Citrus-greening (Huanglongbing)-induced small RNAs are potential early diagnosis markers

Abstract

The present invention provides compositions and methods for detecting Candidatus Liberibacter infection and Huanglongbing disease in a citrus plant by detecting the expression of small RNAs such as miRNA and siRNA. The invention also provides methods for treating Huanglongbing disease in a citrus plant by contacting the plant with a phosphorus containing solution.

Claims

1. A method for detecting the level of expression of one or more RNA in a sample from a citrus plant, the method comprising detecting a miRNA399 RNA in the sample by (i) contacting a sample from a citrus plant suspected of being infected with HLB disease or having a Ca. L. asiaticus-infection with a nucleic acid that specifically hybridizes to the RNA, (ii) contacting a sample from a citrus plant that is not infected with HLB disease or does not have a Ca. L. asiaticus-infection with a nucleic acid that specifically hybridizes to the RNA, detecting increased expression of the miRNA399 RNA in the sample from the citrus plant suspected of being infected with HLB disease by a method selected from Northern analysis, polymerase chain reaction (PCR), reverse-transcription-PCR (RT-PCR), or hydrizing the RNA to a microarray, and contacting the citrus plant suspected of being infected with HLB disease with phosphate or a phosphorus oxyanion; wherein the miRNA399 comprises a sequence selected from UGCCAAAGGAGAUUUGCCCGG (SEQ ID NO:9), UGCCAAAGGAGAGUUGCCCUA (SEQ ID NO:10), UGCCAAAGGAGAAUUGCCCUG (SEQ ID NO:11), or UGCCAAAGGAGAGUUGCCCUG (SEQ ID NO:12).

2. The method of claim 1, further comprising detecting an RNA selected from the group consisting of siRNA1005, siRNA1008 and siRNA1009 in the sample from (i), wherein the siRNA1005 comprises a sequence at least 90% identical to ATAGATAATGGATCAACGGTTATA (SEQ ID NO:13); the siRNA1008 comprises a sequence at least 90% identical to TCGAACAAGGTAAGGATGTCA (SEQ ID NO:14) or CCTTGTTCGAACAAGGTAAGGATGTCATTCTTT (SEQ ID NO:100); and the siRNA1009 comprises a sequence at least 90% identical to CTTCTAATAAACATGCATGAA (SEQ ID NO:15) or CGTCTTCTAATAAACATGCATGAACTTATT (SEQ ID NO:101).

3. The method of claim 1, wherein the method further comprises detecting the mRNA of a ubiquitin-conjugating enzyme E2 (UBC) gene.

4. The method of claim 3, wherein the UBC mRNA comprises a sequence that is at least 90% identical to SEQ ID NO:84.

5. The method of claim 1, wherein the method further comprises measuring phosphate levels in the plant.

6. The method of claim 1, wherein the nucleic acid is labeled with a detectable probe.

7. The method of claim 1, wherein the detecting is performed using a nucleic acid dipstick.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows expression levels of some conserved citrus miRNAs are affected by Las infection. Relative expression levels of some conserved citrus miRNAs were examined at 10 wpi (a) and 14 wpi (b). Expression levels are presented as reads in Las-treated samples over corresponding untreated samples. “+1” indicates a 2-fold induction; “−1” indicates a 2-fold reduction; HLB: Las-infected; wpi: weeks post inoculation. Red bar highlights miR399, which is involved in citrus Pi accumulation. The miRNA IDs are listed at the left side.

(2) FIG. 2 is a graph showing Ca. L. asiaticus-induced miR399 down-regulates its target, a E2-conjugating enzyme gene UBC. UBC mRNA was measured by real-time RT-PCR and actin was used as an internal control. Similar results were obtained from two biological replicates.

(3) FIG. 3 shows that Las infection causes citrus phosphorus and iron deficiency. 15-20 leaves from both untreated and Las-treated trees were collected. For the treated plants, both asymptomatic and symptomatic leaves were collected. For symptomatic leaves a combination of blotchy mottled and small chlorotic leaves of different ages was collected. For asymptomatic and control leaves, a combination of older and younger leaves was collected (if possible). Amount of iron, phosphorus, potassium, zinc, and carbon was examined (mean±SE). Amount of each element in untreated samples was assigned to 1. Experiments were repeated for 2 times with similar results.

(4) FIG. 4 shows Las infection down-regulates csi-PHO2 but up-regulates Phosphate transporters (PTs). Expression levels of csi-PHO2 (UGID: 1423690), csi-PT2 (UGID: 3374895), and csi-PHT2;1 (UGID: 2916608) relative to citrus actin were determined by quantitative real-time PCR (mean±SE). Total RNA from both 10- and 14-wpi samples (untreated and Las-treated) was used for reverse transcription followed by quantitative PCR. Expression level of untreated samples is assigned to 1. Experiments were repeated for 3 times with similar results. HLB: Las-infected; wpi: weeks post inoculation.

(5) FIG. 5 shows the regulatory circuit mediated by miR399 in response to Ca. L. asiaticus infection.

DETAILED DESCRIPTION OF THE INVENTION

Introduction

(6) Citrus greening or “Huanglongbing” (HLB), caused by bacteria Candidatus Liberobacter, is one of the most destructive diseases of citrus. Candidatus Liberibacter (Ca. Liberibacter or Ca. L.) is a Gram negative bacterial pathogen restricted to the phloem. The uneven distribution within trees and the latency of detectable symptoms make detection and confirmation of asymptomatic infections very difficult. Therefore, developing early diagnosis biomarkers and effective reagents is an urgent need for the citrus industry, especially for those in the threatened but un-infected regions, such as California. The recent detection of Psyllids (the insect vector for pathogen infection) at the California-Mexico border underlines the importance of the current invention.

(7) To prevent its further spread, early diagnosis before the appearance of the dreaded symptoms is particularly important. However, the unculturable nature of the bacteria and their low concentration and uneven distribution in the hosts make it extremely difficult to detect Ca. L. infection. Instead of focusing on the bacteria, the disclosure provides a method of early diagnosis by taking advantage of host rapid defense responses to identify unique host biomarkers.

(8) Some host small RNAs are rapidly and specifically induced by pathogens, which makes them one of the most attractive markers for early diagnosis. To identify HLB specific small RNAs, endogenous small RNAs were profiled by high-throughput sequencing of small RNA libraries prepared from HLB positive and uninfected control plants. Those small RNAs that are induced by HLB infection have the potential to serve as early detection diagnosis markers for HLB.

(9) The disclosure also provides the identification of a major cause for the pathogenesis of HLB—phosphorus starvation—from the study of one HLB-induced miRNA, microRNA399 (miRNA399), which is one of the potential early diagnosis markers. miRNA399 is induced specifically in response to Ca. Liberibacter infection, but not to other bacterial pathogen, such as Spiroplasma citri, the causal agent for citrus stubborn disease that has similar symptoms. miRNA399 induction is an important response to phosphorous starvation. The level of phosphorus in Ca. L.-infected plants was only 60-70% of that in uninfected plants. Phosphorus application can largely rescue the symptoms and increase the yield of fruits.

(10) Accordingly, the disclosure provides methods and compositions for treating Ca. L. infection comprising contacting the plant with a phosphorus oxyanion solution. The disclosure also provides methods for treating Ca. L. infection comprising down-regulating expression of the ubiquitin-conjugating enzyme (UBC) gene, which subsequently derepresses phosphorus transporters and increases phosphorus uptake.

(11) By profiling small RNAs from untreated and Ca. Liberibacter-infected Citrus sinensis (sweet orange) samples using high throughput deep sequencing, some Huanglongbing-induced small interfering RNAs (siRNAs) and microRNAs (miRNAs) were identified, providing a method of early diagnosis for infection. Furthermore, one of these markers, miR399, is induced by phosphorus starvation. It targets ubiquitin E2 conjugating enzyme genes involved in phosphorus uptake, which has been demonstrated in Arabidopsis thaliana. This result indicates that phosphorus starvation is one of the causes for HLB disease symptom. Indeed, a markedly decreased level of phosphorus in Ca. Liberibacter infected citrus was observed. Ca. Liberibacter infection causes phosphorus starvation, and subsequently leads to quick induction of miR399, which in turn silences UBC and derepresses the phosphorus transporters and facilitates phosphorus accumulation. Thus, the disclosure also provides a method of treating HLB by applying phosphorus oxyanion solutions (including phosphite and phosphate salt solutions with polymeric forms) to Ca. Liberibacter-infected plants to reduce HLB symptoms.

(12) The disclosure provides a method of determining infection of a plant or tree (e.g., a citrus plant or tree) comprising determining the level of miR399, siR1005, siR1008, and siR1009 in a sample from the plant or tree, wherein an increased level compared to an uninfected or healthy plant indicates that the plant or tree may be infected by Ca. Liberibacter.

(13) The disclosure also provides a method of slowing or inhibiting the spread, growth or infection of a tree or plant with Ca. Liberibacter comprising contacting a plant with HLB disease or infected with Ca. L. with phosphorus oxyanions (which can be in solution), stimulating phosphorus uptake in the plant by suppressing expression of UBC mRNA or UBC polypeptide in the plant, thereby ameliorating symptoms of HLB disease in the plant.

(14) Detection of Infection or Disease

(15) The present methods detect the expression of small RNAs in citrus plants that are induced by infection with bacterial pathogens. In one embodiment, the bacterial pathogen is Ca. Liberibacter. In one embodiment, the bacterial pathogen is Ca. Liberibacter asiaticus (Las).

(16) In some embodiments, infection of a citrus plant by Ca. L. induces or increases the expression level of some small RNAs. In some embodiments, infection of a citrus plant by Ca. L. induces or increases the expression of miRNA399, siRNA1005, siRNA1008 and/or siRNA1009 compared to a control or healthy plant.

(17) In some embodiments, the small RNA molecules include the sequences shown in Table 1.

(18) TABLE-US-00001 TABLE 1 Sequences of small RNA molecules that are increased following infection with Ca. L. SEQ ID small RNA Sequence NO: csi-miR399a: UGCCAAAGGAGAUUUGCCCGG  9 csi-miR399b: UGCCAAAGGAGAGUUGCCCUA 10 csi-miR399c: UGCCAAAGGAGAAUUGCCCUG 11 csi-miR399d: UGCCAAAGGAGAGUUGCCCUG 12 csi-miR399e: UGCCAAAGGAGAAUUGCCCUG 11 csi-siR1005: ATAGATAATGGATCAACGGTTATA 13 csi-siR1008: TCGAACAAGGTAAGGATGTCA 14 csi-siR1009: CTTCTAATAAACATGCATGAA 15

(19) As is well known in the art, in RNA uracil (U) replaces thymide (T). Thus, T can be represented by T or U, depending if the sequence is DNA or RNA.

(20) As is understood in the art, the sequence of siRNAs can vary at the 5′ and 3′ ends. Thus, for the detection of siR1008 and siR1009, the sequence detected can be shifted either 5′ or 3′ relative to SEQ ID NOs: 14 and 15. Thus, for siR1008, the detected sequence can include a sequence from CCTTGTTCGAACAAGGTAAGGATGTCATTCTTT (SEQ ID NO:100), where SEQ ID NO:14 is underlined. Likewise, for siR1009, the detected sequence can include a sequence from CGTCTTCTAATAAACATGCATGAACTTATT (SEQ ID NO:101), where SEQ ID NO:15 is underlined.

(21) In some embodiments, infection of a citrus plant by Ca. L. reduces or decreases the expression level of some small RNAs. In some embodiments, infection of a citrus plant by Ca. L. reduces or decreases the expression level of a small RNA selected from miR408, miR171, miR396, miR398, miR160, and/or miR394 compared to a control or healthy plant.

(22) FIG. 1 shows miRNAs that have increased expression levels in citrus plants infected with Las. Thus, in some embodiments, infection of a citrus plant by Ca. L. induces or increases the expression level of a micro RNA of Table 2.

(23) TABLE-US-00002 TABLE 2 miRNAs showing a relative increase in expression in citrus plants infected by Ca. L. miR399 miR159 miR393

(24) FIG. 1 also shows that infection of a citrus plant by Ca. L. reduces or decreases the expression level of other micro RNAs. Thus, in some embodiments, infection of a citrus plant by Ca. L. reduces or decreases the expression level of a microRNA shown in Table 3. In some embodiments, infection of a citrus plant by Ca. L. reduces or decreases the expression level of a small RNA selected from miR408, miR171, miR396, miR398, miR160, and/or miR394.

(25) TABLE-US-00003 TABLE 3 miRNAs showing relative decrease in expression in citrus plants infected by Ca. L. miR160 miR164 miR398 miR394 miR403 miR396 miR171 miR408 miR172

(26) As is well understood in the art, miRNA molecules are thought to degrade target mRNA or inhibit translation of RNA by precise or imprecise base-pairing with their target RNA molecules. Thus, the sequences of miRNA molecules can vary slightly from their target sequence and still function to inhibit gene expression. Further, the sequences of miRNA and siRNA molecules can vary at either the 5′ or 3′ end by the addition or subtraction of one or two nucleotides and still function to inhibit expression of a given target. Thus, the methods encompass detecting miRNA and siRNA molecules that have slight variations in the nucleic acid sequence.

(27) For example, as shown in Table 4, different sequences were identified for miRNA399 in both uninfected control and Ca. L. infected citrus plants. The miR399 sequences shown in Table 4 are capable of targeting (base-pairing with) the mRNA that encodes the citrus UBC protein (accession number EY742134) or an mRNA substantially identical to the mRNA that encodes accession number EY742134. In one embodiment, the miR399 sequence targets a sequence that is substantially identical to SEQ ID NO:84.

(28) TABLE-US-00004 TABLE 4 Citrus miRNA399 sequences and abundance upon Ca. L. asiaticus infection. SEQ Citrus small RNA sequence ID Alignment (reverse-complemented if NO: Reads subject orientation is -) Untreated   2  2 ath-miR399b TGCCAAAGGAGAGTTGCCCTA 10 wpi  1  1 ath-miR399b TGCCAAAGGAGAGTTGCCCTG HLB  1 12 ath-miR399b TGCCAAAGGAGAGTTGCCCTG 10 wpi  2 10 ath-miR399b TGCCAAAGGAGAGTTGCCCTA  3  1 ath-miR399c TGCCAAAGGAGCGTTGCCCTG  4  1 ath-miR199b TGCCAAAGGAGAGTTGCCATG  5  1 ath-miR399c TGTCAAAGGAGAGTTGCCCTG Untreated  2  2 ath-miR399b TGCCAAAGGAGAGTTGCCCTA 14 wpi  8  2 ath-miR399f TGCCAAAGGAGATTTGCCCGG  6  1 ath-miR399a TGCCAAAGGAGAATTGCCCTG HLB  2 23 ath-miR399b TGCCAAAGGAGAGTTGCCCTA 14 wpi  1  8 ath-miR399b TGCCAAAGGAGAGTTGCCCTG  7  1 ath-miR399c TGCTAAAGGAGAGTTGCCCTA 99 consensus TG C/T C/T AAAGGAG A/C G/T/A TTGCC miR399 C/A T/G A/G

(29) Thus, in some embodiments, the method comprises detecting the expression level of a sequence that is substantially identical to one of the miR399 sequences in Table 4. For example, in some embodiments, a sequence that is at least 60%, 70%, 80%, 85%, 90%, or 95% identical to a sequence in Table 4 is detected. In one embodiment, the method comprises detecting a small RNA having the miR399 consensus sequence TG X.sub.1 X.sub.2 AAAGGAG X.sub.3 X.sub.4 TTGCC X.sub.5 X.sub.6 X.sub.2 (SEQ ID NO:99), where X.sub.1 is C or T, X.sub.2 is C or T, X.sub.3 is A or C, X.sub.4 is G, T, or A, X.sub.5 is C or A, X.sub.6 is T or G, and X.sub.2 is A or G.

(30) In some embodiments, infection of a citrus plant by Ca. L. can be detected by detecting the expression level of a sequence that is substantially identical to an siRNA listed in Table 5. For example, in one embodiment, the expression level of a sequence that is at least 80%, 85%, 90%, or 95% identical to an siRNA listed in Table 5 is detected.

(31) TABLE-US-00005 TABLE 5 HLB induced citrus siRNAs. reads with highest copy number untreated HLB H/U 10 untreated HLB H/U (SEQ ID NO:) 10 wpi 10 wpi wpi 14 wpi 14 wpi 14 wpi Total TTCCAGATAGAAGGCCACTCA (42) 1.0 736.0 736.0 1.0 106.0 106.0   844 TTCCACCAATCGATCAGGATA (43) 1.0 233.0 233.0 1.0 89.0 89.0   324 GCGTATGAGGAGCCATGCATA (44) 1.0 155.0 155.0 1.0 115.0 115.0   272 CTTGGATTTATGAAAGACGAA (45) 1.0 127.0 127.0 1.0 1.0 1.0   130 GGCAGGGCTAGTGACTGGAGTGA 1.0 117.0 117.0 1.0 37.0 37.0   156 (46) ACAGGCCGCAAACATTTTCCT (47) 1.0 87.0 87.0 42.0 143.0 3.4   273 ACAGACCGCACACCTTTTCTT (48) 1.0 59.0 59.0 18.0 62.0 3.4   140 ATTAGGAGCTAAAATTGTTGT (49) 1.0 45.0 45.0 24.0 111.0 4.6   181 ACGAAATGTGAGTAGAGTGGACAG 1.0 45.0 45.0 113.0 135.0 1.2   294 (50) TTCCAAAGGGATCGCATTGA (51) 1.0 42.0 42.0 1.0 77.0 77.0   121 TCATTTAAGGGTTTCGTGTTC (52) 1.0 34.0 34.0 1.0 115.0 115.0   151 ACGCTCGGACGAAGCACATAGATG 1.0 29.0 29.0 58.0 40.0 0.7   128 (53) TATGGGATTTACCTCGGCAAA (54) 2.0 55.0 27.5 1.0 48.0 48.0   106 TGTGTGGATGAATAAGATTTC (55) 8.0 210.0 26.3 7.0 193.0 27.6   418 TATCTGGATAAAAGGCTACCC (56) 212.0 5229.0 24.7 211.0 3480.0 16.5  9132 TCATGGATAAGGTCATGCATT (57) 6.0 137.0 22.8 6.0 188.0 31.3   337 AAAAACTTGGAAGCGTTGGAT (58) 8.0 154.0 19.3 10.0 62.0 6.2   234 TCCTGCCGGGTTGCATAATCA (59) 6.0 85.0 14.2 1.0 19.0 19.0   111 ATAGATAATGGATCAACGGTTATA 26.0 359.0 13.8 37.0 321.0 8.7   743 (60) TCATGGATAAGGTCATGCATC (61) 25.0 308.0 12.3 24.0 403.0 16.8   760 CTGAAAGCTGAGGTTGTCCTT (62) 8.0 44.0 5.5 25.0 30.0 1.2   107 AGTGTCAAAAAGAGCAATGGCGTC 11.0 44.0 4.0 25.0 22.0 0.9   102 (63) AATCCTTGGATTAGGAGTGTGGAG 1.0 4.0 4.0 1.0 112.0 112.0   118 (64) ATCAATAAATCAGGATTGGCGGAA 82.0 288.0 3.5 71.0 216.0 3.0   657 (65) CGTTAGGGAGTCCGGAGACGT (66) 13.0 44.0 3.4 20.0 37.0 1.9   114 GAATAAGACATGGAGTTGGAA (67) 18.0 51.0 2.8 48.0 64.0 1.3   181 AGGAAATGGACGATACGGACGCAT 70.0 176.0 2.5 91.0 1.0 0.0   338 (68) TCAAGTGAGGTTCGGTCTTTGAA 29.0 69.0 2.4 22.0 26.0 1.2   146 (69) TAATCGTGGGAGACGAAGCTG (70) 2184.0 5128.0 2.3 2352.0 2587.0 1.1 12251 CGAAGGTCCGAGGTCGAGGTT (71) 68.0 154.0 2.3 82.0 1.0 0.0   305 AGGTTTGGGCTTGTTGCAAGTAGA 27.0 61.0 2.3 45.0 25.0 0.6   158 (72) TCCGGGCGGAAGACATTGTCA (73) 49.0 109.0 2.2 37.0 76.0 2.1   271 AACGGAAAGAACACAACACGG (74) 736.0 1592.0 2.2 924.0 673.0 0.7  3925 TGTTAGCTTTCTCGGACGCAG (75) 24.0 48.0 2.0 16.0 37.0 2.3   125 TCAAGTGAGGTTCTGTCTTTGA (76) 12.0 24.0 2.0 8.0 74.0 9.3   118 AAAGCAACGATTGTATGGCCA (77) 44.0 87.0 2.0 33.0 25.0 0.8   189 TCAAGTGAGGTTCGGTCTTGA (78) 146.0 260.0 1.8 116.0 253.0 2.2   775 GAATGTGGAATTAAGCGCACCAAA 259.0 455.0 1.8 23.0 71.0 3.1   808 (79) TCGAACAAGGTAAGGATGTCA (80) 197.0 301.0 1.5 180.0 577.0 3.2  1255 CTGGATGCAACTGTGGTACGG (81) 67.0 76.0 1.1 100.0 236.0 2.4   479 GGTGCTTCCGGATCTCAGGAT (82) 1.0 1.0 1.0 1.0 312.0 312.0   315 CACATGGGTTAGTCGATC (83) 1.0 1.0 1.0 1.0 750.0 750.0   753 H/U = ratio of reads from HLB diseased plants and uninfected plants.

(32) In some embodiments, infection of a citrus plant by Ca. L. is determined by detecting the expression of one, two, three, four, five or more of the small RNAs from Tables 1, 2, 3, 4 or 5. In some embodiments, infection of a citrus plant by Ca. L. is determined by detecting the expression of one, two, three, four, five or more small RNAs that are substantially identical to a small RNA from Tables 1, 2, 3, 4 or 5.

(33) Methods of detecting small RNAs are well known in the art and such methods can be adapted to detect miRNA399, siRNA1005, siRNA1008 and/or siRNA1009. In some embodiments, detection can include methods involving hybridization of nucleic acids to small RNAs by base-pairing. Examples include Northern analysis, polymerase chain reaction (PCR), reverse-transcription-PCR (RT-PCR) and microarrays. In some embodiments, the small RNA is detected by DNA oligonucleotides in Northern blot analysis. In some embodiments, the small RNA is detected using a locked nucleic acid (LNA) probe in Northern blot analysis.

(34) In some embodiments, the amount of a small RNA detected is compared to the amount of small RNA detected in a control, uninfected plant to determine if the plant is infected by Ca. L. In some embodiments, the amount of a small RNA detected is compared to a reference value that corresponds to or is indicative of the level of expression of the particular small RNA in a plant that is not infected by Ca. L. In some embodiments, the expression of one, two, three, four, five or more small RNAs is detected.

(35) In some embodiments, the small RNA is detected using a nucleic acid “dipstick” or other rapid detection device.

(36) In some embodiments, the small RNA is detected by sequencing the isolated RNA. Thus, in some embodiments, the small RNA is cloned and sequenced to determine the nucleic acid sequence of the small RNA, thereby detecting the expression of the small RNA in a plant. In some embodiments, the expression of a small RNA is detected by determining the number of sequencing reads that correspond to the individual small RNA sequence. In one embodiment, the number of sequencing reads that correspond to an individual small RNA sequence is compared in infected and uninfected control plants, thereby providing an indication that the infection increased or decreased expression of the small RNA.

(37) In some embodiments, the small RNAs were cloned and sequenced as described in the Examples. For example, total RNA is isolated from a plant tissue, 18-28 nucleotide fragments are recovered, and the purified fragments are ligated to adaptor oligonucleotides at the 5′ and 3′ ends. In one embodiment, the adaptor oligonucleotides serve as binding sites for PCR primers. The RNA fragments with the adaptor oligonucleotides are reverse transcribed and amplified by PCR. The PCR amplified products are sequenced to detect the small RNA molecules expressed by the plant.

(38) The sequencing results revealed that the small RNAs induced by Las infection were from about 18 to about 28 nucleotides in length. In some embodiments, the small RNAs were identified as microRNAs (miRNAs) by aligning the sequence with conserved miRNAs in other plant species (less than or equal to 2 mismatches to a conserved miRNA). In some embodiments, the small RNAs were identified as miRNAs based on whether their precursor RNAs can form stem-loop structures. In some embodiments, the small RNA was identified as a putative small interfering RNA (siRNA) because it did not match any conserved plant miRNA or its precursor RNA did not form stem-loop structures.

(39) In some embodiments, the small RNA comprises a 21, 22, 23, or 24 nucleotide species.

(40) In some embodiments, increased expression is detected when the expression level of a small RNA is at least 10%, 20%, 50%, 100%, 500% or 1000% or more than the expression level detected in a control or uninfected plant. In some embodiments, decreased expression is detected when the expression level of a small RNA is less than 10%, 20%, 50%, 100%, 500% or 1000% or less than the expression level detected in a control or uninfected plant. In some embodiments, the control level corresponds to the expression level of a particular small RNA in a healthy plant that does not have HLB disease. In some embodiments, the control level is a reference value or average value that corresponds to or is indicative of the level of expression of the particular small RNA in a plant or population of plants that is not infected by Ca. L.

(41) Samples

(42) In some embodiments, the expression of a small RNA is detected in a biological sample from a citrus plant. For example, the biological sample can comprise bark or a leaf from an infected or control plant.

(43) Methods of Treating

(44) The present disclosure also provides methods of ameliorating the symptoms of HLB infection of citrus plants. The methods described herein can be used to reduce symptoms caused by HLB infection, including yellowing of leaves, blotchy mottle of the leaves, zinc-deficiency-like mottle, severe chlorosis, and reduced fruit yield. It will be understood that symptoms of HLB vary according to the time of infection, stage of the disease, tree species, and tree maturity, among other things. It will be further understood that the disclosed methods do not necessarily result in eradication or cure of the infection, but can significantly reduce the symptoms caused by HLB infection.

(45) Thus, in some embodiments, the methods provided herein reduce the symptoms of HLB by reducing the yellowing of leaves, resulting in a greener appearance, increasing the growth rate of the plant, and/or increasing the fruit yield of the plant. Thus, in some embodiments, the fruit yield is improved by 5%, 10%, 20%, 30%, 40%, 50%, 60% 70%, 80%, 90%, 100%, 200%, 500% of more compared to a plant that is not treated according to the methods. In some embodiments, the fruit yield is increased to 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% of the yield of a similar plant that was not infected by HLB.

(46) In some embodiments, the methods involve contacting an infected plant with inorganic phosphite and/or phosphate (Pi). In one embodiment, the inorganic phosphite and/or phosphate is in a solution. In some embodiments, the plant is contacted with a phosphorus solution comprising phosphorus oxyanion solutions (including phosphate and phosphate salt solutions). In one embodiment, the phosphorus solution comprises mixed mono- and dipotassium salts of phosphorus acid, with polymeric forms. In one embodiment, the inorganic phosphorus solution is applied by foliar spray.

(47) In some embodiments, the method of treatment can include transforming a plant with an expression cassette that expresses a small RNA described herein. For example, in one embodiment, a citrus plant can be transformed with an nucleic acid that expresses miR399. While not being bound by theory, it is believed that expression of miR399 regulates a conserved phosphate homoestasis pathway, which results in increased transport of Pi into the plant cells. Thus, in one embodiment, the method comprises overexpressing miR399 in a plant. In one embodiment, the method comprises stimulating phosphate fixation or accumulation in the plant.

(48) In some embodiments, the method of treatment includes suppressing or decreasing the expression or activity of a ubiquitin conjugating enzyme (UBC), for example ubiquitin-conjugating enzyme E2. In one embodiment, the citrus ubiquitin conjugating enzyme has accession number EY740382. In one embodiment, the ubiquitin conjugating enzyme is a citrus homolog of the Arabidopsis PHO2 (phosphate 2) ubiquitin conjugating enzyme E2. In one embodiment, the citrus UBC is encoded by the sequence shown in SEQ ID NO:84.

(49) In one embodiment, the method of treatment includes increasing the expression or activity of Pi transporters. For example, in some embodiments, the Pi transporters are citrus homologs of the Arabidopsis thaliana phosphate transporter 2 (AtPT2) and phosphate transporter 2;1 (AtPHT2:1). In some embodiments, the citrus phosphate transporter homologs are represented by Unigene Csi: 14938 (UGID: 3374895) and Unigene Csi: 9842 (UGID: 2916608).

(50) Expression Cassettes

(51) In some embodiments, the present invention provides for expression cassettes comprising a promoter operably linked to a polynucleotide encoding a small RNA of the invention (e.g., as described herein), wherein introduction of the expression cassette into a plant results in the plant expressing a small RNA as described herein. In some embodiments, the promoter is heterologous to the polynucleotide. In some embodiments, the promoter is inducible. In some embodiments, the promoter is tissue-specific.

(52) In some embodiments, introduction of the expression cassette into a plant results in the plant having decreased expression or activity of a UBC as compared to a plant lacking the expression cassette. In one embodiment, the introduction of the expression cassette into a plant results in the plant having decreased expression or activity of the citrus homolog of ubiquitin conjugating enzyme E2. In some embodiments, introduction of the expression cassette into a plant results in the plant having increased expression or activity of Pi transporters as compared to a plant lacking the expression cassette. In one embodiment, introduction of the expression cassette into a plant results in the plant having increased expression or activity of the citrus homolog of phosphate transporter 2 (AtPT2) and phosphate transporter 2;1 (AtPHT2:1).

(53) In another embodiment, the present invention provides for expression vectors comprising an expression cassette of the invention (e.g., as described herein).

(54) Plants

(55) In some embodiments, the plant is a citrus plant. In some embodiments, the citrus plant is an orange tree, a lemon tree, a lime tree, or a grapefruit tree. In one embodiment, the citrus plant is a navel orange, Valencia orange, sweet orange, mandarin orange, or sour orange. In one embodiment, the citrus plant is a lemon tree. In one embodiment, the citrus plant is a lime tree. In some embodiments, the plant is a relative of a citrus plant, such as orange jasmine, limeberry, and trifoliate orange.

(56) In some embodiments, the present invention provides for plants (or a plant cell, seed, flower, leaf, fruit, or other plant part from such plants or processed food or food ingredient from such plants) comprising an expression cassette comprising a promoter operably linked to a polynucleotide encoding a small RNA of the invention (e.g., as described herein). In some embodiments, the plant has decreased UBC expression or activity and/or increased expression or activity of Pi transporters.

(57) Kits

(58) In some embodiments, the disclosure provides kits that are useful for detecting the expression of small RNAs in plants. For example, the kit can include reagents that detect the presence of one or more small RNAs in a sample from a plant. In some embodiments, each reagent detects a different small RNA. In some embodiments, the reagent comprises a ligand that is capable of specifically binding to or hybridizing with a small RNA described herein. In some embodiments, the reagent is a nucleic acid that is labeled with a probe or other moiety that enables detection of the reagent. In some embodiments, the reagents include the nucleotide sequences in Tables 1-5 above. In some embodiments, the reagents are capable of detecting one, two, three, four, five or more of the small RNAs described herein.

(59) In some embodiments, the kits include primers that are useful for amplifying the small RNAs detected by the kits. For example, the kits can include the following primer sequences:

(60) TABLE-US-00006 miR399b (SEQ ID NO: 2): TGCCAAAGGAGAGTTGCCCTA miR399b RT primer (SEQ ID NO: 85): GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACATGGGC (50 nt) PCR primer: miR399b RT-F (SEQ ID NO: 86): GCGGCGGTGCCAAAGGAGAGTT miR399b RT-R (SEQ ID NO: 87): GTGCAGGGTCCGAGGT csi-siR1005 (SEQ ID NO: 13): ATAGATAATGGATCAACGGTTATA csi-siR1005 RT primer (SEQ ID NO: 88): GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACGTTATA (50 nt) PCR primer: csi-siR1005 RT-F (SEQ ID NO: 89): GCGGCGGATAGATAATGGATCAACG csi-siR1005 RT-R (SEQ ID NO: 90): GTGCAGGGTCCGAGGT csi-siR1008 (SEQ ID NO: 14): TCGAACAAGGTAAGGATGTCA csi-siR1008 RT primer (SEQ ID NO: 91): GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACATGTCA (50 nt) PCR primer: csi-siR1008 RT-F (SEQ ID NO: 92): GCGGCGGTCGAACAAGGTAAGG csi-siR1008 RT-R (SEQ ID NO: 93): GTGCAGGGTCCGAGGT csi-siR1009 (SEQ ID NO: 15): CTTCTAATAAACATGCATGAA csi-siR1009 RT primer (SEQ ID NO: 94): GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACCATGAA (50 nt) PCR primer: csi-siR1009 RT-F (SEQ ID NO:95): GCGGCGGCTTCTAATAAACATG csi-siR1009 RT-R (SEQ ID NO: 96): GTGCAGGGTCCGAGGT.

EXAMPLES

Example 1

(61) This example describes methods for detecting and treating citrus plants infected with Ca. L. asiaticus (“Las”).

(62) Materials and Methods

(63) Plant Material Maintains, Inoculation, and RNA Extraction

(64) For examining Las induced small RNA, two-year-old greenhouse-grown ‘Navel’ orange (C. sinensis) scions on Cleopatra mandarin (C. reticulata) rootstocks were inoculated by grafting a combination of tree bark pieces or leaf pieces onto the rootstock portion of each plant. Five plants were inoculated with bacterium-free tissue pieces and 19 plants were inoculated with infected tissue for 10 and 14 weeks post inoculation (wpi), respectively. For infected plants, bark and leaf pieces were obtained from infected greenhouse-grown Valencia scions. For non-infected plants, bark and leaf pieces were obtained from a healthy greenhouse-grown Valencia orange, confirmed negative for Las by PCR. For examining Spiroplasma citri induced small RNA, Spring Navel/Carrizo was grafted to Madam Vinous (C. sinensis) receptor plants, which are positive for S. citri by culturing and PCR using primers to spiralin gene sequences. Before inoculation plants were fertilized using a 17N-6P-10K controlled release fertilizer (Scotts Sierra, Marysville, Ohio, USA). Plants were arranged randomly on the greenhouse bench and kept under natural light conditions at a temperature of 17-25° C.

(65) Small RNA Library Construction

(66) Total RNA was resolved by a denaturing 14% polyacrylamide gel. 18-28 nt RNA fragments were recovered. The purified fragments were 5′ adapted (GUU CAG AGU UCU ACA GUC CGA CGA UCA G; SEQ ID NO:97), gel purified, and 3′ adapted (UC GUA UGC CGU CUU CUG CUU G; SEQ ID NO:98). The RNA fragments adapted at both ends were gel purified and reverse transcribed (SuperScript II, Invitrogen). After PCR amplification for 15 cycles (98° C. for 30 sec; 98° C. for 10 sec; 60° C. for 30 sec), the PCR products were gel purified and sequenced according to Solexa small RNA sequencing protocol.

(67) Mineral Measurement

(68) For mineral analysis, six to eight leaves were collected from each plant for analysis. Leaves from two non-infected plants of the same age and the same scion/rootstock combination were collected for comparison. Leaf tissue was washed for 15 sec each in RO-D (reverse osmosis-distilled) water, 0.01% detergent (Citranox, Alconox, Inc., White Plains, N.Y.) and 0.1 N HCl solution, followed by three more rinses in RO-D water, dried at 80° C. for at least 24 h in a forced-air oven, dry weight recorded, and leaf tissue milled to pass a 20-mesh screen. Leaf tissue (500 mg) was digested in 10 ml of concentrated HNO3 (trace metal grade) at 300 psi and 170° C. for 10 min in a microwave (model Mars 5, CEM Corp., Mathews, N.C.). Leaf digestates were brought to volume in 100 ml volumetric flasks and filtered (no. 41; Whatman Paper, Maidstone, Kent, U.K.). Foliar levels of phosphorous (P), potassium (K), iron (Fe), manganese (Mn), copper (Cu), zinc (Zn), sodium (Na), calcium (Ca), and magnesium (Mg) were determined by inductively couple plasma (ICP [IRIS 1000 HR Duo, ThermoElemental, Franklin, Mass.]) and nitrogen (N) and carbon (C) by dynamic-flash combustion and GC separation (CNS analyzer [NC 2100, CE Elantech, Inc., Lakewood, N.J.]).

(69) Application of Inorganic Phosphorus Solution

(70) The foliar nutrient sprays started with the spring flush in 2008. Applications have been made three times each year and timed with the initiation of the new vegetative flushes in the spring (March), summer (June), and autumn, (September). The foliar spray treatment has been a 3-18-20 analysis liquid fertilizer of potassium poly phosphate with 1.0 pint/gal. mono- and dipotassium salts of phosphorus acid 56% (K-Phite, Plant Food Systems, Inc., Zellwood, Fla. 32798), plus 8.4 lbs. spray grade potassium nitrate (KNO.sub.3), and 5.0 gallons 435 citrus spray oil in 250 gal./acre rate applied by hand gun to the foliage until runoff. Untreated control trees received no foliar applied nutrients. All trees received two ground-applied applications of a controlled-release fertilizer to the soil each year of a 14-0-24 analysis. The nutrients are from sulfur coated urea, sulfur coated sulfate of potash, ammonium nitrate, sulphate of potash magnesia, and muriate of potash. The fertilizer contained 2.3% Mg, 005% B, and 7.57% S. Insect management was determined by scouting and appropriate insecticides were applied by tractor pulled speed sprayer when required. Copper was used as a fungicide for disease control. Irrigation was by micro sprayers under each tree. Systemic and pre-emerge herbicides were used for weed management. Other standard good grove management practices known to favor good production were used.

(71) PCR Detection of Las

(72) The presence of Las was tested monthly and symptom-development was recorded. Petioles were ground in liquid nitrogen with a mortar and pestle. One hundred mg of ground tissue was used for DNA extraction according to the manufacturer's instructions (Qiagen, Valencia, Calif., USA). PCR amplifications were performed as described by Albrecht and Bowman (2008).

(73) RNA Extraction

(74) Total RNA was extracted from fully expanded leaves according to Strommer et al. (1993) with slight modifications. Phenol/chloroform/isoamylalcohol (25:24:1) extraction was followed by one extraction with chloroform/isoamylalcohol and precipitation of RNA with isopropanol at −20° C. overnight. RNA was pelleted by centrifugation at 10,000 g and 4° C. for 1 h. RNA pellets were stored at −80° C. and shipped to UC Riverside for construction of small RNA libraries. Leaf petioles were used for PCR detection of the bacterium.

(75) Small RNA Parsing

(76) After Illumina sequencing, the generated sequences (in FASTA format) were parsed and trimmed in total. An R program was used to identify the small RNA/adaptor boundaries, trim off the adaptors with variant length, and sort the sequences to respective source libraries. After removing the sequences shorter than 18 nt or longer than 28 nt, the remaining unique small RNAs were filtered by BLAST against a citrus non-coding RNA database, which included rRNA, tRNA, snoRNA and snRNA. The citrus non-coding RNA database was generated by combining Rfam database (Release 9.1, http://www.sanger.ac.uk/Software/Rfam/) with NCBI citrus EST sequences (http://www.ncbi.nlm.nih.gov/). The resulted small RNA sequences were interpreted as citrus small RNA and were studied in this research.

(77) Small RNA Cluster Analysis and Annotation

(78) Citrus EST sequences from NCBI (http://www.ncbi.nlm.nih.gov/) and non-redundant short EST sequences (<150 bps in length) parsed from the trace files (provided by Dr. Tim Close; UC Riverside) were assembled using CAP3 to generate a citrus EST Uniset database. The parsed small RNA sequences were then aligned to citrus EST Uniset database with BOWTIE (version 0.10.0, http://bowtie-bio.sourceforge.net/index.shtml), allowing up to 2 mismatches. Such generated sequences were mapped, and the reads for each source library was counted. To compare the expression level between different libraries, small RNA counts were normalized to the original library size. Cluster analysis on small RNAs matching to citrus EST Uniset was done by R program. A small RNA cluster was defined by aligning to the same locus with at least 16 nt overlapping. The clusters featured with forward (F) and reverse/complementary (R/C) alignments were counted in the 4 libraries, respectively. The functional annotation for citrus EST Uniset genes was from NCBI non-redundant protein database, Arabidopsis protein database (TAIR8_pep_20080412) and rice protein database (Version 5.0). Only the Uniset genes aligned with more than 10 small RNA copies were listed in the final small RNA expression table.

(79) Conserved and Novel miRNA Analysis

(80) For identifying conserved microRNAs, candidate small RNAs were analyzed by BLAST against microRNA Registry and Plant MicroRNA database (PMRD; http://bioinformatics.cau.edu.cn/PMRD/), respectively. For predicting novel miRNAs, each EST sequences matched with small RNAs were passed through an R script that detects a 450 bp pile-up matching region, and retrieve the flanking sequences by 200 bp from each side of the region. All retrieved EST fragments were folded with RNAfold (version 1.6.1). Structures of the EST sequences with minimum free energies were further analyzed by using an R and Perl script to retrieve stem-loops from the second structures and to check whether these stem-loops satisfy the following criteria: (1) the length of stem is longer than 20 bp; (2) no more than 4 bugles in a stem; (3) no more than 3 bp mismatches in a bugle; (4) putative microRNA locates on one strand of the stem, while miRNA* on the complementary strand; (5) no small RNA read matches to the loop region. Such predicted miRNAs were subjected to northern blot validation.

(81) Conserved and Novel Citrus MicroRNA Target Prediction

(82) MicroRNA targets were computationally predicted from the citrus EST Uniset using TargetFinder program (Release 1.5, http://jcclab.science.oregonstate.edu/node/view/56334) with default parameter settings. Briefly, potential targets from FASTA searches (+15/−10 match/mismatch scoring ratio, −16 gap penalty, and a RNA scoring matrix) were scored using a position-dependent, mispair penalty system. Penalties were assessed for mismatches, bulges, and gaps (+1 per position) and G:U pairs (+0.5 per position). Penalties were doubled if the mismatch, bulge, gap, or G:U pair occurred at positions 2 to 13 relative to the 5′ end of the microRNA. Only one single-nt bulge or single-nt gap was allowed. Based on a reference set of validated microRNA targets, only predicted targets with scores of four or less were considered reasonable. The functional annotation for targets of conserved and novel microRNA was done from NCBI non-redundant protein database.

(83) Northern Hybridization

(84) Fifteen to 100 micrograms of total RNA were loaded per lane, depending on signal strength. The RNAs were resolved on a denaturing 14% polyacrylamide gel and electro-blotted onto Hybond N.sup.+ membranes (Amersham) overnight in a cold room at a constant 150 milliampere (Bio-Rad). Membranes were cross-linked with 0.15M 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride (EDC) at 60° C. for 2 hours and dried at 80° C. for 2 hours. Citrus miRNA probes were synthesized as their reverse complementary sequences to the mature miRNAs and were labeled at the 5′ end with γ-.sup.32P-ATP using T4 polynucleotide kinase (Biolabs). Membranes were pre-hybridized using Perfect Hyb Plus buffer (Sigma) for 2 hours and hybridized with probes overnight at 37° C. Membranes were exposed overnight either to storage phosphor screens (GE health) or X-ray films after being washed four times (20 min each).

(85) Results

(86) Small RNA Profiling of HLB-Positive and Healthy Sweet Orange

(87) To study citrus small RNAs in responding to Las infection, C. sinensis plants were infected by Las by graft. Samples were collected at an early time point (10 wpi) and a late time point (14 wpi), respectively. Small RNAs ranging from 18 to 28 nucleotides (nt) from both healthy and Las-infected (treated) plants were cloned and sequenced. About half of the sequences can be aligned to the currently available citrus genome that was assembled from different databases. The identified citrus small RNAs were grouped into miRNAs and siRNAs by whether the sequence can be matched to conserved miRNAs (=<2 mismatches) in other plant species (top 20 plant species in miRBase; Table 6), and/or whether their precursor RNAs can form stem-loop structures. Our data showed that about 30% of the deep-sequencing reads belong to miRNA, while another 70% belong to siRNA due to the lack of miRNA characteristics. Since these citrus siRNAs didn't show any sequence similarities to siRNAs identified in other plant species or other organisms, we interpret them as C. sinensis specific siRNAs and named them as csi-siRNAs. Our deep-sequencing results showed that, similar to the model plant such as Arabidopsis, citrus small RNAs are featured by 21 nt and 24 nt species, as well as relatively low-abundant 22 and 23-nt populations. The majority of the identified citrus miRNAs start with a uracil (U), while most of the siRNAs start with both U and A (adenine).

(88) TABLE-US-00007 TABLE 6 conserved citrus miRNAs and abundance in libraries. miRNA 10 wpi 14 wpi families Untreated HLB Untreated HLB Total miR157 31896 34020 46477 58511 170904 miR166 2260 2345 69580 58393 132578 miR167 8546 1916 2621 7820 20903 miR164 4409 1775 2271 1227 9682 miR172 2696 1440 2366 1199 7701 miR165 123 108 4620 2790 7641 miR168 1160 1827 1393 2324 6704 miR159 109 749 321 1326 2505 miR396 707 113 1173 139 2132 miR845 855 234 311 209 1609 miR169 220 84 343 424 1071 miR403 612 104 104 42 862 miR827 223 124 189 156 692 miR160 159 66 406 41 672 miR156 86 43 236 283 648 miR170 215 129 97 195 636 miR162 60 49 161 147 417 miR393 16 66 22 271 375 miR171 191 22 39 15 267 miR394 23 8 101 19 151 miR399 4 26 6 33 69 miR398 10 4 23 5 42 miR408 14 1 7 3 25 miR390 7 4 3 4 18 miR395 2 6 0 5 13 miR397 3 0 3 4 10

(89) Citrus miRNAs were searched against database containing miRNA sequences from Arabidopsis thaliana, Brassica rapa, Populus trichocarpa (western balsam poplar), Gossypium hirsutum (upland cotton), Brassica napus, Glycine max, Vitis vinifera (wine grape), Solanum lycopersicum, Medicago truncatula, Oryza sativa, Zea mays (maize), Triticum aestivum (bread wheat), Sorghum bicolor, Saccharum officinarum (noble cane), Hordeum vulgare subsp. vulgare, Pinus taeda, Ricinus communis, Selaginella moellendorffii, Physcomitrella patens, and Chlamydomonas reinhardtii. miRNAs with more than 10 reads were deemed as valid conserved miRNAs and reported in Table 6.

(90) Research has shown that some Arabidopsis siRNAs that are specifically elicited upon bacterial pathogen infection participate in host innate immunity by regulating the expression level of their targets (Katiyar-Agarwal et al., 2006; Katiyar-Agarwal et al., 2007; Zhang et al., 2011). We investigated whether csi-siRNAs expression is altered upon Las-infection. We compared csi-siRNAs expression levels in both healthy and Las-infected samples and found that, similar in Arabidopsis, some csi-siRNAs also showed varied expression levels by Las infection (Table 5). Csi-siRNAs identified in this research were sorted according to their expression levels between healthy and treated samples and 10 representatives (csi-siR1001-10) were selected for experimental validation. Csi-siR1005 was induced upon Las infection both at 10- and 14-wpi. Csi-siR1005 can target a citrus protein without known function. Similarly, csi-siR1008 and csi-siR1009 also showed weak but noticeable induction upon Las infection. Csi-siR1008 can target a citrus protein homolog to an Arabidopsis putative disease resistance protein (CC-NBS class) with perfect match, whereas csi-siR1009 targets a putative disease resistance protein with 2 mis-matches. It suggests that csi-siR1008 and csi-siR1009 may play potential roles in citrus innate immunity by regulating the expression of disease resistance proteins. These inductions are specific to Las infection since when samples were treated with S. citri that causes citrus stubborn disease, no significant induction was observed.

(91) We focused our study on citrus miRNAs, which have been more extensively studied compared to siRNAs. A potential advantage of focusing on miRNAs is that currently available information in other systems may facilitate the study in citrus. Our analysis showed that most of the citrus miRNAs (conserved plus novel miRNAs) are 21-nt long, and dominantly favored a U at the 5′ end. Among the citrus miRNAs identified, about 240 miRNAs belonging to at least 26 miRNA families (reads>=10) could also be found in other plant species including Arabidopsis thaliana (Table 6), indicating that regulatory mechanisms in other plant species may also apply to the citrus system. Beyond these conserved miRNAs, 13 identified citrus miRNAs could not be matched to any currently known plant miRNAs, and therefore were termed as novel miRNAs. Similarly, they were named as csi-miRNAs (Table 7).

(92) TABLE-US-00008 TABLE 7 Novel citrus miRNAs. microRNA microRNA Candidate sequence star untreated HLB 10 untreated HLB 14 ID (SEQ ID NO:) (SEQ ID NO:) 10 wpi wpi 14 wpi wpi Csi- UGAAGCUGCCAGC AUCAUCUGGCAGU 7713.00 9946.31 625.47 776.91 miR5001 AUGAUCU (16) UUCACC (17) Csi- UAGAUAAAGAUGA UUUUCUCUUAUCG 604.00 104.13 675.33 235.24 miR5002 GAGAAAAA (18) UUAUCUGU (19) Csi- UUUGUUGCAUGAU CUACCCGCAUCAU 99.00 212.42 227.75 53.65 miR5003 GCUGAUAA (20) GCAACAAA (21) Csi- AGUGUUAGGUGUA UCUCGUACUUCUC 125.00 122.87 132.57 99.05 miR5004 GAGAAGCACGA UUCACCAAGCA (22) (23) Csi- AUUCGGGACGAGU CGUAAACUCGUCU 83.00 97.88 100.85 50.56 miR5005 UUACAAC (24) CGUACUU (25) Csi- AAUGGCUGGAUCC ACAAUUGGAUUUA 51.00 42.69 48.72 25.79 miR5006 AGCUGUGG (26) GCCAUUAA (27) Csi- AUGCCGUAUCACG CCCACUUGAUGUG 44.00 49.98 35.13 27.86 miR5007 UGGGAG (28) UCAUUC (29) Csi- UUUGAUGCCUUCU GAUUAAGGAGAGU 5.00 10.41 18.13 10.32 miR5008 UUAGUCGC (30) UUUCAGU (31) Csi- GACAGAAGAGAGT GCUCGCUCCUCUUC 0.00 5.21 0.00 31.98 miR5009 GAGCAC (32) UGUCAG (33) Csi- GUAUAUAUCUUGC GUAUGUAAGAUAC 12.00 9.37 5.67 2.06 miR5010 AUGCAUG (34) AUCCCC (35) Csi- UUUCUCUUAUCGU UAGAUAAAGAUGA 17.00 1.04 3.40 3.10 miR5011 UAUCUGU (36) GAGAAAAA (37) Csi- UUGUUGUUGAGUG ACAUAAAUACUUA 8.00 8.33 2.27 2.06 miR5012 UGUAUGUUA (38) AUAAUAAUC (39) Csi- UCGUCCUUCUCUC AGAGUGGGUGGGU 1.00 4.17 6.80 1.03 miR5013 AUAUUUUU (40) GGAGAGG (41) *Reads are normalized to library sizes. Csi = citrus sinensis. wpi = weeks post infection.

(93) The computationally predicted novel csi-miRNAs were experimentally validated using radioactively labeled probes. The Northern hybridization results showed a faithful agreement with the deep-sequencing results. For example, csi-miRNA5001, 5002, 5003, 5004, 5005, 5006, and 5007, which showed moderate or high reads in our deep-sequencing analysis, could be easily detected by their corresponding probes in Northern blots. In contrast, other csi-miRNAs that were predicted as low abundant appeared to be too weak to be detected (data not shown). This was confirmed by increasing the detecting limit by employing a locked nucleic acid (LNA) probe: when probed by a LNA probe, csi-miRNA5009 was clearly detected. However, there is no noticeable variation between healthy and treated samples, or between 10 wpi and 14 wpi time points, indicating that under the examined conditions, these csi-miRNAs may not be subject to expression alteration.

(94) Some Citrus miRNAs and siRNAs are Differentially Expressed in Healthy and HLB-Positive Plants

(95) Computational analysis of our data revealed that some of the conserved citrus miRNAs showed elevated expression upon Las infection, while others showed reduced expression. For example, at 10 and 14 wpi, miRNAs such as miR159, 399, 393 are noticeably induced (>2 fold), while miRNAs such as miR160, 396, 394, 398, 171, 403, and 408 are clearly reduced (>2 fold), as shown in FIGS. 1a and 1b. This implies that these csi-miRNAs may be subject to the influence of Las infection, or even play some potential roles in Las-elicited plant immunity. We tested some of the miRNAs by using Northern hybridization. Our results showed that all these alterations in miRNA expression upon Las infection were successfully validated using sequence-specific probes against the family member with the highest reads (data not shown). This induction is specifically due to Las infection since S. citri infection didn't show noticeable alteration on miR399 and miR159 expression level.

(96) Induction of miR399 Revealed Phosphorus Deficiency in HLB-Positive Plants

(97) In spite of its relatively low abundance, miR399 showed distinguishable induction upon Las challenge (FIG. 1). MiR399 has been shown to be involved in phosphate (Pi) homeostasis in Arabidopsis (Fujii et al., 2005; Lin et al., 2008; Hsieh et al., 2009). Some studies have shown that miR399 is the phloem-mobile long-distance signal involved in phosphate starvation response (Buhtz et al., 2008; Pant et al., 2008; Buhtz et al., 2010). In Arabidopsis thaliana, miR399 has multiple target sites in the 5′UTR of the transcript of a gene encoding a putative ubiquitin-conjugating enzyme (PHO2; At2g33770), which in turn negatively regulates Pi transporters (PT) (Fujii et al., 2005; Bari et al., 2006; Lin et al., 2008). Upon Pi deficiency, the miR399-mediated Pi homeostasis mechanism was turned on: the Pi-deficiency-induced miR399 down-regulates PHO2, which releases its inhibitory role on PTs; the increased PTs transport more Pi into the cells as a consequence, which alleviates Pi deficiency (Fujii et al., 2005; Lin et al., 2008). Research showed that the same Pi regulatory mechanism might also exist in other plant species such as rapeseed (Buhtz et al., 2008) and pumpkin (Pant et al., 2008). Interestingly, Bari et al. (2006) identified potential orthologs of PHO2 from orange (Citrus aurantium), implying that the regulatory mechanism maybe also conserved in citrus plants.

(98) The variation on citrus miR399 level between healthy and Las-treated citrus samples, and the potential conservation of the Pi regulatory mechanism prompted us to investigate the Pi level in these samples. We hypothesize that if the same Pi-regulatory mechanism also applies in C. sinensis, then the Las-infected C. sinensis would show deficiency in Pi level and elevated miR399 expression level; and we would also observe reduced PHO2 (citrus homolog) mRNA and increase PTs.

(99) We collected leaves from both healthy and Las-infected plants (PCR-confirmed). Our measurement showed that the phosphorus level in leaves of treated plants was more than 35% lower than in leaves from untreated plants. We also observed iron (Fe) deficiency in Las-infected plants, which indicates the phloem localized bacterial pathogen confers similar negative influence on several mineral elements. However, homeostasis of entire mineral elements is not negatively affected since some elements showed no noticeable variation, such as potassium (K) and zinc (Zn) (FIG. 4a). In contrast, copper (Cu) accumulated more in Las-infected plants (data not shown).

(100) We further tested our hypothesis by detecting PHO2 and PTs transcripts in both healthy and Las-infected plants by real-time PCR. After database searching, we identified one citrus PHO2 and two citrus PTs: csi.2677 (UGID: 1423690; homolog to Arabidopsis PHO2 [At2g33770; identity=69.9%]), csi.14938 (UGID: 3374895; homolog to AtPT2 [Arabidopsis thaliana phosphate transporter 2; AT2G38940; identity=85.2%]), and csi.9842 (UGID: 2916608; homolog to PHT2;1 [phosphate transporter 2;1; AT3G26570; identity=88.3%]). These citrus homologs therefore are referred as csi-PHO2, csi-PT2, and csi-PHT2;1 hereafter. We examined gene expression levels of PHO2 and PTs by real-time PCR. As shown in FIG. 4, we observed reduced csi-PHO2 expression level at both 10- and 14-wpi, agreeing with the increased miR399 expression at both time points, as well as the reduced Pi level; we also observed elevated expression of csi-PT2 and csi-PHT2;1, which are opposite to csi-PHO2 and in agreement with csi-PHO2's repressive role on csi-PTs. Taken together, our hypothesis was validated by the observation in concord with all the predictions: Pi-deficiency and induced miR399 in Las-infected plants, opposite expression profiles of csi-PHO2 (reduced) and csi-PT2 and -PHT2;1 (induced) at both time points of the Las-infected plants. Therefore, we propose that the miR399-PHO2 regulatory machinery is a conserved Pi homeostasis regulatory mechanism, at least between Arabidopsis and citrus.

(101) Application of Inorganic Phosphorus Solution Largely Reduces HLB Symptoms

(102) If Las infection leads to host Pi deficiency, which consequently caused observed symptoms such as yellowing, blotchy mottle, zinc-deficiency-like mottle, severe chlorosis, and most importantly, reduced fruit yield, then applying Pi to the infected plants should alleviate at least some of the HLB symptoms. We foliarly applied potassium poly phosphate (mono- and di-potassium salts of phosphorus acid 56%, plus potassium nitrate [KNO.sub.3], and citrus spray oil) to Las-infected plants for two years. As controls, we also applied KNO.sub.3 and citrus spray oil to infected plants, as well as untreated control trees that received no foliarly applied nutrients. After applying potassium poly phosphate for two years, the Las-infected plants showed dramatically reduced disease symptoms. Compared to untreated control trees and trees treated with nutrients only, the phosphate-treated trees have greener appearance, vigorous growth, and increased fruit yield. On the leaves from trees applied potassium poly phosphate, there is hardly any yellowing or blotchy mottle visible. In contrast, these symptoms are quite obvious on leaves from untreated trees or trees treated with nutrients only. The observed reduced symptoms are not due to application of potassium since the control plants also received KNO.sub.3. Therefore, we conclude that applying Pi, but not potassium nutrients, to Las-infected citrus can relieve HLB symptoms. Our results indicate that Pi application may be employed for HLB management and restore fruit yield in infected regions.

(103) Discussion

(104) Citrus is an important economic plant globally, which is currently threatened by the very destructive citrus disease-HLB (Bove, 2006). One of the current challenges for HLB management is an effective HLB disease control reagent that can replace the costive eradiation procedure, which is the only option currently available. Plants have evolved multiple levels of immune responses, including basal defense triggered by virulent pathogens in susceptible hosts and resistance (R) gene-mediated resistance activated by avirulent pathogens in resistant hosts (Chisholm et al., 2006; Jones & Dangl, 2006). Infection of bacterium Ca. L. in different genotypes of citrus plants causes different degrees of disease and symptoms. Although there is no known complete resistance in Citrus spp., tolerant citrus cultivars that have a very mild or no obvious disease symptom and with low bacterial titer were identified (Folimonova et al., 2009). Thus citrus hosts are capable of recognizing Ca. L. infection and responding to the pathogen in different degrees. The significance of this research is investigating the involvement of the citrus innate immune system in Ca. L. infection. Studying and understanding how citrus hosts use innate immunity as tools against Ca. L. should lead to utilizing of these innate immune tools as efficient biological reagents against Ca. L. in future. This is going to have a huge impact on the citrus industry considering HLB management-related cost is estimated about 40% higher than pre-HLB costs in the United states (Irey et al., 2008) and about 12.65-38.73% of the total operational costs in Sao Paulo, Brazil (Belasque et al., 2010).

(105) Previous studies have shown that small RNAs are involved in basal defense and R gene-mediated resistance (Katiyar-Agarwal et al., 2006; Navarro et al., 2006; Katiyar-Agarwal et al., 2007; Zhang et al., 2011), as well as plant fitness adjusting mineral homeostasis (Jones-Rhoades & Bartel, 2004; Fujii et al., 2005; Burkhead et al., 2009). Our investigation of citrus small RNA populations revealed that citrus possess many conserved miRNAs, which can also be found in other plants, such as Arabidopsis. In most of the cases, bioinformatics analysis data agreed with our experimental very well. For example, there is a very good match between the bioinformatics prediction and experimental validation of the expression level of citrus conserved miRNAs; most of the miRNAs that were predicted more than 2-fold increase or decrease in both 10- and 14-wpi (FIG. 1) were successfully experimentally validated. This indicates that our bioinformatics approaches are not only powerful, but also reliable in investigating the dynamic small RNA populations upon pathogen challenge. Our results showed that some citrus miRNAs, such as miR 160, 396, 398, and 399, demonstrated distinguished expression patterns between healthy and Las-infected plants, among which miR399 and miR396 are the ones with potential practical value due to their relative high abundance.

(106) Furthermore, the Las elicited miR399 induction and Pi deficiency suggests there might be connections between HLB and Pi level. Research has shown that in Arabidopsis the miR399/PHO2 regulatory machinery is turned on upon Pi starvation, and over-expression of miR399 induces accumulation of Pi (Fujii et al., 2005; Lin et al., 2008; Buhtz et al., 2010). Using public genomic DNA and expressed sequence tag data, Bari et al (2006) assembled potential orthologs of PHO2 from rice, Medicago truncatula, poplar, wheat (Triticum aestivum), soybean (Glycine max), cotton (Gossypium hirsutum), apple (Malus domestica), and orange (Citrus aurantium), but not in the Physcomitrella patens EST database or the genome sequences of P. patens and C. reinhardtii. These results suggest that the regulatory mechanism may be conserved across angiosperms and that it may have emerged during the evolution of higher plants.

(107) If the citrus and Arabidopsis miR399 are really functionally conserved, as suggested by our data and observations by Bari et al (2006), this can be interpreted as that Las infection reduces host Pi level, which in turn triggered miR399 induction, as observed in this study. If this is true, then citrus miR399 should be induced upon Las infection, should participate in host Pi up-regulation pathways (by regulating PHO2 directly, and csi-PT2 and -PHT2;1 indirectly), and applying Pi should alleviate HLB symptoms.

(108) Several lines of evidence support our hypothesis. First, as we demonstrated in this study, when citrus was affected by Las, Pi deficiency was observed (FIG. 4a), which is about >30% lower than in the healthy leaves. Some mineral nutrients, such as zinc and potassium, did not show noticeable variations. Second, citrus miR399 was specifically induced upon Las (FIG. 1), but not S. citri challenge. Although the overall expression level is relatively low in citrus, elevation in expression level could be detected unambiguously and consistently. Third, we proved that citrus miR399 is functionally conserved with its Arabidopsis homolog, which regulates Pi accumulation through down-regulating csi-PHO2 (FIG. 4). We also observed elevated expression of two Pi transporters-csi-PT2 and -PHT2;1, respectively. The cognate induction of citrus miR399, reduction on csi-PHO2 level, and induction of csi-PT2 and -PHT2;1 after Las infection suggests there is a functional miR399/PHO2/PTs pathway controlling Pi homeostasis, as being revealed in other species such as Arabidopsis, rapeseed, and pumpkin (Fujii et al., 2005; Buhtz et al., 2008; Pant et al., 2008). Fourth, field practice showed that when Pi is applied, HLB symptom is alleviated. This indicates that Las infection caused Pi deficiency contributes to observed HLB symptoms, including yellowing and reduced yields, and when normal Pi level is restored by applying Pi (or by endogenous mechanisms, such as the miR399 pathway), Las-infected citrus would show reduced symptom.

(109) Based on our results, we propose a model in which citrus miR399 may play a role in citrus response to Las infection. In this model, Las infection causes a Pi deficiency on the host. This is consistent with current observation that Las is restricted to phloem, and Las infection usually cause phloem congestion (Folimonova & Achor, 2010). Reduced Pi level induces miR399 expression, which in turn down-regulates its target, csi-PHO2, a citrus ubiquitin conjugating enzyme. Based on its homology to Arabidopsis PHO2, csi-PHO2 may be a suppressor that restricts citrus phosphate transporter activity at normal conditions. Upon Las infection, the miR399-mediated reduction of csi-PHO2 consequently leads to induction of phosphate transporters, which recharge host cells' Pi reservoir. This model is similar to the model proposed in Arabidopsis, where mR399 and PHO2 play important roles in response to Pi deficiency (Lin et al., 2008). Foliar applying Pi to the Las-infected plants, which reduced HLB symptoms and enhanced fruit yield, validated our model. Our model may explain the filed practice that applying nutrition to infected tree can maintain tree health and productivity. Studies showed that Ca. L. infection restricted either nutrient uptake or transport and that foliar applied minerals could prolong tree life and reduce yield losses (Pustika et al., 2008).

(110) Our model does not exclude the possibility that miR399 can be directly induced by Las infection. Research has shown that miR393 and miR393b* can be induced by bacterial pathogens, and their induction contributes to antibacterial resistance (Navarro et al., 2006; Zhang et al., 2011). We speculate that over-expression of miR399 in Las-infected citrus should reduce HLB symptoms. Whether normal or enhanced Pi level in citrus plants have an inhibitory effect on infection and/or propagation of Las in hosts remains to be investigated in future. It should be noted that besides phosphorus, we also observed reduced iron level after Las infection (FIG. 4a). It seems that after Las infection, the congested phloem also affects homeostasis of other mineral elements.

(111) This example demonstrates the Ca. L. asiaticus infection of citrus plants can be detected by detecting the expression of miRNAs and siRNAs, and that HLB disease symptoms can be ameliorated by treatment of plants with phosphorous.

(112) It is understood that the examples and embodiments described herein are for illustrative purposes only and that various modifications or changes in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this application and scope of the appended claims. All publications, sequence accession numbers, patents, and patent applications cited herein are hereby incorporated by reference in their entirety for all purposes.