, we estimate that 1 kg N ha−<sup>1</sup> might be replaced by an increase of approximately 0.6 kg ha−<sup>1</sup> SR. Our study suggested that using a combination of N and SR (N235 + SR180) could attain maximum GY and improve NUE parameters.

**Keywords:** high yield; nitrogen application; N use efficiency; seed rate; winter wheat

#### **1. Introduction**

Wheat is the main staple crop globally and plays a crucial role in challenging food security, with a total production of 736.1 million tons. Wheat grain yield is not only dependent on genetic potential (variety) and environmental constraints [1] but also depends on management practices [2,3]. Nitrogen is the essential nutrient for wheat growth and production [4,5] which is necessary for maintaining plant growth, biomass, and grain yield [6]. N deficiency in cereal crops reduces fertile tiller numbers [7–9], grain number, and kernel weight [10,11]. However, overuse of N results in environmental problems including N leaching, runoff, and volatilization [12], and reduces overall N use efficiency (NUE) [13]. In China, a rapid increase in wheat yield was in parallel with the dramatic use of N fertilizer since the 1950s [14]. Importantly, in the last 20 years, N input beyond the threshold level only caused prolonged yield improvement while severe environmental pollution [12,15,16]. Thus, the China government proposed the Double Reduction Plan [15]. A low dose of N with high NUE must be one of the main research goals in plant nutrition [16].

However, enhancing crop profitability and NUE simultaneously is necessary for sustainable agriculture [17] and it is a key challenge to improve viable agriculture in the next

**Citation:** Mahmood, H.; Cai, J.; Zhou, Q.; Wang, X.; Samo, A.; Huang, M.; Dai, T.; Jahan, M.S.; Jiang, D. Optimizing Nitrogen and Seed Rate Combination for Improving Grain Yield and Nitrogen Uptake Efficiency in Winter Wheat. *Plants* **2022**, *11*, 1745. https:// doi.org/10.3390/plants11131745

Academic Editors: Przemysław Barłóg, Jim Moir, Lukáš Hlisnikovský and Xinhua He

Received: 17 May 2022 Accepted: 24 June 2022 Published: 30 June 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

decades [18]. N fertilizer is often widely overused to obtain ideal productivity while the adjustment of SR is neglected, which usually synchronously improves productivity and NUE. When the N rate was decreased or cut, the output would be lost by reducing tillering and fertile tillers, grain number, and kernel weight [19]. The increase in SR could partially compensate for the decrease in fertile tillers and spike numbers and final productivity [20]. Increasing plant density from 135 to 405 plants m−<sup>2</sup> [21] or 75 to 300 plants m−<sup>2</sup> [11] significantly increased grain yield and other parameters. However, there must be an optimum SR to compensate for the negative effects of decreasing N for balanced high yields and improved NUE in wheat. Therefore, it is necessary to investigate the compensatory effect of increasing SR on the decreasing N input on wheat productivity and N use efficiency. It is also necessary to clarify their combination effects on the physiological and agronomical performance of wheat to reveal the underlying rules for balanced high grain yield and improved NUE in winter wheat.

A field experiment with different N and SR levels in two successive years was carried out to determine the optimum combination of N and the seed rate leading to improvement in grain yield and N use efficiency.

#### **2. Results**

#### *2.1. Physiological Traits*

#### 2.1.1. Photosynthetic Capacity, Chlorophyll Content, and Leaf Area Index

Pn, Gs, SPAD, and LAI were highest at the anthesis stage, followed by the jointing stage, 10 days after anthesis (10 DAA) and 20 DAA, respectively. N application and SR have significant effects on various physiological traits, viz., leaf photosynthetic capacity, chlorophyll content (SPAD value), as well as leaf area index (LAI) at all growth stages (Figures 1–3). Photosynthetic capacity (Pn), stomatal conductance (Gs), and SPAD were significantly increased by increasing the N rate or decreasing SR. A significant increment was observed as the N application rate increased from 0 to 235 kg ha<sup>−</sup>1. When N increased from 235 to 290 kg ha−1, there was no significant increase for Pn, Gs, and SPAD in both growing seasons. The Pn, Gs, and SPAD values were decreased significantly when SR increased from 135 to 225 kg ha−<sup>1</sup> in both growing seasons at all sampling stages. Moreover, LAI was significantly increased with an increase in both factors (N + SR). A significant effect was observed when N increased up to N235 treatment. There was no significant difference between the N235 and N290 treatments in LAI in both growing seasons (Figure 3). SR also significantly increased the LAI in both growing seasons from SR135 to SR180, beyond this level there was no significant effect.

#### 2.1.2. Enzymatic Activities of NR and GS

Nitrate reductase (NR) and glutamine synthesize (GS) play an essential role in N metabolism assimilation and regulation. NR and GS s enzyme activities were significantly increased by increasing N and decreasing SR (Figure 4). However, the activities of NR and GS differed at the various levels of N application. Furthermore, there was a significant variation in NR and GS activities between N0, N180, and N235, whereas there was no significant difference between the N235 and N290 treatments. The highest value of the NR and GS activities appeared as the combination of N290 and SR135 treatments (Figure 4).

**Figure 1.** Effects of the combination of N and SR on photosynthesis (Pn) and stomatal conductance (Gs) during four differnet growing stages in 2017–2018 and 2018–2019. Different letters represent significant differences in mean values of three replicate plots at *p* ≤ 0.05 levels according to Duncan's test.

**Figure 2.** Effects of N and SR on chlorophyll content (SPAD) during four differnet growing stages in 2017–2018 and 2018–2019. Different letters represent significant differences in mean values of three replicate plots at *p* ≤ 0.05 levels according to Duncan's test.

**Figure 3.** Effects of the combination of N and SR on the values of the leaf area index (LAI) during four differnet growing stages in 2017–2018 and 2018–2019. Different letters represent significant differences in mean values of three replicate plots at *p* ≤ 0.05 levels according to Duncan's test.

**Figure 4.** Combination effects of N and SR on nitrate reductase (NR) and glutamine synthesize (GS) activities during four differnet growing stages in 2018–2019. Different letters represent significant differences in mean values of three replicate plots at *p* ≤ 0.05 levels according to Duncan's test; \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively.

#### *2.2. Grain Yield and Related Agronomic Characteristics*

The grain yield (GY) was significantly influenced by N level and seed rate (N and SR), as well as by their interaction in both years (Table 1). With the increase in the N application rate the GY significantly increased. At SR of 180 kg ha−1, GY increased from 6.7 to 8.3 t ha−<sup>1</sup> when N application increased from N180 to N235 kg ha−<sup>1</sup> in 2018–2019. There was no significant difference between N290 and N235 kg ha−<sup>1</sup> in both growing seasons. SR significantly affected GY in both growing seasons. Under the same amount of N application (N235 kg ha<sup>−</sup>1), GY increased from 5 to 5.7 t ha−<sup>1</sup> in the first growing season and from 7.5 to 8.3 t ha−<sup>1</sup> in the second growing season (Table 1). The highest GY 8.3 t ha−<sup>1</sup> was obtained

with the combination of N235 + SR180 kg ha−<sup>1</sup> in 2018–2019. N and SR significantly affected the agronomic parameters viz., number of spikes (NS), 1000-grain weight (TGW), number of grains per spike, harvest index (HI), and plant height (PH). All agronomic parameters (NS, TGW, NGS, HI, and PH) were significantly increased with the increasing N application rate. Compared to the treatment with SR180 N0, NS was significantly increased by 13.2%, 23.8%, and 22.4 % and NGS was increased by 21.8%, 36.8%, and 37.8% in treatments SR180N180, SR180N235, and SR180N290, respectively, in the second growing season. Except for the first growing season in which PH was scarcely increased, TGW and HI were decreased when N application rate increased from N235 to N290. At the second growing season, none of the agronomic characteristics were significantly influenced beyond N235 kg ha<sup>−</sup>1.

**N kg ha**−**<sup>1</sup> SR kg ha**−**<sup>1</sup> TY t ha**−**<sup>1</sup> NS** × **104 ha**−**<sup>1</sup> TGW(g) NGS HI PH (cm) 2017–2018** N235 SR135 5.05 <sup>b</sup> 357 <sup>c</sup> 39.7 <sup>a</sup> 35.63 <sup>a</sup> 0.41 <sup>c</sup> 73.4 <sup>c</sup> SR180 5.78 <sup>a</sup> 414 <sup>b</sup> 39.5 ab 35.38 <sup>a</sup> 0.44 ab 74.33 bc SR225 5.7 <sup>a</sup> 425 ab 39.2 bc 34.27 <sup>b</sup> 0.45 <sup>a</sup> 75.74 ab N290 SR135 5.03 <sup>b</sup> 361 <sup>c</sup> 39.1 bc 35.63 <sup>a</sup> 0.41 <sup>c</sup> 76.43 ab SR180 5.77 <sup>a</sup> 420 ab 39 bc 35.3 <sup>a</sup> 0.43 bc 77.23 <sup>a</sup> SR225 5.77 <sup>a</sup> 434 <sup>a</sup> 38.9 <sup>c</sup> 34.18 <sup>b</sup> 0.42 bc 77.57 <sup>a</sup> F-Value N 0.113 2.475 13.65 \*\* 0.098 7.478 \* 19.88 \*\* SR 107.1 \*\* 105.9 \*\* 3.38 \* 21.65 \*\* 11.46 \*\* 2.999 N\*S 0.331 0.123 0.63 0.025 1.637 0.426 N0 SR135 4 <sup>g</sup> 391 <sup>g</sup> 35.7 bc 28.8 <sup>d</sup> 0.35 <sup>e</sup> 62.93 <sup>d</sup> SR180 4.8 <sup>f</sup> 479 ef 35.6 bc 28.1 <sup>d</sup> 0.4 cde 63.62 <sup>d</sup> SR225 4.8 <sup>f</sup> 503 de 35.5 <sup>c</sup> 27.2 <sup>d</sup> 0.4 cd 64.07 <sup>d</sup> 2018–2019 N180 SR135 5.6 <sup>e</sup> 448 <sup>f</sup> 36.4 ab 34.5 <sup>c</sup> 0.38 de 72.87 <sup>c</sup> SR180 6.7 <sup>d</sup> 543 bc 36.3 abc 34.1 <sup>c</sup> 0.41 cd 74.02 bc SR225 6.6 <sup>d</sup> 555 <sup>b</sup> 36.3 abc 33.1 <sup>c</sup> 0.42 bc 74.47 <sup>b</sup> N235 SR135 7.5 <sup>c</sup> 518 cd 36.7 <sup>a</sup> 39.6 ab 0.45 ab 73.97 bc SR180 8.3 <sup>a</sup> 593 <sup>a</sup> 36.7 <sup>a</sup> 38.3 ab 0.46 <sup>a</sup> 75.15 ab SR225 8.2 <sup>a</sup> 597 <sup>a</sup> 36.5 ab 37.5 <sup>b</sup> 0.46 <sup>a</sup> 75.75 <sup>a</sup> N290 SR135 7.8 bc 529 bc 36.6 <sup>a</sup> 40.4 <sup>a</sup> 0.45 ab 73.95 bc SR180 8.2 <sup>a</sup> 587 <sup>a</sup> 36.4 ab 38.6 ab 0.44 ab 75.22 ab SR225 8.1 ab 591 <sup>a</sup> 36.4 ab 37.7 ab 0.45 ab 75.97 <sup>a</sup> F-Value N 690 \*\* 74.4 \*\* 9.66 \*\* 242.5 \*\* 43.55 \*\* 598.3 \*\* SR 60.1 \*\* 83.94 \*\* 0.66 12.19 \*\* 4.48 \* 17.8 \*\* N\*SR 2.64 \* 1.36 0.027 0.394 1.77 0.25

**Table 1.** Effects of N and SR on GY and agronomic characteristics in 2017–2018 and 2018–2019.

Note: N, SR, TY, NS, TGW, NGS, HI, and PH indicate nitrogen application, seed rate, theoretical yield number of spikes, thousand-grain weight, the number of grains per spike, harvest index, and plant height, respectively. Different letters in the same column represent significant differences in mean values of three replicate plots according to Duncan's test; \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively.

SR significantly influenced all of the agronomic parameters. Increasing SR significantly increased NS, HI, and PH during both growing seasons and decreased NGS and TGW. As an example, compared with the N235SR135 treatment, NS was significantly increased by 10.8% and 11.6%, while NGS was decreased by 4.5% and 6.7% in treatments N235SR180 and N235SR225, respectively, in the second growing season (Table 1).

Increasing the application of N increased the fodder part and the yield of wheat. Increasing the seed rate can increase the spike number and compensate for reducing the N application rate. Linear regression was used to assess replacing N with SR for balancing GY and NUE parameters. According to the equations obtained from linear regression, the increase in grain yield by 1 ton ha−<sup>1</sup> needs to increase the seed rate by 7669 kg ha−<sup>1</sup> or increase the fertilization with N by 12,807 kg ha−1. This means that 7669 kg ha−<sup>1</sup> seed rate is equivalent to 12,807 kg N; therefore, 0.598 kg ha−<sup>1</sup> seed rate would approximately replace 1 kg of N ha−<sup>1</sup> (Figure 5). Based on our result, increasing SR from 135 to 180 kg ha−<sup>1</sup> was observed to increase by 12.9% and 9.1% for GY and NUPE, respectively.

**Figure 5.** Estimating the regression line to compare N and seed rate for balancing grain yield and improving NUE. \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively. The dots represent the mean value of grain yield under each seed rate (left) or nitrogen rate (right).

#### *2.3. Accumulation, Translation, and Partitioning of DM*

DM accumulation (DMA) was significantly affected by the rate of N application and SR (Table 2). Increasing the rate of application of N significantly increased DMA in all growth stages. The highest amount of DMA appeared during the jointing to the anthesis stage. SR significantly increased DMA at all growth stages. The significant effects of SR were up to 180 kg ha−1, and beyond SR180 kg ha−<sup>1</sup> was not further influenced (Table 2). Similarly, when N fertilizer increased beyond N235 kg ha<sup>−</sup>1, the values of DMA were not significantly increased. At maturity, the highest increase in DM was found with N235SR180 treatment.

**Table 2.** Combination effects of N and SR on DM accumulation, translocation in 2018–2019.


Note: SO-JT, JT-An, An-M, SO–M, PTA, CPT, PAA, and CPA represent sowing to jointing, jointing to anthesis, anthesis to maturity, sowing to maturity, pre-anthesis DM translocation amount, contribution of pre-anthesis translocation to grain, post-anthesis accumulation amount, and contribution of post-anthesis DM accumulation to grain, respectively. Different letters in the same column represent significant differences in mean values of three replicate plots according to Duncan's test; \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively.

The translocation and contribution of DM were significantly affected by the application rate of N and SR (Table 2). With increasing N rate, pre-anthesis translocation (PTA), post-anthesis accumulation (PAA), and contribution of post-anthesis to grain (CPA) were significantly increased, while the contribution of pre-anthesis translocation to grain (CPT) was significantly decreased. The maximum value of PTA appeared in the N235 treatment, which was significantly higher than the N290 treatment. Furthermore, the value of PTA increased with increasing SR, while PAA and CPA significantly increased up to SR180 kg ha<sup>−</sup>1. The CPT values first decreased and then increased as the SR increased.

The partition of DM into different parts of the plant differed between the N and SR treatments (Table 3). At the anthesis stage, the DM of culm + sheath was higher than the DM of rachis + glumes and the DM of rachis + glumes was higher than the DM of the leaves. However, at the harvesting stage, the grain DM was higher than the DM of the culm + sheath, the DM of culms + sheaths was higher than the DM of rachis + glumes, and the DM of the rachis + glumes was higher than the DM of leaves. The proportion of distribution of DM of grains, rachis + glumes, culms + sheathes, and leaves ranged from 37.1% to 45.8%, 12.7% to 14.1%, 33% to 41.4%, and 7.8% to 8.5%, respectively, at harvesting.

**Table 3.** Effects of N and SR on DM accumulation and partitioning at the anthesis and maturity stages in 2018–2019.


Note: N and SR represent nitrogen application and seed rate, respectively. Different letters in the same column represent significant differences of mean values of three replicate plots at *p* ≤ 0.05 levels according to Duncan s test; \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively.

#### *2.4. Accumulation, Translocation, and Partitioning of N*

Accumulation of N (NA) at all parts of the plant was significantly increased by increasing N application and SR up to certain levels (N235, SR180) at both anthesis and maturity stages. At the maturity stage, the total content of N compared to control (N0) treatment increased by 76.9%, 134%, and 139% in the treatments N180, N235, and N290. The N content compared to SR135 was increased by 8.8% and 5% in the SR180 and SR225 treatments (Table 4). As well, N translocation before anthesis to grain (NTA), N accumulation after anthesis (NAA), and the contribution rate of NA after anthesis to grain (CAG) were significantly increased with the increase in N application. NTA increased 32.18% and 2.9% when N increased from N180 to N235 from N235 to N290, respectively (average of three SR treatments). Furthermore, the contribution rate of pre-N translocation to grain (CTG) had the same trend as CPT. Furthermore, by increase in SR up to 180 kg ha<sup>−</sup>1, NTA, NAA, and CAG were significantly increased, while CTG was at first significantly decreased then increased (Table 4).


**Table 4.** Combination effects of N and SR on total N accumulation and translocation in 2018–2019.

Note: SO-JT, JT-An, An-M, SO–M, NTA, CTG, NAA, and CAG represent jointing, jointing to anthesis, anthesis to maturity, sowing to maturity, pre-anthesis N translocation, contribution rate of N translocation to grain, nitrogen accumulation amount, and contribution rate of N accumulation to grain, respectively. Different letters in the same column represent significant differences in mean values of three replicate plots according to Duncan's test; \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively.

Plant N partitioning was also influenced by N and SR. Compared with the control treatment (N0), the N180, N235, and N290 treatments increased the N content by 53.8%, 153%, and 144% in the rachis + glumes, by 84.2%, 149.2%, and 157.6% in the culms + sheathes, and by 65.8%, 107.7%, and 109.4% in the part of leaves, respectively, at anthesis stage. However, this increase in N at the harvesting stage was by 78.7%, 140.3%, and 145.6% for grains, by 52.9%, 98.5%, and 103% for rachis + glumes, by 61.5%, 88.1%, and 99% for culms + sheathes, and by 113.7%, 182%, and 186% for leaves in treatments N180, N235, and N290 compared to N0 treatment, respectively. Additionally, compared to the SR135 treatment, the SR180 and SR225 treatments increased N content by 8.6% and 1% at the rachis + glumes, by 1.7% and 2% at the culms + sheathes, and by 16% and 3.7% at the parts of leaves, respectively, at anthesis stage. Compared to SR135 treatment, the N content of the SR180 and SR225 treatments increased by 9.8% and 6.2% in grains, by 5.5% and 1.8% in the rachis + glumes, by 5.1% and −2% in the culms + sheathes, and by 5.5% and 2.7% in the leaves, respectively, at the maturity stage (Table 5). The range of the N distribution ratio of different parts, i.e., rachis + glumes, culms + sheathes, and leaves were from 17% to 21.9%, from 39.1% to 47.1%, and from 33.5% to 41.6% at anthesis, respectively. The maturity stage range of N distribution range was 75.4% to 79.2%, from 5.8% to 7.4%, 8.8% to 11.9%, and from 5% to 6.4% for grains, rachis + glumes, culms + sheathes, and leaves, respectively (Table 5).

**Table 5.** Combination effects of N and SR on N partitioning during anthesis and maturity stages in 2018–2019.



**Table 5.** *Cont.*

Note: N, SR, and NA represent nitrogen application, seed rate, and N accumulation, respectively. Different letters in the same column represent significant differences in mean values of three replicate plots according to Duncan s test; \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively. Revision as above.

#### *2.5. N Use Efficiency (NUE) Parameters*

N rates, SR, and their interaction had a significant effect on N agronomy efficiency (NAE), N uptake efficiency (NUPE), and N partial factor productivity (NPFP). Increasing N level up to N235 kg ha−1, NAE, NUPE, N recovery efficiency (NRE), and N harvest index (NHI) were significantly increased. However, increasing N levels beyond N235 kg ha−1, NAE, NRE, and NUPE were significantly decreased. The NPFP values were decreased at all N levels. Furthermore, the NAE, NRE, NUPE, NPFP, and NHI values decreased by 12.8%, 10.4%, 17.3%, 13.6%, and 0.4%, respectively, at N290 treatment compared to N235 treatment. Similarly, SR had a considerable effect on the NUE parameters, but the effect of SR was less compared to the N treatment. Maximum values for the parameters NAE, NRE, and NUPE were observed from the combination of treatment with N235 and SR180 (Table 6).

**Table 6.** Combination effects of N and SR on N use efficiency parameters in 2018–2019.


Note: N, SR, NAE, NRE, NUPE, NPFP, and NHI represent nitrogen rate, seed rate, nitrogen agronomy efficiency, N recovery efficiency, N uptake efficiency, N partial factor productivity, and N harvest index, respectively. Different letters in the same column represent significant differences in mean values of three replicate plots according to Duncan s test; \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively.

#### *2.6. Correlation Analysis*

2.6.1. Correlation of GY with Agronomic and Photosynthesis Traits

The key relationships between the GY-related parameter variables are shown in Table 7. There was a significant positive correlation between GY and NS, NGS, TGW, PH, HI, Pn, Gs, SPAD value, and LAI. However, there was no significant relationship between GY and PH (Table 7).


**Table 7.** Correlation analysis of grain yield with agronomic and photosynthesis traits.

Note: GY: grain yield; NS: number of spikes; NGS: number of grains per spike; TGW: 1000 grain weight per gram; PH: plant height (cm); HI: harvest index; Pn: net photosynthesis; Gs: stomatal conductance; SPAD: leaf greenness; LAI: leaf area index. \* and \*\* mean significantly correlation at \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively.

#### 2.6.2. Relationship between GY and Enzyme Activities and NUE Parameters

Grain yield had a significant and positive relationship with nitrogen reductase (NR) and glutamine synthesis (GS) enzymes, N agronomic efficiency (NAE), N recovery efficiency (NRE), and N uptake efficiency (NUPE). However, GY had a significantly negative relationship with N translocation's contribution before anthesis to grain (CTG) (Figure 6).

Increasing N application from 235 to 290 kg ha−1, GY did not increase significantly (0.4%) while NAE, NRE, and NUPE were decreased 12.8%, 10.4%, and 17.3%, respectively. Maximum GY and highest values of NUE parameters, particularly NAE, NRE, and NUPE, were observed from the combination of treatment with N235 and SR180 (Tables 1 and 6).

**Figure 6.** Regression analyses among GY and nitrate reductase (NR)/(**A**) glutamine synthesis (GS)/(P**B**), N agronomy efficiency (NAE)/(**C**), N recovery efficiency (NRE)/(**D**), N uptake efficiency (NUPE)/(**E**), and contribution of N translocation to the grain after the anthesis stage (CTG)/(**F**), respectively. \* *p* ≤ 0.05 and \*\* *p* ≤ 0.001, respectively.

#### **3. Discussion**

DM and N translocation are well-known to greatly contribute to the final GY [22,23]. However, there is a lack of research on the combined effects of N and SR on the DM and N translocation, NUE parameters, growth physiological parameters, and their relationship with GY in winter wheat. The effects of the excessive rate of N and N s compensation by increasing SR on final GY and NUE parameters were unknown.

#### *3.1. Physiological Characteristics*

The plant leaf s photosynthesis capacity plays a crucial role in plant growth and grain yield [24], and approximately 70% of productivity is derived from post-anthesis photosynthesis. In the present study, increasing N application and decreasing SR, resulted in a significantly increased Pn, Gs, and SPAD values in both growing seasons. Furthermore, LAI was significantly increased by increasing N application and SR (Figures 1–3). However, it has shown that the maximum values for Pn, Gs, and SPAD parameters were exhibited from the combination of N290 + SR135 treatment but there were no significant differences between the mentioned and N235 + SR180 treatment. The main reason for decreasing the Pn, Gs, and SPAD values by increasing SR might be due to more competition and the overcrowded shading effect as reported by [25]. In addition, N reductase (NR) and glutamine synthesize (GS) increased significantly with the N application rate and the declaration of SR. Notably, there was no significant difference for the value of both mentioned enzyme activities when N fertilizer amount increased from N235 to N290. These findings suggest N and SR s optimization benefits for maintaining strong photosynthesis capacity and N assimilation ability in wheat plants.

#### *3.2. Grain Yield (GY) and N Use Efficiency (NUE)*

Simultaneous improvement in GY and NUE of wheat is an important objective in modern agriculture management. Here, we investigated the suitable combinations of N and SR to obtain higher GY and NUE. The maximum value was observed with the treatment of N235 + SR180 kg ha−<sup>1</sup> in both growing seasons. The previous finding can explain that too high plant density had no significant effect on wheat grain yield [11,26]. According to our results, yield loss caused by reducing the N rate can be compensated by increasing SR (Figure 2). It was estimated by a linear regression that every decrement of 1 kg N ha−<sup>1</sup> can be replaced by adding 0.6 kg ha−<sup>1</sup> SR. The GY obtained by adding SR is clearly attributed to the increasing spike number (SN) (Table 1). Moreover, the yield in 2017–2018 (Y1) was much lower than that in 2018–2019 (Y2), which may be due to the adverse weather conditions. There was excessive rainfall and less sunshine during grain filling in the first growing season (Figure 1), leading to lower yield components and GY [1].

NUE results from the incorporation of N-uptake efficiency (NUPE) and N-utilization efficiency (NUTE) [27]. In detail, NUPE the plant's capacity to extract N from the soil, and depending on the root structure and the relation of N transporters [28]. In this study, increasing N application from N235 to N290 resulted in a decrease in NAE, NRE, NUPE, and NPFP by 12.8%, 10.4%, 17.3%, and 13.6%, respectively. Similarly, a result was obtained by [29] that treatment with N240 and N300 compared to treatment with N180, NPFP were decreased by 24.5% and 37.4% and NAE were decreased by 23.5% and 31.9%, respectively. The decrease in NUE after the optimal rate might be due to more losses by increasing N application according to the previous finding of [11]. Here, our results further confirmed that NUE components were significantly increased up to a certain amount of SR (180 kg ha<sup>−</sup>1), which is in good agreement with the previous study [30]. This increase in the NUE response to the high SR could be due to an increase in the density of the roots in the soil, which enhanced N from deeper parts of the soil. Therefore, it is not surprising that maximum values of NAE, NRE, and NUPE were also observed from the optimal combined treatment (N235 + SR180), which was also the case for grain yield (Table 2). Our study thereby provides a practical management method approaching higher GY and NUPE by the optimization of N and SR.

#### *3.3. Accumulation and Translocation of DM and N*

Total dry matter accumulation and partitioning into separate parts of the plant were significantly affected by combined N and SR s combined treatments. The maximum value of total dry matter and individual parts especially grains, rachis + glumes, and leaves, was also obtained from the combination of N235 + SR180 treatment. Recent studies reported that the contribution of pre-anthesis translocation to grain was significantly decreased when the N application rate increased [31,32]. Similarly, the result of the current experiment showed that the contribution of pre-anthesis DM translocation was significantly decreased with increasing N rate, while pre-anthesis DM translocation, post-anthesis DM accumulation, and contribution of post-anthesis DM accumulation to grain were significantly increased. The main reason for the decreasing contribution of pre-anthesis DM translocation to grain might be that early senescence occurred due to N deficiency, which would speed up the pre-translocation from leaf and stem to spike. Furthermore, increasing seed rate significantly increased pre-anthesis dry matter translocation, postanthesis DM accumulation, and contribution of post-anthesis DM accumulation to the grain. Parallel to our finding, it was reported that post-heading DM and N accumulation was significantly increased with increasing SR [33]. It should be noted that an excessive amount of N application (N290 kg ha<sup>−</sup>1) as well as SR (SR225 kg ha<sup>−</sup>1) did not increase the amount of DM translocation.

Total N accumulation, partitioning, and translocation showed the same trend with the above part of DM. Both N and SR up to optimal levels (N235, SR180) significantly increased the total N content of individual parts. The same finding reported that no further increase was observed in the uptake of N at N fertilizer and the density of the plant beyond 240 kg N ha−<sup>1</sup> and 405 plants m−<sup>2</sup> [21,34]. N translocation, postanthesis N accumulation, and postanthesis N accumulation contribution of post-anthesis N accumulation to the grain were higher in the high N treatments compared to control and low N treatment. A similar result was found that N translocation and post-anthesis N accumulation were enhanced with increasing N application rate [23]. In the current experiment, N translocation, N accumulation after anthesis, and contribution of post-anthesis to grain response to seed rate were significantly increased from SR <sup>135</sup> to SR <sup>180</sup> kg ha<sup>−</sup>1.

#### *3.4. Relationship of GY with Related Parameters*

Our results showed that GY has a significant and positive correlation with Pn, Gs, SPAD value, LAI, and other GY components (Table 7). This is similar to the results from the study by Jiang et al. [24]. Furthermore, the regression analyses also revealed that GY had a positive correlation with NR, GS, NAE, NRE, and NUPE while showing a negative correlation with CTG (Figure 6). It was also observed that NR and GS activities were highly positively correlated with photosynthesis capacity, which is consistent with previous studies [35,36]. Here, we found that GY had a significant positive correlation with NUE parameters such as NAE, NRE, and NUPE. This was not in agreement with the previous finding that GY showed a negative correlation with NUE [37,38]. The reason might be due to a certain amount of N + SR (N235 + SR180), in which both NUE and GY were significantly higher up to the previous partnership. In conclusion, we found that N and SR s improper rate cannot increase GY, but significantly decreased NUE. In this regard, to achieve the maximum GY and NUE, it would be better to use the optimal amount of both N and SR, which is the result of the current experiment, and the suitable combined treatment was N235 + SR180. Furthermore, by using a suitable combination of N and SR (SR180 N235), replacing N to SR especially for balancing GY and NUE would be the best method for sustainable agriculture. According to our findings, we infer that, based on low SR (SR135), 1 kg ha−<sup>1</sup> N could be saved by increasing approximately 0.6 kg ha−<sup>1</sup> SR.

#### **4. Materials and Methods**

#### *4.1. Plant Material and Experimental Site*

The field experiment was conducted during two successive growing seasons (2017–18 and 2018–19) at the XuYi Rice and Wheat demonstration center (118◦43 N and latitude 32◦59 E) in Jiangsu province, China. The winter wheat cultivar Ningmai 13 was used in both growing seasons. The soil type was clay loam and the pH was 6.8. It contained 31.07 g kg−<sup>1</sup> of organic matter, 2449 g kg−<sup>1</sup> of available N, 27.3 mg kg−<sup>1</sup> of available phosphate, and 240 mg kg−<sup>1</sup> of available potassium. Seeds were sown on 31 October 2017, and 1 November 2018, and the crop was harvested on 3 June 2018, and 6 June 2019, respectively.

#### *4.2. Experiment Design*

The experiment was carried out according to the split-plot design with three replicates. The main plot consisted of three seed rates (SR135, SR180, and SR225 kg ha−1) and the subplot comprised two levels of doses of N in the first year (N235 and N290 kg ha−1) and three N levels in second year of the experiment (N180, N235, and N290 kg ha<sup>−</sup>1). At the first growing season, there was no significant effect between N235 and N290 for GY, because of that we added N180 kg ha−<sup>1</sup> treatment at the second growing season to determine the N effect as well to determine the optimum rate of N for GY. An N-control plot (N0) was also used at the second growing season per replication for the calculation of NUE parameters. N fertilizer was applied as urea (46%), phosphorus (P) and potassium (K) fertilizer as calcium superphosphate (15%), and potassium chloride (60%) at the rates of 120 (P2O5) and 120 (K2O) kg ha−1. All of the phosphorus and potassium fertilizer and 70% of the total amount of N fertilizer were spread by hand before plowing at the time of sowing. The remaining N fertilizer (30%) was applied at the first node (31) according to BBCH. Each plot size was 4 × 3 m and consisted of 12 rows with a row-to-row distance of 25 cm.

In this experiment, the rice–wheat rotation system was undertaken for the long term. Rice cultivation techniques such as puddling, transplanting, and flooding, and the whole amount of straw returned to the field almost in the last decade. Only 5 cm of rice straw remains above the ground. Moldboard plough was followed by rotary plough as primary and secondly tillage. The depth of moldboard tillage was 20 cm and that of the rotary plough was 10 cm. Wheat seeds were sown with a seed drill precise machine with surface stubble plowing and roll compaction. For high-yield production, insects, diseases, and weeds were controlled two times by spraying insecticide (Biscaya), fungicide (Capalo), and herbicide (sulfosulfuron) during both growing seasons.

#### *4.3. Grain Yield and Yield Components*

Uniform plants at the flowering stage were tagged with labels and were sampled at a later stage. At the maturity stage, ears/spikes f rom the area of 0.5 m2 (without taking a sample) of each plot were collected to determine GY and yield components.

#### *4.4. Photosynthesis, SPAD, LAI, N, and Enzyme Activities in Leaves*

Photosynthesis (Pn), stomatal conductance (Gs), SPAD, and LAI were measured at the jointing stage, anthesis, 10 days after anthesis (DAA), and 20 DAA. Pn and GS were measured by a portable gas exchange analyzer (LI-6400XT;LI-COR-Inc., Lincoln NE, USA) at 9:00–11:00 a.m. on a sunny day. The concentration of CO2 in the leaf chamber, light intensity, and relative humidity were set as 380 μmol mol<sup>−</sup>1, 1000–1100 μmol m<sup>−</sup>2s−1, and 500 mL min <sup>−</sup>1, respectively. The SPAD value was determined by Minolta 502 chlorophyll meter (Minolta, Japan). LAI was measured with using a leaf area meter (LI-3100, LI-COR, Lincoln, NE, USA). N concentration was determined by the micro Kjeldahl method [39]. Nitrate reductase (NR) and glutamine synthesis (GS) were determined according to the method previously described by [36] and [40].

#### *4.5. Dry Matter (DM) and N Translocation*

DM and N accumulation (DMA and NA), translocation, and their contribution were estimated according to [31] and [41] by using the following equations (Table 8).

**Table 8.** Equations for estimating dry matter and nitrogen translocation.


#### *4.6. Use Efficiency (NUE) Parameters*

NUE parameters were determined using the following equations described by [17], and [42] (Table 9).



#### *4.7. Weather Condition*

Monthly average temperature, rainfall, and sunshine at the experimental site over two successive years (2017–2018 and 2018–2019) are presented in Figure 7. There was considerable variation between the two growing seasons. At the active tiller stage (from late January to the end of the first week of February), the minimum temperature in 2017–2018 was lower (−4.5 ◦C) compared to that of the second growing season (−0.5 ◦C). At the anthesis stage, the average rainfall in the first growing season was 538 mm, which was 64.02% higher than that in the second growing season (328 mm).

**Figure 7.** Metrological data: (monthly average temperature, rainfall (mm), and sunshine per hour) in two successive growing season.

#### *4.8. Statistical Analysis*

Two-way ANOVA (SPSS version 17.1) was used for analyzing the variance among different treatments. The means were tested with the least significant difference at the 0.05 probability level (*p* ≤ 0.05 by Duncan's). Pearson's correlation between grain yield and related parameters were calculated through the SPSS version17.1. All graphs and linear regression analyses was done by sigmaplot 14.0 software (Chicago, IL, USA).

#### **5. Conclusions**

In summary, the GY, DMA, NAC, NUE parameters and physiological parameters increased significantly with the combination of N235 and SR180. However, the excessive rate of N application cannot increase GY and other parts of plant DM but it decreased NAE, NRE, and NUPE. Our result confirmed that maximum GY and higher NUE components could be achieved via avoiding excessive use of N, and optimizing the compensation effect of increasing SR for reducing N application.

**Author Contributions:** M.H., Q.Z. and D.J. designed the experiments. M.H. analysed data, interpreted data, and wrote the original draft of the manuscript. M.H., A.S. and M.S.J. carried out the experiments, curated and analyzed data. M.H., J.C., X.W., Q.Z. and T.D. were involved in the management of the experiment. D.J. participated in supervision of the project. M.S.J., Q.Z. and D.J. participated in the critical reading and discussion of the manuscript, H.M. did formal analysis and methodology. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the projects of National Key Research and Development Program of China (2021YFF1000204, 2020YFE0202900), National Natural Science Foundation of China (32030076, U1803235, 31901458, 32172116), Jiangsu Agriculture Science and Technology Innovation Fund (JASTIF, CX(20)3086), Natural Science Foundation of Jiangsu Province (BK20190509), the China Agriculture Research System (CARS-03), Jiangsu Collaborative Innovation Center for Modern Crop Production (JCIC-MCP), Jiangsu 333 program, and the 111 Project (B16026). The first author is thankful to Chinese Government Scholarship Council for providing scholarship to pursue an Msc at Nanjing Agricultural University.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

<sup>1.</sup> Slafer, G.A.; Savin, R.; Sadras, V.O. Coarse and fine regulation of wheat yield components in response to genotype and environment. *Field Crops Res.* **2014**, *157*, 71–83. [CrossRef]


## *Article* **Agro-Physiological Traits of Kaffir Lime in Response to Pruning and Nitrogen Fertilizer under Mild Shading**

**Rahmat Budiarto 1,\*, Roedhy Poerwanto 2, Edi Santosa 2, Darda Efendi <sup>2</sup> and Andria Agusta <sup>3</sup>**


**Abstract:** Mild shading has been reported to increase leaf production in kaffir lime (*Citrus hystrix*) through the improvement of agro-physiological variables, such as growth, photosynthesis, and water-use efficiency; however, there is still a knowledge gap concerning its growth and yield after experiencing severe pruning in harvest season. Additionally, a specific nitrogen (N) recommendation for leaf-oriented kaffir lime is still unavailable due to its lesser popularity compared to fruit-oriented citrus. The present study determined the best pruning level and N dose based on agronomy and the physiology of kaffir lime under mild shading. Nine-month-old kaffir lime seedlings grafted to rangpur lime (*C. limonia*) were arranged in a split-plot design, i.e., N dose as a main plot and pruning as a subplot. Comparative analysis resulted in 20% higher growth and a 22% higher yield in the high-pruned plants by leaving 30 cm of main stem above the ground rather than short ones with a 10 cm main stem. Both correlation and regression analysis strongly highlighted the importance of N for leaf numbers. Plants treated with 0 and 10 g N plant−<sup>1</sup> experienced severe leaf chlorosis due to N deficiency, while those treated with 20 and 40 g N plant−<sup>1</sup> showed N sufficiency; thus, the efficient recommendation for kaffir lime leaf production is 20 g N plant<sup>−</sup>1.

**Keywords:** *Citrus hystrix*; chlorosis; leaf production; photosynthesis; shading

#### **1. Introduction**

Citrus is one of the leading, popular horticultural fruit commodities commonly used for fresh food and beverages [1–3]. Published biogeographic, genomic, and phylogenetic analyses determine the Southeast Himalayan foothills as the center of origin for most citrus species [4]. As one of the interesting *Citrus* taxa, lime is reported to have highly polymorphic characteristics, derived from four major *Citrus*, namely *C. medica*, *C. maxima*, *C. micrantha* and *C. reticulata*) [5]. In contrast, the kaffir lime is classified as a relatively minor citrus and apparently wild, native to central Malesia or the Southeast Asian region [6]. This lime is well-designed as a leaf-oriented target due to its aromatic leaves being used as spices in numerous Asian dishes [7–9]. Aside from its fragrance, the bifoliate characteristics of kaffir lime leaves can be used to differentiate this species from others [10,11]. The leaf of the kaffir lime is also famous for producing essential oils [12] that possess several bioactivities, such as antifeedant [13], antibacterial [14], and larvicide [15]. Due to its importance as a spice and an essential oil raw material, the demand for kaffir lime leaves is potentially increased. Thus, some effort is required to meet those needs. General strategies to boost plant production could employ both input intensification and land expansion [16].

Kaffir lime generally grows naturally or is planted in polyculture in the yard by residents with a non-intensive cultivation system [7,9]. In nature, kaffir lime typically grows from seed and shows a high tree appearance, long and large thorns, dense canopy and

**Citation:** Budiarto, R.; Poerwanto, R.; Santosa, E.; Efendi, D.; Agusta, A. Agro-Physiological Traits of Kaffir Lime in Response to Pruning and Nitrogen Fertilizer under Mild Shading. *Plants* **2023**, *12*, 1155. https://doi.org/10.3390/ plants12051155

Academic Editors: Przemysław Barłóg, Jim Moir, Lukáš Hlisnikovský and Xinhua He

Received: 5 February 2023 Revised: 24 February 2023 Accepted: 27 February 2023 Published: 3 March 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

branches, so leaf harvest is less effective and efficient. On the other hand, several local communities in Tulungagung, Indonesia have implemented monoculture and semi-intensive kaffir lime cultivation [9]. Previous research has also proven that modification of cultivation under low-level shading (light reduction of about 23%) can produce beneficial stress, increasing the growth rate and yield of kaffir lime leaves by 84 and 63%, respectively [17]. Although that increasing response has been reported in the first harvest period, further research is urgently needed to confirm yields in the following period.

Kaffir lime experiences heavy pruning during its harvesting season. Heavy pruning is carried out by cutting most of the canopy and leaving a portion of the main stem for successive growth in the next period [9]. The remaining part of the main stem should be studied further since no standard has been set. Based on the results of interviews with several farmers, the previous studies reported a variation in the height of the postharvest main stem of between 10 and 30 cm above ground level [9]. The difference in that height is associated with the severity level of pruning. The height of the remaining main stem should be studied further as it relates to food reserves for growth in further seasons. Pruned plants may likely experience a different source and sink balance condition than unpruned ones. Too severe pruning may result in improper vegetative growth. Earlier studies reported a vegetative growth reduction of mandarin citrus due to heavy pruning [18]. Concerning leaf-oriented citrus such as kaffir lime, vegetative growth inhibition can directly threaten yields and profits. Therefore, there is an urgency to obtain the best pruning level for kaffir lime.

In addition to the importance of determining the pruning severity, optimizing leaf production in kaffir lime agribusiness needs to be supported by the best nitrogen (N) fertilizer doses. N is the imperative macronutrient for normal plant growth and development [19]. More specifically, this nutrient is required for producing chlorophyll, proteins, nucleic acids [20], amino acids, and sugar [21]. Thus, plant productivity highly depends upon N fertilization [22,23]. Previous studies have proven the role of N in increasing the vegetative growth of Eureka lemons and Maltese sweet citrus [24]. Concerning the kaffir lime, previous studies reported a strong and positive correlation between leaf nitrogen status and leaf oil production [25], strengthening the argument that N is urgently needed to produce good quality kaffir lime leaves. Unfortunately, there is no specific N-fertilizer dosage recommendation for leaf-oriented citrus such as kaffir lime. In Indonesia, the citrus research center has issued a recommendation only for fruit-oriented citrus, with a range of 10–20 g N per plant, for 1-year-old plants [26]. This dosage can be used as a reference for compiling specific recommendations for leaf-oriented kaffir lime cases.

Interestingly, there are some knowledge gaps pertaining to kaffir lime leaf production, i.e., the confusion resulting from a variation in the height of the postharvest main stem, the lack of information on the growth and yield in the following post-pruning period under similar mild-shading conditions, and the missing specific N recommendations for leaf-oriented production. Therefore, the present study aimed to determine the best pruning level and N dose based on the agro-physiological characteristics of kaffir lime under mild shading.

#### **2. Results**

#### *2.1. Nitrogen Status in Soil, Leaves, and Canopy N Uptake*

The present experiment measured both N status in leaves and soil at 90 DAT, and the results showed no significant differences in total soil N levels under different doses of N fertilizer applied (Figure 1A). The total N content of the soil in all the treatments ranged from 0.19 to 0.20%. In contrast, N-fertilizer doses had a noticeable effect on the total N content of the leaf tissue, with the highest N–leaf tissue in the highest N-fertilizer dose (40 g plant<sup>−</sup>1), while the lowest N–leaf tissue in control/no N fertilizer applied. Compared to the control, the increase in N–leaf tissue was varied, by about 13% on 10 g N plant−1, 17% on 20 g N plant−1, and 34% on 40 g N plant−1. Similarly, the increase in N-fertilizer doses was surely followed by a significant improvement in canopy N uptake (Figure 1B).

**Figure 1.** (**A**) Bar chart of nitrogen status on kaffir lime leaves and soil, and (**B**) regression curve of kaffir lime canopy nitrogen uptake under different nitrogen-fertilizer dosage. Note: mean values followed by the same letter are not significantly different based on DMRT at α 0.05.

The results of the regression analysis showed a high coefficient of determination (R2) of 0.99. It was likely that the N-fertilizer dose variable could be used to estimate kaffir lime canopy N uptake by employing the provided mathematical equation. Additionally, correlation analysis also confirmed (i) the insignificant and weak correlation between N–soil and N–leaf tissue; and (ii) the significant, positive, and strong correlation between N–leaf tissue and canopy N uptake (r 0.97) (Table 1). The importance of N, as represented by canopy N uptake, for plant growth and physiological processes was also proved by correlation analysis, since canopy N uptake showed significant, positive, and strong correlation to certain variables of plant growth and physiology, namely relative growth rate, shoot number, leaf numbers, plant fresh weight, and leaf chlorophyll content.

**Table 1.** Pearson correlation analysis between nitrogen status, characteristics of production and physiology of kaffir lime.


Note: NTS—nitrogen total in soil, NLC—nitrogen content in leaves, CNU—canopy nitrogen uptake, RGR relative growth rate, PH—plant height, SN—shoot numbers, LN—leaf numbers, LFW—fresh weight of leaves, PFW—plant fresh weight, PR—photosynthetic rate, SC—stomatal conductivity, TR—transpiration rate, CA chlorophyll α, CB—chlorophyll β, CT—chlorophyll total.

#### *2.2. Growth Performance under Different Pruning Levels and N Dosages*

Growth performance seemed to be significantly affected by the single factor effect, rather than a combination of both N dosage and pruning factor (Table 2). The results of the analysis of correlation highlighted the positive, strong and significant correlation between relative growth rate (RGR) to plant height (r 0.97), number of shoots (r 0.96) and number of leaves (r 0.99) (Table 1). Another statistical analysis, namely DMRT, also

reported that the RGR, plant height, and number of leaves were increased significantly, along with the increase in the dose of N fertilizer given. Plants treated with 40 g N per plant showed the best growth performance, especially compared to the control, by displaying significant improvements of about 99% on RGR, 48% on plant height, and 146% on leaf numbers. However, the best growth performance on the 40 g N plant−<sup>1</sup> treatment was not significantly different to the 20 g N plant<sup>−</sup>1.

**Table 2.** Kaffir lime growth performance under different pruning levels and nitrogen-fertilizer dosages.


Note: mean values within the same column and same factor followed by the same letter are not significantly different based on DMRT at α 0.05. RGR—relative growth rate, SP—short pruning, HP—high pruning, N\*P—the interaction of nitrogen-fertilizer dosages and pruning levels, Ns—not significant.

Harvesting activity employing a high pruning type (leaving the main stem at 30 cm above the ground) stimulated greater support to subsequent kaffir lime growth, rather than the short pruning type (leaving the main stem at 10 cm above the ground). The high pruning type had 20% greater growth rate than the short ones. However, the short pruning type successfully induced more shoots than the high ones. In terms of leaf numbers, the result was not significantly different between the treatments.

#### *2.3. Plant Production under Different Pruning Levels and N Dosages*

The statistical analysis depicted an insignificant interaction effect of N dose and pruning on all the variables observed related to plant production. Kaffir lime production was solely influenced either by N dose or pruning. Concerning N-fertilizer dosage, the best results of plant fresh weight were observed in plants fertilized with 40 g N that experienced an increase of 55.13 g compared to control. However, it was not markedly different from those fertilized with 20 g N (Table 3). The fresh weight of plants fertilized with 10 g N was 30% higher than controls. However, it was still 25% lower than those fertilized with 40 g N. Concerning the pruning levels, the short-pruned type is thought to have a lower assimilated reserve than the high-pruned ones, leading to a lower production response. Pruning by leaving 30 cm of the main stem above the ground resulted in the improvement of plant and leaf production by about 28 and 22%, respectively, compared to the short pruning type.

The present experiment also revealed the partition of biomass in the entire plant body in response to N dose and pruning factor. In the absence of N fertilizer, the stem became the most dominant part, representing more than 50%. In the presence of N fertilizer, the more N applied, the more dominant the leaf biomass, exceeding the stem portion, in contrast to previous case (Figure 2A). Unlike the root and stem parts, the leaves are the most commercially valuable part of the kaffir plant. Therefore, the context of yield in kaffir lime is associated with the number and weight of leaves harvested in a certain unit of growing area. The estimation of leaf production extrapolated from the present findings, showed that plants fertilized with 40 g N and 20 g N received the best treatment, with an increase in yield of more than 2× compared to control (Figure 2B). In addition, regression analysis also re-confirmed and identified a strong association between (i) leaf fresh weight

and N dose (Figure 3A) and (ii) leaf numbers and N dose (Figure 3B). The application of N fertilizers at various doses showed a quadratic pattern on both the fresh weight of leaves (R2 = 0.9864) and leaf numbers (R2 = 0.9998). The fresh weight of leaves in plants fertilized with 10 g N, 20 g N and 40 g N increased over controls, by 82, 174 and 198% respectively. The number of leaves in plants fertilized with 10 g N, 20 g N and 40 g N increased over controls, by 72, 123 and 146% respectively. However, there was no significant difference in fresh weight of leaves and number of leaves between 40 g N plant−<sup>1</sup> and 20 g N plant<sup>−</sup>1.

**Table 3.** Plant production (fresh weight) of kaffir lime under different pruning levels and nitrogenfertilizer dosages.


Note: mean values within the same column and same factor followed by the same letter are not significantly different based on DMRT at α 0.05. SP—short pruning, HP—high pruning, N\*P—the interaction of nitrogenfertilizer dosages and pruning levels, Ns—not significant.

**Figure 2.** (**A**) Biomass partition and (**B**) estimated leaf production (ton per ha) of kaffir lime under different nitrogen-fertilizer dosages and pruning levels.

**Figure 3.** Regression curve of kaffir lime (**A**) fresh weight of leaves and (**B**) leaf numbers under different nitrogen-fertilizer dosages. Note: mean values followed by the same letter within the same curve are not significantly different based on DMRT at α 0.05.

#### *2.4. Physiological Response of Kaffir Lime under Different Pruning Levels and N Dosages*

The alteration of growth and production of kaffir lime in response to different Nfertilizer dosages and pruning levels was followed by the variable of plant physiology. Higher leaf production in N-fertilized plants was associated with an increase in the rate of plant photosynthesis. The rate of photosynthesis between plants fertilized with 20 g N and 40 g of N was insignificant; however, there was a noticeable increase of about 15% compared to both control (0 g N) and 10 g N (Table 4). In contrast, there was no significant difference in N-fertilizer dosage on stomatal conductance, transpiration rate, stomatal limitation to CO2 uptake, and water-use efficiency (WUE) in the kaffir lime plants. Stomatal conductance, transpiration rates, intrinsic WUE and instantaneous LUE in the present experiment varied in the range of 0.37–0.38 mol H2O m−<sup>2</sup> s<sup>−</sup>1, 6.25–6.59 mmol H2O m−<sup>2</sup> s<sup>−</sup>1, 2.68–3.23 μmol CO2 mmol H2O<sup>−</sup>1, and 0.58–0.98 μg lux<sup>−</sup>1, respectively.

**Table 4.** Kaffir lime physiological response under different pruning levels and nitrogen-fertilizer dosages.


Note: mean values within the same column and same factor followed by the same letter are not significantly different based on DMRT at α 0.05. Pn—photosynthetic rate (μmol CO2 m−<sup>2</sup> s−1), Tr—transpiration rate (mmol H2O m−<sup>2</sup> s−1), Sc—stomatal conductance (mol H2O m−<sup>2</sup> s−1), Sl—stomatal limitation to CO2 uptake (mol m−<sup>2</sup> s−1), WUE—intrinsic water-use efficiency (μmol CO2 mmol H2O−1), LUE—instantaneous light-use efficiency (μg lux<sup>−</sup>1). Ns—not significant, N\*P—the interaction of nitrogen-fertilizer dosages and pruning levels.

As the only variable that was significantly affected by N-fertilizer dose, the rate of photosynthesis in kaffir lime was strong and positively correlated to the content of chlorophyll, i.e., chlorophyll a (r 0.86) and chlorophyll b (r 0.85) (Table 1). The chlorophyll content increased along with the increase in N fertilizer applied to kaffir lime plants and was actually supported by the morphological fact that could be seen directly in the field. Based on field observations, the yellowish color on kaffir lime leaves fertilized with 0 g N and 10 g plant−<sup>1</sup> was an early symptom of N-deficiency stress. Meanwhile, plants with 20 g N and 40 g N plant−<sup>1</sup> experienced normal green leaves, presumably not experiencing a Ndeficiency condition (Figure 4). Such external leaf color variations between the N-fertilizer dosages used could be reconfirmed by the results of pigment analysis.

**Figure 4.** Kaffir lime appearance under different pruning levels and nitrogen-fertilizer dosages.

The statistical analysis of the pigment content of kaffir lime revealed the significant effect of N dosage on plant chlorophyll content, both chlorophyll α and chlorophyll β. The absence of N-fertilizer application on the control treatment showed the lowest pigment chlorophyll content of all (Table 5). The higher the N-fertilizer dose applied, the higher the content of chlorophyll, chlorophyll β and chlorophyll total in kaffir lime leaves. In the 40 g N treatment, the content of chlorophyll α and chlorophyll β and its total chlorophyll increased significantly, by about 1.37 mg g−1, 0.48 mg g−1, respectively and 1.85 mg g−<sup>1</sup> compared to the lime without fertilizer.


**Table 5.** Kaffir lime leaves' pigment content under different nitrogen-fertilizer dosages.

Note: mean values within the same column followed by the same letter are not significantly different based on DMRT at α 0.05.

#### **3. Discussion**

Pruning is the agricultural technique used to regulate plant growth, reduce pest and disease incidence, and increase horticultural management effectiveness [27–31]. Canopy rejuvenation is the foremost important benefit obtained from such a technique. In a rejuvenated canopy, new leaves grew to immediately restore the lost foliage [32] and these leaves possess much more productivity [33] due to the higher potential of carbon assimilation [32]. A previous study reported the use of light pruning in the form of pinching to induce robust canopy growth in early seedlings of kaffir lime [17]. In the postpruning period, the massive growth of lateral shoots is caused by lowering apical dominance [34–36]. In the fruit-oriented major citrus of mandarin, heavy pruning was applied to rejuvenate the canopy, but as a result, there was a decline of vegetative growth due to a severe decline in plant resource capacity [18]. Thus, mild to moderate pruning was highly recommended for those kinds of citrus [37,38]. However, to harvest the leaf yield of kaffir lime, growers have to apply heavy pruning. Due to the nature of citrus plants which are sensitive to leaf disturbance [17,18], the pruning level should be adjusted to be lighter. Comparative analysis on the pruning factor resulted in a significant increase in growth rate, plant and leaf production by about 20, 28 and 22%, respectively, in the high pruning compared to the short pruning that left only 10 cm of the main stem. The higher assimilate reserve in the existing main stem could likely support a higher growth rate, and final yield, regardless of nitrogen doses. Therefore, the recommendation for pruning in kaffir lime is the high pruning type by leaving the main stem 30 cm above the grafted join spot.

Aside from pruning, nitrogen management was also evaluated to provide the first N-fertilizer recommendations for kaffir lime grown under artificial mild shading. Most citrus growers use leaf tissue rather than soil as the basis for fertilizer application. Leaf tissue is a more representative proxy for estimating a tree's nutritional status for mobile nutrients, such as nitrogen [39–41]. The result of the leaf-based nutritional test should be compared with the optimal range of that nutrient [42,43]; thus, there is an urgency to estimate the optimal range of nutrients for achieving a profitable citrus yield [44]. A previous study [45] found the variation in optimal leaf nutrient contents in four fruitoriented popular citrus, namely oranges, mandarin, grapefruit and pomelo. The existence of variations in the optimal leaf nutrient content is thought to be related to differences in cultural practices, edaphic, climatic, and genetic factors [44]. In fact, leaf nutrient-based citrus fertilizer recommendation guidelines have been intensively studied for seventy years in the USA [46–48], and have been updated several times by numerous researchers [49–51].

Concerning leaf nitrogen, published studies have produced recommendations based on total leaf-N concentration for popular fruit-oriented citrus species [45,50,52]. Interestingly, the present study proposed the total leaf N range (2.06–2.36% N total) for the leaforiented minor citrus, kaffir lime, grown under mild shading. The correlation and regression analysis were adopted in the present experiment since earlier studies frequently used it to estimate the relationship between citrus yield and leaf nutrient concentrations [45,50,53]. Our findings highlighted the positive and strong correlation between leaf-N status and relative growth rate (r 0.93) and final leaf numbers harvested (r 0.93). Regression analysis also found positive quadratic patterns between (i) N-fertilizer dose and fresh weight of leaves (R2 0.9864) and (ii) N-fertilizer dose and leaf numbers (R2 0.9998).

Due to its quadratic pattern, a dose of 40 g N plant−<sup>1</sup> does not automatically become the best fertilizer recommendation. It has already been reported that plants fertilized with both 40 g N and 20 g N doubled the yield of the control. However, 20 g N plant−<sup>1</sup> is more efficient with a relatively similar effect to 40gNplant−<sup>1</sup> for increasing growth rate, photosynthetic rate, fresh weight of leaves and leaf numbers of kaffir lime under mild-shading conditions. Mild shading was previously reported to produce a beneficial stress instead of a harmful one [17]. Best practice of N-fertilizer application under lightly shaded conditions may become a combo booster for kaffir lime leaf production. Similarly, the success of N fertilizer and beneficial shading for boosting plant growth performance was also reported by previous researchers, as indicated by larger and broader leaves [54], more dominant vegetative growth [55], and higher yield [56]. In contrast, slower plant growth leading to lower production performance was observed both in the control and the 10 g N plant−<sup>1</sup> treatment. Vegetative growth inhibition is a common plant response under N-deficient conditions [57].

Aside from vegetative improvement, another advantage of best N-fertilizing practice is the regulation of assimilate translocation priority. N adequacy seemed to alter the assimilate translocation priority in kaffir lime plants. A N-sufficient plant, treated with both 20 g N plant−<sup>1</sup> and 40 g N plant<sup>−</sup>1, showed a dominant portion of leaves, while a deficient plant likely had a large portion of stem. Concerning agribusiness profit, the leaf part of the kaffir lime is more valuable and profitable than the stem or even the root [7,9,10], due to the content of various beneficial phytochemicals such as citronellal, citronellol, citronellyl acetate, linalool and caryophyllene that contribute to the strong aromatic formed [12,58,59].

The N-sufficient condition in the 20 g N plant−<sup>1</sup> and 40 g N plant−<sup>1</sup> displayed a good leaf–N total (>2%) associated with proper growth performance. That growth improvement is mainly caused by a higher photosynthetic rate and chlorophyll content. Similarly, the relationship between RGR to photosynthetic and chlorophyll content was confirmed by the correlation results, with coefficients of correlation of about 0.94 and 0.91, respectively. A N-sufficient status is vital for constructing optimal leaf photosynthesis. The leaf-N status represents the protein content for the Calvin cycle and thylakoids that are subsequently associated with leaf photosynthetic capacity [60]. Concerning the normal leaf cells of C3 plants, including kaffir lime, N is allocated mostly in the chloroplast, at about 75%, with 10% in the cell wall, and 5% in mitochondria [61]. The variation of those partitions may occur in N-deficient conditions.

The N-deficient plant, as observed in the control and 10 g N plant−1, displayed a low leaf–N total (<2%) and exhibited a yellow leaf appearance, implying a chlorosis phenomenon. The data of the chlorophyll test also displayed a significant reduction compared to the N-sufficient plant. Leaf chlorophyll content was previously reported to be crucial for assimilation rate since it positively and strong correlates to leaf photosynthetic rate [62]. The degradation of chlorophyll, called chlorosis, begins to appear on the lower leaves prior to spreading over the entire canopy, and even in severe cases, it can cause necrosis on old leaves [57]. Chlorosis, as a popular N-deficient symptom, is caused by a failure to form chlorophyll pigments [63]. A published study in oranges and pomelo described the main reason behind lowering CO2 assimilation, i.e., impairment of the thylakoid structure and photosynthetic electron transport chain (PETC) in the leaves and declining leaf photosynthetic pigment levels [64]. Moreover, N-deficient leaves are proven to have smaller chloroplasts and no starch granules. In contrast, N-sufficient leaves have large chloroplasts, with fully formed grana components and larger starch granules for performing greater assimilation [65]. In contrast, mild-shaded and N-fertilized plants may have a more robust photosynthetic apparatus, as evidenced by a large number of thylakoids per granum and an abundance of grana per chloroplast [54].

#### **4. Materials and Methods**

#### *4.1. Study Site*

The experiment was carried out at Pasir Kuda experimental field of IPB University, Bogor, Indonesia (6◦36 36 S, 106◦46 47 E, 263 m above sea level) from November 2018 to March 2019. The soil description of the Pasir Kuda experimental field was a sandy clay latosol soil with an actual pH, C-organic, N total, P total, and K total of about 6.7, 2.37%, 0.19%, 240 mg P2O5 100 g<sup>−</sup>1, and 160 mg K2O 100 g−1. During the study period, the experimental field was exposed to the rainy season, with monthly rainfall intensity ranging from 230 mm to 318 mm (x-bar 289 mm).

#### *4.2. Planting Materials*

Plant materials were nine-month-old seedlings obtained by the grafting technique that combined kaffir lime (*Citrus hystrix* DC) scions onto rangpur lime (*C. limonia* Osbeck) rootstock. Before field transplanting, seedlings underwent initial selection to confirm only selected plants were involved in the present experiment, with certain requirements, such as bifoliate leaves, pest and disease-free, normal growth, dormant apical bud, uniform in leaf numbers (30 ± 2 leaves) and plant height (60 ± 4 cm).

#### *4.3. Research Procedure*

The present study employed a split-plot experimental design, with N dosage as the main plot and pruning level as the subplot. Four levels of N dosage were tested, viz., 0 g N plant−1, 10 g N plant−1, 20 g N plant−1, and 40 g N plant−1. Two levels of pruning were also evaluated, namely short and high pruning. Short pruning was technically defined as leaving only 10 cm of the main stem above the ground, whereas high pruning gave the cutting point at 30 cm from above the ground. Six replications were provided for each combination treatment; thus, 46 experimental units in the form of kaffir lime seedlings were counted in total.

Kaffir lime seedlings were raised in a monoculture cropping system under mildshading conditions that were artificially formed by installing a black shading net 2 m above the soil surface. In a previous, similar study, this treatment resulted in (1) a reduction of sunlight, ambient temperature, and soil temperature of about 23, 6.3, and 6.5%, respectively; and (2) the improvement of ambient relative humidity by about 2%, compared to open field monoculture system [17]. Kaffir lime transplanting to the field was conducted in November 2018, with a 50 cm × 50 cm planting distance.

Pruning was applied, according to the treatment, in December 2018, or 30 days after planting (DAP). The cutting point for pruning was actually determined based on the grafted join spot. That spot was normally found 15 cm above the stem base of the rootstock variety. However, the joining spot seemed to equal the soil surface due to the soil banking technique for suppressing undesired shoot growth from the rootstock variety.

Inorganic fertilizers, apart from N, such as phosphorus (P) and potassium (K) were applied uniformly at 33 DAP, in the form of 15 g P2O5 and 10 g K2O, following the national citrus agency recommendations [26]. N fertilization was carried out simultaneously with P and K, with a dose adjusted to the treatment. All mentioned fertilizer was delivered in the morning (7.00 am) of a sunny day through a soil drench surrounding the seedling (10 cm away from the main stem). Hand-weeding was routinely applied every month. Pest and disease inspection was conducted weekly, and the damage was chemically managed. Harvesting was scheduled at 120 DAP in March 2019.

#### *4.4. Measured Variables*

Measured variables were N status in soil and leaves, canopy N uptake, growth performance, plant production (fresh weight), and physiological responses. The N content in the soil and leaf samples was analyzed by using Kjeldahl method at 115 DAP. Plant production (fresh weight) was measured on harvesting day (120 DAP) by weighing either the whole plant or individual parts in the analytical balance (Hwh, China). Plant samples were then dried using an oven at 80 ◦C for 3 days to obtain plant dry weight by using a similar analytical balance. Canopy N uptake was obtained from the multiplication of the canopy dry weight and the N content of leaves. The relative growth rate (RGR) was calculated based on the ratio of the increase in plant dry weight to the number of weeks from the 1st pruning (at 30 DAP) to the 2nd pruning (120 DAP), i.e., 13 weeks. Plant height, shoot number, and leaf numbers were also observed on harvest day using a roll meter (Kenmaster, North Jakarta, Indonesia), and hand counter (Kenko, North Jakarta, Indonesia), respectively.

Plant physiological variables such as photosynthetic rate (μmol CO2 m−<sup>2</sup> s−1) and transpiration rate (mmol H2O m−<sup>2</sup> s−1) were measured at 94 HSP at 10.00 am (sunny day) using the Li-6400XT portable photosynthesis system (Licor Inc, Lincoln, NE, USA). In addition, the present experiment also measured (i) intrinsic water-use efficiency (μmol CO2 mmol H2O<sup>−</sup>1) and (ii) instantaneous LUE (μg lux<sup>−</sup>1) by (i) dividing the rate of photosynthesis by the rate of transpiration rate and (ii) dividing the fresh weight of leaves by the perceived sunlight amount, respectively. The amount of perceived sunlight was measured by Lux-28 portable digital lux meter (Danoplus, Hong Kong, China). Leaf pigment was measured at 86 DAT in the present experiment by following the Sims and Gamon method [66].

#### *4.5. Data Analysis*

Quantitative data obtained in the present experiment was subjected to the analysis of variance (ANOVA) and any significant difference found was further analyzed by the Duncan's multiple range test (DMRT) at α 0.05. Pearson correlation analysis was carried out to find the closeness of the relationship between the observed variables, such as N status, plant growth, production, and physiological response. In addition, regression analysis was also performed to elucidate the association between N dose and certain important yield-related variables, such as canopy N uptake, fresh weight of leaves and leaf numbers. All statistical analysis was performed using the Statistical Tool for Agricultural Research (STAR) version 2.0.1.

#### **5. Conclusions**

The present experiment succeeded in determining the best practice for pruning and N fertilizer to boost leaf production under mild-shading conditions. Pruning by leaving 30 cm of the main stem above the ground resulted in a significant improvement of plant and leaf production by about 28 and 22%, respectively, compared to the short pruning that left only 10 cm of the main stem. The higher assimilate reserve in the existing main stem likely produced higher support for recovering the lost foliage. Concerning N management, N-fertilizer dosage had a noticeable effect on the total N content of the leaf tissue, with the highest N–leaf tissue content achieved from the highest N dose. Both correlation and regression analysis confirmed that N is crucial for plant growth and plant yield due to the role of this nutrient in chlorophyll content and photosynthetic rate. Kaffir lime treated with 0 and 10 g N plant−<sup>1</sup> experienced N-deficient conditions, as indicated by leaf chlorosis, leading to lower chlorophyll content, photosynthetic rate, relative growth rate and then leaf yield harvested. A N-sufficient condition was achieved as the effect of 20 and 40 g N plant−<sup>1</sup> application, producing a great growth and yield performance due to a high carbon assimilation rate. However, a dose of 40 g N plant−<sup>1</sup> does not automatically become the best fertilizer recommendation, since 20 g N plant−<sup>1</sup> is more efficient with a relatively similar effect for increasing kaffir lime leaf production.

**Author Contributions:** Conceptualization, R.B. and R.P.; methodology, R.B. and E.S.; software, R.B.; validation, D.E. and A.A.; formal analysis, R.B.; investigation, R.B.; resources, D.E. and A.A.; data curation, R.B.; writing—original draft preparation, R.B.; writing—review and editing, R.P., E.S., D.E. and A.A.; visualization, R.B.; supervision, R.P., E.S., D.E. and A.A.; project administration, R.B.; funding acquisition, R.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding. The APC was fully funded by Universitas Padjadjaran.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** Authors acknowledge technical support given by Nandang Hasanudin for assisting with the Li-6400XT test on kaffir lime. In addition, authors would also like to thank all the management team of the Pasir Kuda experimental garden for their administrative support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
