Metabolomic comparative analysis of the phloem sap of curry leaf tree (Bergera koenegii), orange jasmine (Murraya paniculata), and Valencia sweet orange (Citrus sinensis) supports their differential responses to Huanglongbing

ABSTRACT

Orange jasmine, Murraya paniculata and curry leaf tree, Bergera koenegii are alternative hosts for Diaphorina citri, the vector of Candidatus Liberibacter asiaticus (CLas), the pathogen of huanglongbing (HLB) in citrus. D. citri feeds on the phloem sap where CLas grows. It has been shown that orange jasmine was a better host than curry leaf tree to D. citri. In addition, CLas can infect orange jasmine but not curry leaf tree. Here, we compared the phloem sap composition of these 2 plants to the main host, Valencia sweet orange, Citrus sinensis. Phloem sap was analyzed by gas chromatography-mass spectrometry after trimethylsilyl derivatization. Orange jasmine was the highest in proteinogenic, non-proteinogenic amino acids, organic acids, as well as total metabolites. Valencia was the highest in mono- and disaccharides, and sugar alcohols. Curry leaf tree was the lowest in most of the metabolites as well as total metabolites. Interestingly, malic acid was high in Valencia and orange jasmine but was not detected in the curry leaf. On the other hand, tartaric acid which can prevent the formation of malic acid in Krebs cycle was high in curry leaf. The nutrient inadequacy of the phloem sap in curry leaf tree, especially the amino acids could be the reason behind the longer life cycle and the low survival of D. citri and the limitation of CLas growth on this host. Information obtained from this study may help in cultivation of CLas and development of artificial diet for rearing of D. citri.

Introduction

Asian citrus psyllid, Diaphorina citri (Hemiptera: Liviidae), is considered one of the most dangerous pests in citrus. D. citri transmits the huanglongbing pathogen, Candidatus Liberibacter asiaticus (CLas), while feeding on the citrus phloem sap. CLas is transmitted by D. citri in a persistent, circulative, and propagative manner.¹ Once CLas is inoculated into the citrus plants, it multiplies within the phloem sap and moves to other parts of the plants.² The proliferation of CLas in the citrus phloem sap indicates that it contains the entire nutrient required for CLas.

Citrus and its close relatives are the main hosts for D. citri. Field observations showed that sweet oranges and orange jasmine (Murraya paniculata (L.) Jack) are the preferred hosts for D. citri.³ In addition, studies also showed that the ornamental Rutaceous specie (Bergera koenigii) was also a good host for D. citri.⁴ Because orange jasmine and curry leaf tree (Bergera koenigii) are widely grown in many parts of the world as ornamental plants, there has been a significant concern about their role in spreading HLB disease.⁵ Consequently, many comparative laboratory studies were designed to test the preference of D. citri to these 2 ornamental plants.

Tsai & Liu (2000)⁶ tested the preference of D. citri to 4 hosts and showed that grapefruit was the preferred host, followed by rough lemon, orange jasmine, and sour orange. Teck et al. (2011)⁷ conducted a comparative life cycle study of D. citri on 3 different host plants (orange jasmine, curry leaf tree, and mandarin orange). The greenhouse study showed that D. citri can be reared on all 3 host plants and orange jasmine was the preferred host.⁷ The life cycle of D. citri on jasmine orange, citrus, curry leaf was 18.5, 19.0, and 23.0 days, respectively.⁷ Teck et al. (2011)⁷ also showed that females prefer to lay their eggs on young flushes.

Westbrook et al. (2011)⁸ assessed seedlings of 87 Rutaceae genotypes including orange jasmine and curry leaf tree for infestations of D. citri by surveying the tested population for the 3 life stages (eggs, nymphs, and adults). Orange jasmine and curry leaf tree were highly colonized by D. citri.⁸ Alves et al. (2014)⁹ evaluated the suitability of different citrus species (Hamlin, Valencia, Natal, Ponkan, Pêra, and orange jasmine) on the development of D. citri and found that Valencia and orange jasmine were the best hosts; the highest viability for eggs stage, total survival rates, and the net reproductive rates were observed on Valencia and orange jasmine.⁹

Orange jasmine and curry leaf tree are usually grown as ornamental plants in citrus production areas and were found to be good hosts for D. citri. This suggested that these plants may also serve as reservoirs for CLas.⁵ Early studies showed that orange jasmine can be infected by CLas through grafting and dodder transmission; however, the relation between CLas and curry leaf tree was not clear in early reports.⁵ Damsteeg et al. (2010)⁵ showed that D. citri can transmit CLas to orange jasmine, but not to curry leaf tree. Although disease symptoms were not clear in CLas-infected orange jasmine, the presence of CLas was confirmed by polymerase chain reaction (PCR) and back-inoculation to sweet oranges.⁵

Herein, we hypothesize that the phloem sap of orange jasmine is more rich in nutrients (diversity and/or quantity) than curry leaf tree which makes it a good host for both D. citri and CLas. To test this hypothesis, we studied the phloem sap composition of these 2 plants compared to sweet orange.

Material and methods

Valencia sweet orange, orange jasmine, and curry leaf tree (12 months old, about 0.8 m height) were kept in a temperature-controlled greenhouse (28–32°C). Phloem sap was collected using the centrifugation method described by Hijaz and Killiny (2014).¹⁰ Ten µL of each biological sample was placed into a 200-µL fused insert vial (National Scientific, MSCert 4000–30LVW) with 100 µL of extraction solvent (8:1:1 methanol, chloroform, water) and 10 µL of internal standard (IS) (1000 ppm azelaic acid) was added to each tube. One blank sample containing only the extraction solvent and IS solution was prepared for each biological replicate. The samples were dried under a nitrogen stream to complete dryness. The dried samples were derivatized to their TMS derivatives and analyzed by GC-MS as described by Hijaz and Killiny (2014). Compounds were identified based on their relative retention times and comparison of their ion spectra using Wiley 9^th ed. and NIST 2011 mass spectral libraries. For further confirmation, 57 analytical standards were derivatized and used as reference compounds. Quantitation to millimolar (mM) was made using linear calibration curves of analytical standards derivatized in the same manner as the phloem sap samples.

Statistical analyses were performed using JMP version 9.0 (SAS Institute, Inc.,). Data were normally distributed. Analysis of variance (ANOVA) followed by Post hoc pairwise comparisons using Tukey honestly significant different tests were used to compare the level of metabolites among different plant species.

Results

Total metabolites of phloem sap followed the pattern: Orange jasmine>valencia sweet orange>curry leaf tree

A representative gas chromatography-mass spectrometry (GC-MS) chromatogram of the TMS-derivatized phloem saps is shown in Fig. 1A. The concentrations of the detected metabolites are listed in Table 1. A total of 79 metabolites were detected in the selected plants. The metabolites were belonging to 9 groups; proteinogenic amino acids (PAAs), non-proteinogenic amino acids (NPAAs); organic acids, fatty acids, phenolics, mono- and disaccharides, sugar alcohols, sugar acids, phosphate compounds. The percentage composition of the main metabolites in 3 plants are shown in Fig. 1B. As percentages, mono- and disaccharides were the main group in Valencia, whereas NPAAs was the major group in orange jasmine. The total concentration of the detected metabolites in Valencia sweet orange, orange jasmine, and curry leaf tree were 300.1±212 mM, 411.9 ± 147.4 mM, and 123.8 ± 61.06 mM, respectively (Table 1).

Figure 1.

A total ion chromatogram (TIC) of the phloem sap of Valencia sweet orange, orange jasmine, and curry leaf after derivatization with methoxamine hydrochloride (MOX) and N-methyl-N-trimethylsilyl-trifluoroacetamide (MSTFA). Annotations for main methoxyaminated (MEOX) and/or trimethylsilylated (TMS) metabolite groups are displayed on the chromatogram (A). Percentages of the main groups of metabolites in Valencia sweet orange, orange jasmine, and curry leaf tree (B).

Table 1.

Concentrations of different metabolites detected in Valencia sweet orange (C. sinensis), orange jasmine (Murraya paniculata), and curry tree (Bergera koenegii) phloem sap (mM ± SD, n=6), using GC-MS.

compound	Valencia	Orange jasmine	Curry leaf tree		Valencia	Orange jasmine	Curry leaf tree
Amino acids				Mono- and disaccharides			ND
L-Alaninea	0.28 ± 0.08	0.73 ± 0.33	0.42 ± 0.14	Threose	ND	ND	ND
L-Valinea	0.19 ± 0.02	0.36 ± 0.10	0.18 ± 0.10	β-L-Arabinopyranose 1	0.49 ± 0.38	0.37 ± 0.13	0.16 ± 0.13
L-Isoleucinea	0.18 ± 0.02	0.02 ± 0.06	0.05 ± 0.09	Arabinosea	0.47 ± 0.18	ND	ND
L-Prolinea	2.75 ± 1.83	25.55 ± 8.29	2.46 ± 2.11	Xylosea	0.25 ± 0.07	0.73 ± 0.49	0.32 ± 0.06
Glycinea	0.38 ± 0.16	0.81 ± 0.47	0.60 ± 0.17	β-L-Arabinopyranose 2	0.36 ± 0.21	ND	ND
L-Serinea	0.55 ± 0.14	3.94 ± 0.55	1.12 ± 0.53	Unk Sugar furanose	ND	0.58 ± 0.63	0.21 ± 0.13
L-Threoninea	2.19 ± 0.91	7.53 ± 1.71	1.39 ± 0.77	Fructosea	27.15±22.00	19.02 ± 7.67	6.71 ± 5.63
L-Argininea	ND	7.00 ± 7.49	0.41 ± 0.20	Mannosea	0.52 ± 0.11	1.24 ± 0.59	0.47 ± 0.04
L-Aspartic acida	0.57 ± 0.40	1.38 ± 0.21	0.77 ± 0.80	Glucosea	51.61±32.83	18.94 ± 5.85	7.43 ± 5.36
L-Cysteinea	0.46 ± 0.35	0.00 ± 0.00	0.00 ± 0.00	Unk disaccharide (m/z 361)	0.13 ± 0.02	0.13 ± 0.04	0.11 ± 0.01
L-Asparagine 1a	1.86 ± 0.94	0.58 ± 0.15	0.57 ± 0 .13	Sucrosea	27.67±26 .81	39.77±18 .06	20.67 ± 5 .35
L-Ornithinea	ND	6.96 ± 3 .19	0.28 ± 0 .69	Turanosea	3.79 ± 1 .55	0.14 ± 0 .03	0.13 ± 0 .02
L-Glutamic acida	0.27 ± 0 .08	0.16 ± 0 .12	0.17 ± 0 .09	Trehalosea	ND	ND	ND
L-Phenylalaninea	0.21 ± 0 .05	0.53 ± 0 .17	0.18 ± 0 .09	Maltosea	ND	0.19 ± 0 .08	0.16 ± 0 .04
L-Asparagine 2a	0.54 ± 0 .25	ND	ND	TOTAL	112.44 ± 84 .16	81.1 ± 33 .57	36.37 ± 16 .77
TOTAL	10.43 ± 5 .23	55.55 ± 22 .84	8.6 ± 5 .91	Sugar alcohols
Other Amines				Erythritola	0.18 ± 0 .00	0.57 ± 0 .14	0.22 ± 0 .03
γ-Aminobutyric acida	1.80 ± 1 .18	8.50 ± 1 .69	1.71 ± 1 .75	Xylitola	0.25 ± 0 .07	0.22 ± 0 .03	0.22 ± 0 .04
Synephrinea	1.65 ± 0 .98	ND	0.40 ± 0 .32	Glucitola	0.47 ± 0 .23	ND	ND
Putrescinea	5.13 ± 4 .78	122.19±16 .38	1.41 ± 0 .70	Mannitola	0.27 ± 0 .06	0.27 ± 0 .03	0.15 ± 0 .16
2-Aminopropanol	1.44 ± 1 .18	0.38 ± 0 .44	0.45 ± 0 .24	chiro-Inositola	22.64±16 .28	13.93 ± 8 .14	3.50 ± 1 .04
N-Carboxy-glycine	0.45 ± 0 .14	0.27 ± 0 .10	0.25 ± 0 .05	Unk S Alc (m/z 319)	2.90 ± 1 .65	1.51 ± 0 .58	0.90 ± 0 .04
N-Methyl-L-proline	1.01 ± 0 .37	0.59 ± 0 .13	2.10 ± 3 .47	scyllo-Inositola	9.70 ± 7 .02	2.09 ± 1 .29	1.10 ± 0 .28
N-Acetyl-L-Proline	0.29 ± 0 .14	ND	ND	Myo-Inositola	11.91 ± 5 .46	7.07 ± 2 .20	4.85 ± 1 .94
Unk Amine (m/z 116)	0.86 ± 1 .15	4.85 ± 5 .61	3.60 ± 1 .34	Unk S Alc (m/z 319) (BK)	ND	1.09 ± 0 .55	0.42 ± 0 .46
TOTAL	12.63 ± 92	136.78±47 .20	9.90±7 .80	Unk S Alc (m/z 290)	1.62 ± 0 .56	2.07 ± 1 .68	0.94 ± 0 .06
Organic acids				TOTAL	49.94 ± 31 .33	28.82 ±14 .64	12.3 ± 4 .05
Pyruvic acid	0.20 ± 0 .23	0.08 ± 0 .03	0.05 ± 0 .03	Sugar acids
Lactic acid	0.42 ± 0 .66	0.33 ± 0 .07	0.19 ± 0 .11	Glutaric acida	1.14 ± 0 .32	ND	ND
Glycolic acida	0.14 ± 0 .17	0.24 ± 0 .11	0.11 ± 0 .09	2-Ketoglutonic acid	1.92 ± 1 .32	0.00 ± 0 .00	2.85 ± 0 .47
Oxalic acida	0.10 ± 0 .03	0.19 ± 0 .07	0.13 ± 0 .06	2-Ketoglutaric acida	1.14 ± 0 .32	0.11 ± 0 .29	0.40 ± 0 .44
Dimethylmalonic acida	0.09 ± 0 .08	ND	ND	2-Hydroxyglutaric acid	1.46 ± 0 .20	ND	ND±
Malonic acida	0.10 ± 0 .12	0.27 ± 0 .16	0.07 ± 0 .04	Ribonic acida	0.90 ± 0 .92	0.94 ± 0 .65	0.25 ± 0 .22
Benzoic acida	2.30 ± 3 .89	0.15 ± 0 .04	0.32 ± 0 .12	Galacturonic acid 1a	ND	2.82 ± 1 .69	0.22 ± 0 .20
Butanoic acid	1.09 ± 0 .52	ND	0.33 ± 0 .31	Galacturonic acid 2a	2.36 ± 4 .56	ND	ND
Maleic acida	0.17 ± 0 .11	0.50 ± 0 .17	0.13 ± 0 .03	Galactonic acid	0.25 ± 0 .15	0.05 ± 0 .07	0.13 ± 0 .08
Succinic acida	0.27 ± 0 .17	0.49 ± 0 .10	0.34 ± 0 .15	Gluconic acida	ND	0.35 ± 0 .39	0.06 ± 0 .02
Propanoic, 1-oxo	0.12 ± 0 .08	0.46 ± 0 .11	5.90 ± 1 .49	Saccharica	12.45±11 .03	0.09 ± 0 .07	1.79 ± 1 .55
Fumaric acida	0.26 ± 0 .18	0.56 ± 0 .20	6.80 ± 5 .22	Galactaric acida	0.74 ± 0 .70	0.32 ± 0 .24	0.13 ± 0 .07
Malic acida	12.78 ± 9 .02	12.29±11 .83	ND	Unk SA 1 (MP)	0.26 ± 0 .19	4.72 ± 2 .69	ND
Threonic acida	5.34 ± 0 .55	4.66 ± 0 .11	0.52 ± 0 .31	Arabino-hexaric acid	0.54 ± 0 .37	0.98 ± 0 .46	0.38 ± 0 .14
Tartaric acida	ND	0.10 ± 0 .13	12.47 ± 3 .15	Glycerol-Glycoside	0.15 ± 0 .07	0.24 ± 0 .24	0.05 ± 0 .04
Shikimic acid	0.38 ± 0 .23	1.75 ± 2 .83	ND	Glucuronic acid	0.10 ± 0 .11	0.62 ± 0 .57	0.01 ± 0 .03
Citric acida	2.84 ± 0 .82	41.75±12 .08	5.17 ± 4 .94	Unk SA 2 (Val)	0.68 ± 0 .36	ND	ND
Isocitric acid	0.98 ± 0 .79	ND	ND	Unk SA 3 (Val)	1.63 ± 1 .80	ND	ND
Quinic acida	35.25 ±21 .27	10.14 ± 6 .05	0.06 ± 0 .03	Unk SA 4 (Val)	1.34 ± 2 .08	0.52 ± 0 .72	0.03 ± 0 .04
TOTAL	62.83 ± 38 .92	73.96 ± 34 .09	32.59 ± 16 .08	Unk SA 5 (Val)	3.68 ± 3 .14	ND	ND
Fatty acids				Unk SA 6 (Val)	4.26 ± 4 .30	ND	ND
Myristic acida	3.41 ± 2 .28	0.00 ± 0 .00	0.15 ± 0 .37	TOTAL	35±31 .94	11.76 ± 8 .08	6.3±3 .3
Palmitic acida	1.28 ± 0 .50	6.21 ± 2 .82	7.15 ± 2 .87	Phosphates
Stearic acida	4.45 ± 1 .15	9.74 ± 4 .41	8.51 ± 2 .84	Phosphoric acid	5.02 ± 4 .99	3.95 ± 0 .98	1.68 ± 0 .87
TOTAL	9.14 ± 3 .93	15.95 ± 7 .23	15.81 ± 6 .08	Glycerophosphate	0.00 ± 0 .00	0.00 ± 0 .00	ND
Phenolic compounds				Unk Sugar Phosphate	0.00 ± 0 .00	0.89 ± 0 .61	ND
Coumaric acid	1.63 ± 1 .42	0.13 ± 0 .06	0.04 ± 0 .04	TOTAL	5.02 ± 4 .99	4.84 ± 1 .59	1.68 ± 0 .87
Ferulic acida	0.68 ± 0 .37	2.97 ± 0 .86	0.14 ± 0 .07	Total metabolites	300.15 ± 212 .33	411.9 ± 147 .35	123.8±61 .06
Caffeic acid	0.41 ± 0 .12	0.06 ± 0 .04	0.05 ± 0 .02
TOTAL	2.72 ± 1 .91	3.16 ± 0 .96	0.23 ± 0 .13

^aCompound was identified by its identifier ions (m/z) and confirmed with references standard; otherwise tentatively identified by LRI and a matching score of 700 higher using NIST 2011 and Wiley 9^th ed. mass spectral databases and by comparison with entries in the Golm Metabolome Database (http://gmd.mpimp-golm.mpg.de/).

Amino acids

Nine other NPAAs were detected in the phloem saps. The total concentration of these NPAAs was 136.7 ± 47.2 mM, 12.6 ± 9.9 mM, 9.9 ± 7.8 mM in orange jasmine, Valencia sweet orange and curry leaf tree respectively (Table 1). The concentrations of the main NPAAs; putrescine, γ-Aminobutyric acid, and _L-Ornithine were higher in orange jasmine than Valencia and curry leaf.

In addition to the NPAAs, 17 PAAs were detected in the phloem saps. Although _L-Glutamine, _L-Leucine, _L-Lysine, _L-Methionine, and _L-Tyrosine standards were easily detected after TMS derivatization, none of them was detected in the phloem sap indicating that they were present below the limit of detection. Orange jasmine was the highest in total PAAs (55.5 ± 22.8 mM), followed by Valencia (10.4 ± 5.2 mM) and curry leaf (8.6 ±5 .9 mM). Essential amino acids (_L-Valine, L-threonine, _L-Phenylalanine, and _L-Isoleucine) were found at lower concentrations compared to NPAAs in all of the 3 host plants. In the same manner, orange jasmine was the highest in essential amino acids (8.4 ± 2.0 mM), followed by Valencia (3.3 ±1 .2 mM) and curry leaf (1.8 ±1 .05 mM). The concentrations of the essential amino acids (_L-Valine, _L-Threonine, _L-Phenylalanine, and _L-Isoleucine) in orange jasmine were higher than that found in Valencia sweet orange and curry leaf tree. In addition, _L-Proline, _L-Serine, _L-Alanine in orange jasmine were also higher than the other two hosts.

Organic acids, fatty acids and phenolic compounds

Nineteen organic acids were detected in the phloem saps (Table 1). The total amount of organic acids was 74.0 ± 34.1 mM, 62.8 ± 38.9, and 32.6 ± 16.1 mM in orange jasmine, Valencia sweet orange, and curry leaf tree respectively. Citric acid and maleic acid were higher in orange jasmine while quinic and butanoic acids were higher in Valencia sweet orange (Table 1). Interestingly, malic acid was not detected in curry leaf tree, whereas it contained high levels of tartaric and propanoic acid (Table 1).

Three fatty acids and 3 phenolic compounds were detected in the phloem sap (Table 1). Palmitic acid was lower in Valencia than orange jasmine and curry leaf tree (Table 1). Phenolics were found at low levels. Caffeic acid was highest in Valencia, whereas ferulic acid was highest in orange jasmine (Table 1).

Sugars, sugar alcohols, and sugar acids

Fifteen mono- and disaccharides were detected in the phloem saps (Table 1). The total concentrations of these sugars were the highest 112.44 ± 84.16 mM, 81.1 ± 33.6 mM, and 36.4 ± 16.8 mM in Valencia sweet orange, orange jasmine and curry leaf tree respectively. Glucose and fructose were the main monosaccharides in the phloem saps. Fructose was not significantly different among the 3 plants. However, glucose level in curry leaf tree was lower than Valencia (Table 1). Arabinose was found in trace amounts in Valencia, but was not detected in orange jasmine and curry leaf tree (Table 1). Mannose was also found at low concentration (<1.2 mM) and its level in orange jasmine was higher than Valencia sweet orange and curry leaf tree (Table 1). Sucrose was the main disaccharide and it was not significantly different among the 3 species. Turanose level in Valencia was higher. (Table 1). Maltose was found in trace amounts and its level in Valencia was higher than orange jasmine and curry leaf tree.

Ten sugar alcohols were detected in the phloem sap (Table 1). The total concentrations of sugars alcohols in Valencia sweet orange, orange jasmine, and curry leaf tree were 49.9 ± 31.3, 28.8 ± 14.6, and 12.3 ± 4.1 mM, respectively. Interestingly, the levels of the 3 inositol isomers (chiro-, scyllo-, and myo-inositol) in curry leaf tree were lower than Valencia. The level of these inositol isomers in orange jasmine was moderate; slightly higher than curry leaf tree and lower than Valencia. Other sugar alcohols were found in low amounts. The erythritol level in orange jasmine was higher than Valencia and curry leaf tree. Glucitol was only found in Valencia (Table 1).

Twenty sugar acids were detected in the derivatized samples. The total concentrations of these sugar acids were 35.0 ± 32.0 mM, 11.8±8.0 mM, and 6.3 ± 3.3 mM in Valencia sweet orange, orange jasmine, and curry leaf tree. Glucaric and 2-hydroxyglutaric acids were detected in only in Valencia sweet orange. In addition, 2-ketoglutaric acid in Valencia was higher than orange jasmine and curry leaf tree (Table 1).

Discussion

Insect survival and development are negatively affected by the presence of toxic compounds and nutritional deficiency of the host plant.¹¹ The nutritional status of the host plant, especially the concentration of amino acids plays an important role in the development of D. citri.⁷ Because essential amino acids are found at low amounts in the phloem sap compared to sugars, phloem sap sucking insects consume huge amounts of this fluid in order to get more of these essential amino acids.¹² In addition, to overcome amino acid deficiency, phloem sap sucking-insects also depend on their symbiotic bacteria as another source for amino acids.¹²

In addition to their role in the synthesis of protein, amino acids are an important source of energy ¹³. Amino acids can be converted to trehalose (synthesis of trehalose by gluconeogenesis) or directly oxidized during flight to produce energy.¹³ Alves et al. (2014) suggested that the nutritional differences among the host plants affect the survival and the development of D. citri.⁹ Our current results confirmed the presence of nutritional differences among the 3 tested hosts of D. citri.

Interestingly, the phloem sap of curry leaf tree was lower in essential, non-essential, and total amino acids than Valencia and orange jasmine. On the other hand, the concentrations of putrescine and _L-ornithine in orange jasmine were higher than Valencia and curry leaf. These polyamines could also be beneficial for D. citri.

Feeding studies showed that polyamines were found to increase the lifespan of fruit flies, worms and yeast.¹⁴ The nutrient inadequacy of the phloem sap in curry leaf tree, especially the amino acids, could be the reason behind the longer life cycle and the lower survival of D. citri on this host.

Amino acids present in the phloem sap are also important for the growth and multiplication of CLas. Genome sequencing of CLas showed that CLas has the tricarboxylic acid (TCA) cycle genes and hence it can use a wide range of amino acids to produce energy.¹⁵ CLas can metabolize many amino acids including aspartate, methionine, serine, alanine, arginine, proline, threonine, tryptophan, glutamate, glycine, tyrosine, cysteine, histidine, and phenylalanine.¹⁵ Our results showed that the levels of alanine, threonine, and phenylalanine in curry leaf were lower than orange jasmine. On the other hand, CLas genome sequencing indicated that CLas is not able to synthesize leucine, isoleucine, tryptophan, valine, and tyrosine and from their intermediates.¹⁵ Only isoleucine and valine were detected in the phloem sap after TMS derivatization which indicated that these amino acids are found in very low concentration. Valine level in curry leaf was also lower than orange jasmine. Our previous study showed that methyl chloroformate (MCF) derivatization was better than TMS for determination of amino acids because MCF does not react with sugars which are abundant in the phloem sap.¹⁰ The high level of putrescine, and L-ornithine in orange jasmine could also be beneficial for CLas; however, additional research is needed to test this assumption.

Many organic acids including malic, maleic, succinic, citric, and fumaric were detected in the phloem sap. These compounds are intermediates in the TCA cycle so they could also be incorporated in the TCA cycle to produce energy. Malate, succinate, citrate, and fumarate also play an important role in proline metabolism and synthesis.¹³ Succinic, citric, quinic, and malic acid were detected in the honeydew of D. citri.¹⁶ The level of malic acid in honeydew was relatively high, but the other organic acids were found in low amounts.¹⁶ This result indicated that organic acids, especially those present in low concentration in the phloem sap, could be a limiting factor for D. citri development. The low levels of the organic acids in curry leaf tree could also be responsible for the low survival and longer life cycle of D. citri. Interestingly, citric acid which is an intermediate in the TCA in was high in orange jasmine.

In the same manner, CLas can directly integrate many of the organic acids (malic, succinic, citric, and fumaric) in the TCA cycle and use them as a source of energy. Malate synthase and isocitrate lyase are absent in the CLas genome, and therefore CLas should use exogenous fumarate, malate, succinate, and aspartate as carbon substrates for the TCA cycle.¹⁷ The low levels of organic acids in curry leaf could also be another limiting factor for the growth of CLas. Interestingly, a high amount of tartaric acid was found in curry leaf tree while malic acid was absent. Malic acid is the main organic acid in most plants; however, occasionally tartaric, oxalic, or citric acid is the most dominant organic acid.¹⁸ The presence of tartaric acid in high amounts in the phloem sap of curry leaf tree could inhibit the TCA cycle in CLas. In fact, tartaric acid inhibits the fumarase enzyme which converts fumarate to malate and consequently results in a low quantity of malic acid and malfunction of the TCA cycle.¹⁹

In general sugars are found in high levels in the phloem sap hence phloem sap-sucking insects must be able to eliminate the extra sugar and tolerate the high osmotic pressure of the phloem sap. The total sugars in curry leaf was lower than orange jasmine and Valencia. Since D. citri produce large amounts of honeydew on curry leaf tree, it is unlikely that sugars are the limiting factor for D. citri survival and development. The level of total sugar alcohols in curry leaf was also lower than Valencia and orange jasmine. Inositol is required for immature stages of some locust species.²⁰ The level of sugar acids was also lower in curry leaf tree; however, no information is available about the role of sugar acids in the development of insects.

Sequencing of CLas showed that glycolysis was the major pathway for the catabolism of monosaccharides and it could metabolize fructose, glucose, and xylose.¹⁵ Although the level of glucose in curry leaf was lower than orange jasmine and Valencia, it is not expected to be the limiting ingredient for CLas growth because its level in curry leaf was only marginally lower than orange jasmine. In fact, some gram-negative α-proteobacteria such as the pathogenic strains of Cronobacter can utilize sorbitol, mannitol and inositol as carbon sources.²¹ Scyllo-inositol was negatively correlated with citrus tolerance to CLas (non-published data). Sugar acids in Valencia and orange jasmine were also higher than those curry leaf. Sugar acids were also negatively correlated with citrus tolerance to CLas (non-published data).

In conclusion, the phloem sap of curry leaf tree contains lower amounts of amino acids, organic acids, sugars, and total metabolites than Valencia and orange jasmine but contains higher amount of tartaric acid. In addition, previous studies showed that curry leaf tree contains many compounds with antimicrobial and anti-fungal activity such as murrayanine, girinimbine and mahanimbine.²² This may explain why curry leaf tree does not harbor CLas and it is not an excellent host for D. citri.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Acknowledgments

This work was supported with grant # 2016-70016-24844 for NK received from SCRI-NIFA-USDA. I thank Shelley E Jones for the technical assistance and the English revision of the manuscript.