Carbohydrate concentration in different Mexican lemon tissues

To assess starch concentrations, leaf and tree roots with HLB and healthy trees were selected. Fifty leaves per treatment separated veins and leaf laminae of mature leaves (120 days) and immature leaves (30 days). Starch content in secondary roots was also evaluated.

Starch extraction was performed separately for leaf and root from 20 mg of lyophilized and ground tissues of each sample. 500 μL of distilled water was added and incubated at 100 °C for 10 min. Subsequently, the mixture was centrifuged at 10 774 rpm for 2 min. 300 μL of supernatant was recovered in an Eppendorf tube and 900 μL of pure alcohol was added and centrifuged at 9 449 rpm for 5 min.

The supernatant was discarded and the precipitated starch was resuspended in 1 mL of sterile double-distilled water (Milli-Q) and stirred in a vortex for 3 min (^{González et al., 2011}). For quantification of starch, 30 μ0L of the starch extract and 50 μL of iodine solution was added to a 96-well optical plate and the absorbance at 595 nm was read in triplicate on a spectrophotometer (Thermo Scientific ™ Multiskan GO, USA). A standard curve was made from pure rice starch (Sigma-Aldrich, USA).

Because the main sugars present in the citrus phloem are sucrose and glucose with 64 and 20% respectively (^{Hijaz and Killiny, 2014}). The concentrations of these were determined in leaf tissues in rib and leaf lamina of 120 and 30 days. Quantified with commercial kits (Sigma-Aldrich, USA).

Chlorophyll concentration in leaves

The concentration of chlorophyll in Mexican lemon leaves with different degrees of HLB involvement was determined (Figure 1). Five stages of disease development in mature leaves were chosen and chlorophyll a, chlorophyll b and total chlorophyll were quantified according to (^{Warren, 2008}).

Figure 1 Levels of HLB symptoms in Colimex lemon leaves of trees in production.  

The SPAD index was measured on the same leaf due to its high correlation with the actual chlorophyll content in the leaves (^{Intrigliolo et al., 2000}). The readings were made with a portable chlorophyll meter SPAD 502 (Soil Plant Analysis Development, Minolta).

Foliar nutrient analysis

Concentrations of macro- and micronutrients in mature leaf (120 days) were evaluated. From healthy trees and with HLB. 100 mature leaves of each treatment were collected. For each sample a compound mixture and three repetitions were performed. The samples were dried in a Felisa^® brand oven at a temperature between 65 to 70 °C for 72 h. Subsequently, they were dried at constant weight and processed in Thomas Scientific^® brand electric mill, with 40 cavity mesh, finally obtaining fine powder, which was used for digestions.

The elements analyzed were total nitrogen (N), phosphorus (P), potassium (K), calcium (Ca) and magnesium (Mg); and the microelements iron (Fe), copper (Cu), zinc (Zn), manganese (Mn) and boron (B). Data were collected and analyzed by analysis of variance (Anova), the comparisons of means were made using the Tukey test (α≥ 95%) with the statistical analysis system InfoStat Version 2018e.

Concentration of carbohydrates in veins and leaf blade of Mexican lemon

In the test of means the leaves of trees of lemon Colimex infected with CLas were statistically superior in concentration of starch in leaf of leaves of 120 days in comparison with leaves without HLB. 30 and 120 day veins showed no statistical differences between healthy and diseased trees with HLB (Figure 2). In sweet orange trees infected with CLas, the accumulation of starch in leaves has been described compared to healthy leaves (^{Rosales and Burns, 2011}).

Different letters on the bars indicate significant statistical differences (p< 0.05).

Figure 2 Effect of CLas on the concentration of starch in Mexican lemon tissue in production. 30 and 120 day leafs. Without the presence of HLB (-HLB) and infected with HLB (+HLB). 

^{Schneider (1968)}, observed high accumulation of starch in leaves and petioles compared to healthy tissues. This accumulation of starch is caused by the obstruction of the transport of photoassimilates in the foliar vascular system and accelerated rates of leaf starch synthesis. In orange trees infected with HLB, the plugging of sieve tubes by deposition of calose was reported (^{Etxeberria et al., 2009}; ^{Kim et al., 2009}). Which has an important role in the citrus defense response to the HLB (^{da Graça et al., 2015}), thus affecting the transport of carbohydrates and causing massive accumulation of starch in leaf plastids (^{Schneider, 1968}).

Also, accumulation of leaf starch due to Zn deficiency has been reported (^{Sagaram and Burns, 2009}). These results agree with ^{Esquivel-Chávez et al. (2012)} that found a higher concentration of starch in Mexican lemon leaves with HLB in the greenhouse. However, the effect of HLB on starch concentration was mainly observed in the 120-day leaf blade of Colimex lemon in production, due to the higher concentration of chloroplasts in this tissue.

Figure 3 shows the comparison of means where the leaf veins presents a higher concentration of sucrose and is statistically different from the leaf blades in the two stages of development evaluated in both infected and healthy tissues. The highest concentration of sucrose occurred in the leaf veins of 120 days with HLB. In contrast, the lowest concentration was obtained in the leaf blade of healthy 30-day leaves. These results are due to CLas causing imbalance in the distribution of carbohydrates and accumulation of sucrose in leaf (^{Fan et al., 2010}).

Different letters on the bars indicate significant statistical differences (p < 0.05).

Figure 3 Effect of CLas on the concentration of sucrose in Mexican lemon tissue in production. 30 and 120 day leafs. Without the presence of HLB (-HLB) and infected with HLB (+HLB). 

Due to the concentration of CLas in veins and the decrease in cellular activity within the cambium and radial parenchyma in mature leaf petioles infected with HLB, which results in a significant failure in the carbohydrate transport route (^{Brodersen et al., 2014}). In lemon Colimex with HLB sucrose accumulates mainly in leaf veins of 120 days.

Glucose concentrations based on the comparison of means have significant statistical differences, showing higher concentrations in tissues with HLB. These results show the highest accumulation of glucose in 30-day veins with HLB (Figure 4). This agrees with ^{Fan et al. (2010)} who described that HLB causes imbalance in carbohydrate distribution and glucose accumulation. Because CLas behaves like Spiroplasma citri, preferably using fructose increasing the activity of invertase and consequently accumulate glucose.

Different letters on the bars indicate significant statistical differences (p < 0.05).

Figure 4 Effect of CLas on the concentration of glucose in Mexican lemon tissue in production. 30 and 120 day leafs. Without the presence of HLB (-HLB) and infected with HLB (+HLB). 

In complement ^{Li et al. (2009)} reported a higher concentration of CLas in the veins of infected leaves compared to the leaf blade. HLB causes glucose accumulation mainly in the 30-day rib of lemon Colimex.

As observed in Figure 5, the roots of trees infected with HLB show significant statistical differences with healthy roots in starch concentration. With a decrease of 40%. Lemon Colimex shows a decrease in the concentration of starch in tree roots with HLB. Because they suffer from a nutritional disorder due to the interruption of photosynthesis transport from root outbreaks (^{Zhang et al., 2011}). Causing a decrease in starch reserves in the roots of trees infected with CLas. This also causes them not to develop and suppress the growth of new roots (^{Etxeberria et al., 2009}).

Different letters on the bars indicate significant statistical differences (p< 0.05).

Figure 5 Effect of CLas on starch concentration in Mexican lemon tree roots in production. Without the presence of HLB (-HLB) and infected with HLB (+HLB). 

Chlorophyll concentration in leaves

The levels of involvement from asymptomatic leaves to level 4 of HLB symptoms showed significant statistical differences with regard to the total chlorophyll concentration as shown in Figure 6. Due to the progressive accumulation of starches that cause the disintegration of the chloroplast thylakoid system (^{Etxeberria et al., 2009}; ^{González et al., 2011}) and damages photosynthetic membranes by decreasing the concentration of chlorophyll (^{Sagaram and Burns, 2009}). Although, the starch inside the chloroplast does not break the outer membranes, but the structure of the internal grana is destroyed which causes chlorosis (^{Achor et al., 2010}).

Different letters on the bars indicate significant statistical differences (p< 0.05).

Figure 6 Concentration of chlorophyll (Chlorophyll a, Chlorophyll b and total Chlorophyll) in Mexican lemon tree leaves in production with different levels of HLB involvement. 

It has also been described that increasing sucrose/glucose levels can lead to repression of genes involved in photosynthesis and cause chlorophyll decrease (^{Fan et al., 2010}). Therefore, in Colimex lemon leaves the decrease in the concentration of chlorophyll a, b and total in leaf was observed depending on the degree of involvement by HLB.

When performing the SAPD readings at the HLB involvement levels evaluated, no significant statistical differences were found between the asymptomatic leaves, level 1 and level 2. In contrast, these were statistically different at levels 3 and 4 (Figure 7). The SPAD reading is only taken from a partial part of the leaf so that the destruction of chloroplast is not uniform (^{Achor et al., 2010}), the sensor can capture areas with different concentrations of chlorophyll on the same leaf.

Different letters on the bars indicate significant statistical differences (p< 0.05).

Figure 7 SPAD index in Mexican lemon tree leaves in production with different levels of HLB involvement. 

So, the statistical analysis only detects the differences in leaves with marked chlorosis in most of the leaf. In Lemon Colimex SPAD only detects statistical differences in the degrees of greatest involvement by HLB with respect to asymptomatic leaves.

Foliar nutrient analysis

Table 1 shows the nutrient data that presented significant statistical differences between leaves of trees with HLB and healthy leaves (potassium, calcium, magnesium, copper, iron, zinc and boron). The same table shows the percentage of decrease, highlighting the copper with more than 57%. Nitrogen, phosphorus and manganese did not show significant statistical differences.

There are works in several regions of the world with different citrus fruits that show contrasting results. The decrease of Ca and Zn in trees infected with HLB. Like ^{Razi et al. (2011)} showed lower foliar concentrations of Mg and Fe in trees with HLB and no difference in K, P and Zn. ^{Etxeberria et al. (2009)} described deficiencies of Zn and Fe in symptomatic trees. Like ^{Tian et al. (2014)} describe Zn deficiency in leaves with HLB. In grapefruit infected with HLB, concentrations of Ca, Mg, Fe, Zn, Mn and Cu decreased. Agreeing with this work with the exception of Mn (^{Nwugo et al., 2013}).

The HLB in lemon Colimex affected the concentrations of K, Ca, Mg, Cu, Zn and B due to the loss or restriction of the root, which causes drought stress and changes in the mineral status of the outbreaks (^{Spann and Schumann, 2009}). The results differ from what was reported, possibly to be different species and the use of containers and fertigation in this experiment. Since differences in the management of nutrition, soil, climate, variety/rootstock combination, state of plant development and nutrient reabsorption efficiency during citrus experiments infected with HLB have an effect on the data obtained (^{Razi et al., 2011}; ^{Cao et al., 2015}).