Differential Expression of Arabidopsis thaliana Acid Phosphatases in Response to Abiotic Stresses

Document Type: Research Paper

Authors

Department of Plant Biotechnology, National Institute for Genetic Engineering and Biotechnology, P.O. Box 14155-6343, Tehran, IR Iran

Abstract

The objective of this research is to identify Arabidopsis thaliana genes encoding acid phosphatases induced by phosphate starvation. Multiple alignments of eukaryotic acid phosphatase amino acid sequences led to the classification of these proteins into four groups including purple acid phosphatases (PAPs). Specific primers were degenerated and designed based on conserved sequences of PAPs isolated from plants, fungi and animals. RNA profiles of Pi-fed and Pi-starved A. thaliana roots were compared with two methods established for gene-family-directed differential display of pap genes. Having analyzed the differentially displayed fragments, seven pap encoding cDNA clones were isolated. One of the clones was a trans-splicing product of two genes that encodes an acid phosphatase carrying a zinc-finger domain. Six other clones were predicted to encode secretory phosphatases. Reverse-northern blotting and semi-quantitative RT-PCR revealed distinct expression patterns for each gene under diverse environmental conditions such as Pi starvation, high-salt concentration, cold shock, nitrogen and sulfur deprivation. The presented data can provide some clues for dissecting the possible roles of PAPs in Pi acquisition.

Keywords


INRTODUCTION

Phosphorous is one of the most essential macronutrients that contributes to the structure of key biomolecules such as DNA, RNA, phospholipids and phosphoproteins as well as energy transfer components such as pyrophosphate, ATP, ADP or AMP (Marshner 1995). It also plays vital roles in metabolic reactions, gene expression and signaling pathways (Marshner 1995).
 In soil, high quantity of phosphorous is present in various forms of inorganic and organic compounds ranging from 400 to 1200 mg/kg (Rodriguez and Fraga 1999). Despite, plants only absorb soluble inorganic phosphate (Pi) which is the least readily available nutrient in the rhizosphere (Marshner 1995; Robinson et al., 1999). To deal with sub-optimal Pi concentrations in soil, a number of adaptive strategies have been evolved in plants. Recent genome-wide transcriptional analysis by the use of cDNA microarrays and oligonucleotide chips, have shown that about one third of the Arabidopsis genes exhibit altered expression patterns within the first three days of Pi starvation (Wu et al., 2003; Hommand et al., 2003; Misson et al., 2005). These include many genes encoding enzymes of nitrogen assimilation and carbon fixation indicating that cell rescue system is turned on when the Pi starvation is prolonged.
 Among these, the up-regulation of acid phosphatases (Apases) is believed to be the primary means for the releasing, recycling and scavenging of Pi from both internal and external resources (Goldstein et al., 1988a and b; Duff et al., 1994). Purple acid phosphatases (PAPs) are a group of APases that catalyze the hydrolysis of a wide range of phosphate esters and anhydrides in plants, fungi and animals (Shenck et al., 2000a; Li et al., 2002; Flanagan et al., 2006). Structurally, plant PAP proteins are categorized to high molecular weight (HMW) and low molecular weight (LMW) phosphatases all of which share metal-ligating conserved sequences (Shenck et al., 2000a). The former is thought to be functional in homodimeric form and carries two b-a-b-a-b motifs in its carboxyl end; while the later is typically monomeric with a sequence similar to carboxyl segment of HMW proteins but carrying only one of the structural motif (Klabunde et al., 1996; Zhang et al., 1996; Huang et al., 1997; Mertz et al., 1997; Uppenberg et al., 1999). Comparison of primary sequences of PAP proteins from both eukaryotic and prokaryotic organisms has revealed five blocks with seven invariant residues required for metal coordination (DXG/GDXXY/GNH(D/E)/VXXH/GHXH; bold letters represent metal ligating residues). Using metal-ligating conserved sequences as motifs, Li and colleagues (2002) reported the identification of 29 PAPs in A. thaliana complete genome sequence. Later, Zhu et al. (2005) compared the expression pattern of 28 PAP encoding genes in five organs.
 In this research, we have established a couple of methods for gene-family-directed differential display (DD) with the use of degenerate primers designed based on PAP proteins conserved sequences. Using these methods, we have reported the identification of PAP genes that are responsive to environmental conditions, particularly Pi starvation.


MATERIALS AND METHODS

Plant materials, culture conditions and treatments: The seeds of Arabidopsis thaliana ecotype Col-1 were obtained from Arabidopsis Biological Resource Center (ABRC, Ohio State University, Ohio). Seed sterilization and hydroponics growth conditions were as described by Malboobi and Lefebvre (1997). Prior to the treatments, 14-day old seedlings were transferred into 15 ml of half-strength MS medium (Murashig and Skooge, 1962) supplemented with 1% sucrose and incubated for two days to achieve identical nutritional states. For all investigated stresses other than Pi starvation, the initial concentration of KH2PO4 was kept at 5 mM. For Pi deprivation, 5 mM KCl replaced KH2PO4 in half-strength MS medium. For high-salt treatment, NaCl solution was added to a final concentration of 100 mM, a sub-lethal concentration for A. thaliana seedlings (Saleki et al., 1993). For minus nitrogen treatment, KNO3 and NH4NO3 were replaced by equal molarity of KCl. For medium lacking sulfur, equal molarity of MgCl2 was added in place of MgSO4. Cold treatment was applied by incubating 27 days plants at 4°C for 24 hr. For elicitor treatment, 28 days plants were incubated in the 20 ml fresh medium containing 100 μg/ml chitin for 4 hrs. In all cases, the culture medium was refreshed twice a week. All plants were harvested after a 14 day treatments; except for plants grown without nitrogen which were harvested after 9 days of treatment due to onset of severe deprivation symptoms. All tissue samples were stored at -70°C.

Database search and clustering of acid                phosphatases: Initially, a key word search was performed using the phrase “acid phosphatase” against Swiss-PROT and PIR databases. Second ,  PAP1 (GenBank Acc. No. ATU48448) protein sequence isolated from A. thaliana Var. Landsberg erecta (Patel K.S., Lockless S.W and McKnight T.D. unpublished data) was used for BLASTP search in the above databases using the default setting. Sequences with identities above 95% were deleted, a total of forty amino acid sequences of Apases isolated from plants, fungi and animals’ species were compilied by the end of year 1998. The retrieved Apase amino acid sequences were aligned using several softwares including ClustalW 1.8 (http://www.ebi.ac.uk/clustalw/) method with BLOSUM matrix. For clustering, the alignment outcomes were imported to PhyloDraw 0.8 software (http://pearl.cs.pusan.ac.kr/phylodraw/) and analyzed with the use of neighbor-joining algorithm.

Primer design: The protein alignment was used to identify the conserved motifs, that were selected to design degenerate primers  PAF1, PAF2 , PAF3 and specific primers P1 and P2 (Table 1). Specific primers for the identified PAP genes  were designed within the unique sequences of each one (Table 1). To sort out the unique parts, the predicted open reading frame nucleotide sequences were searched against Arabidopsis genome database and segments with over 75 percent similarities were deleted.

RNA extraction and differential display: Total RNA was isolated from treated and non-treated A. thaliana roots as described by Chirgwin and colleagues (1979). 20 mg of total RNA was reverse transcribed in a 20-ml reaction containing 50 units of Expand reverse transcriptase (Roche Biochemical, Mannheim, Germany), 40 units of RNase inhibitor (Roche Biochemical) and 40 pmole of either P2 or oligo(dT)15 primers which was incubated at 42°C for 2 hrs. Two optimized methods used for gene-family- directed differential display were as below:

1) Low-stringency DD: PCR amplification of cDNA molecules synthesized with oligo(dT)15 primer was performed using three combinations of degenerate and specific primers including PAF1and P2, PAF2 and P2 or PAF3 and P2 pairs. PCR reaction was composed of 0.1 ml of cDNA products, 0.25 mM of each dNTP, 2.5 mM MgCl2, 10 pmole of each primer, 1X Taq polymearase buffer and 1 unit Taq DNA polymerase (Cinnagen, Tehran, Iran) in a total volume of 20 ml. The first PCR program was 4 cycles of 1 min denaturation at 94°C, 1 min annealing at 45°C and 1 min extension at 72°C. This was followed by the second PCR program which was 1 min denaturation at 94°C, 1 min annealing at 50°C, 1 min extension at 72°C for 36 cycles and a final extension at 72°C for 5 mins.

2) Slow-ramping DD: two-step PCR amplification of cDNA molecules synthesized with P2 primer was performed with the above combinations of degenerate and specific primers. The initial PCR reaction was consisted of 0.2 ml cDNA products, 0.25 mM of each dNTP, 2.5 mM MgCl2, 20 pmole of each primer, 1X Taq polymearase buffer and 0.5 unit Taq DNA polymerase in a total volume of 20 ml. The first round of PCR reaction was carried out for 30 cycles with the following program: 1 min denaturation at 94°C, ramping from 94°C to 40°C at the rate of 8 sec/°C, 1 min annealing at 40°C and 1 min extension at 72°C. After addition of 10 μl solution containing 0.25 mM of each dNTP, 1X Taq polymearase buffer and 0.5 unit Taq DNA polymerase, the second round of PCR was performed for 20 cycles of 94°C for 1 min, 50°C for 1 min (without slow ramping) and 72°C for 1 min proceeded. The amplified DNA fragments were separated on 1% agarose gel and stained with ethidium bromide.

Cloning and sequence analysis: Differentially         displayed bands were cut and eluted by the use of Agarose gel DNA extraction Kit (Roche Biochemical). Purified cDNA fragments were cloned into pTZ57RT vector supplied in T/A cloning Kit (Fermentas, Vilnius, Lithuania) and recombinant plasmids were used to transform competent E. coli strain DH5a cells (Sambrook, Fritsch and Maniatis 1998). Specific features of the sequenced PAP clones were predicted by the use of online databases and relevant software including Pfam (Bateman et al., 2000), Prints (Attwood et al., 2000), Blocks (Pierrokovski et al., 1996), SMART (Schultz et al., 2000; Letunic et al., 2002), InterPro (http://www.ebi.ac.uk/InterProScan/) and PSORT  (Nakai and Horton, 1999).

Reverse-northern blotting: Reverse-northern blotting was performed following the procedure of Mohsenzadeh and colleagues (2005). Duplicate H+ nylon membranes (Roche Biochemical) were prepared by placing dots of reamplified, alkaline-denatured DNA inserts of the isolated clones. To synthesize labeled cDNA molecules, 10 μg heat-denatured total RNA isolated from Pi-fed and Pi-starved plants were added into cDNA synthesis reactions consisted of 80 units of Expand reverse transcriptase, 50 mM Tris-HCl, pH 8.5, 8 mM MgCl2, 150 mM DTT, 1.25 units/μl RNase inhibitor, 0.5 mM dNTP mixture, 0.13 mM dTTP, and 0.07 mM DIG-dUTP (Roche Biochemical) incubated at 42°C for 2 hrs. Prehybridization, hybridization, stringency washes, and detection steps were carried out according to the DIG Labeling and Detection Kit instructions (Roche Biochemical) using hybridization buffer composed of 7% SDS, 0.25 M NaH2PO4, 1 mM EDTA. The intensity of each dot was quantified by TotalLab software (Phoretix International, Tyne House, Newcastle, UK). All data were normalized by dividing the given volume value for each dot by that of alpha-tubulin. The relative expression level was calculated as the ratio of dot intensities hybridized with labeled cDNA molecules derived from Pi-starved plants by those of the Pi-fed ones.

Semi-quantitative RT-PCR: In order to evaluate the transcriptional responses of the isolated pap genes to Pi deprivation and other stress conditions, a series of semi-quantitative RT-PCR were conducted. For all samples, the same amounts of total RNA (10 μg) were used in reverse transcription reactions. Pairs of specific primers that anneal to the unique sequences of pap genes and alpha-tubulin were used (Table 1) in the PCR reactions. PCR reactions were performed in a solution of 50 mM KCl, 15 mM Tris-HCl (pH 8.4), 2 mM MgCl2, 0.25 mM dNTP mix, 2 pmole of each primer, 0.1-1 μl cDNA and 1unit Taq DNA polymerase (CinnaGen) on a thermocycler (Perkin-Elmer, GeneAmp 9600) at the appropriate annealing temperature (see Table 1). Prior to PCR amplification of Arabidopsis PAPs, the cDNA content of all reverse transcription reactions were normalized by amplification of alpha-tubulin transcript using primers TubF and TubR (Table 1). Ethidium bromide stained gel images were analyzed by TotalLab software (Phoretix International) to obtain numeric data representing the band intensities. All calculations were as mentioned above and performed on Microsoft Excel software.


RESULTS

Classification of acid phosphatases: In the first attempt, we tried to retrieve as many Apase protein sequences as available by the year 1998. At this stage, the main goal was to identify the conserved motifs in eukaryotic Apase sequences for subsequent primer design. Removing redundant sequences with more than 95% identities, 40 Apase sequences were compiled by databases searches, 6 from plants, 19 from animals and 15 from fungi. Multiple sequence alignments by Clustal W led to classification of Apases into four groups: group 1, mainly animal histidine phosphatases (15 sequences); group 2, fungi histidine phosphatases (9 sequences); group 3, high and low molecular weight PAPs (12 sequences) and group 4, class B acid phosphatases (3 sequences).
 The alignments of the PAP sequences (Figure 1, group 3) revealed at least four conserved motifs with some invariant amino acid residues (Figure 2). A part of PAF1 motif was used for degenerate primer design (Table 1) which is located at N-terminal sequence of plants and fungi PAPs. This motif is also a part of fibronectin type III domain commonly found in many plant and animal proteins as well as extracellular enzymes of soil bacteria (Tsyguelnaia and Doolittle, 1998). Three other degenerate primers corresponded to the short sequences of PAF2, PAF3 and PAF4 motifs (Table 1, Figure 2) known to contain metal- ligating residues (Schenk et al., 2000a). P1 and P2 specific primers were designed based on nucleotide sequences of PAP1 gene encoding GDLSYA and GHVHAY motifs that are highly conserved among plant PAPs (Figure 2).

Differential display of PAP family members:  In order to identify PAP genes preferentially expressed in Arabidopsis plants in response to long term harsh Pi stress, two methods for gene-family-directed DD were established to illustrate the expression profiles of a subpopulation of transcripts encoding PAP family members. In both methods, total RNA samples were prepared from both Pi-fed and Pi-starved roots after 14 days of treatments. For the first method, the starting material was cDNA molecules prepared by reverse transcription with oligo(dT)15 primer that were used in the PCR reactions conducted at low stringency. As shown in Figure 3A, only a few bands were amplified with this method. Four combinations of degenerate and specific primers yielded five differentially expressed cDNA fragments. For the second method, reverse transcription was carried out with a PAP-specific primer (P2) proceeded with slow ramping (8 sec/°C from 95°C to 40°C) using pairs of degenerate and gene-specific oligos. After addition of fresh Taq DNA polymerase, another round of PCR was performed at higher annealing temperature. With this method, an average of 8-12 bands per lane was displayed (Figure 3B). Six other preferentially expressed cDNA fragments in Pi -starved samples were identified with this method.

Sequence analysis of the identified PAPs: Differentially expressed fragments were cloned for further analysis. The partial cDNA sequences were searched against A. thaliana genomic sequence and expressed sequence tag (EST) databases. An inventory of relevant data for the isolated gene fragments is presented in Table 2. Only two fragments isolated by the first DD method were similar to the predicted pap genes whereas five fragments obtained through the second method were representing pap gene sequences. Those two fragments, L8 and M3, isolated by the first method correspond to pap9 and pap18 genes sequences, respectively. Four fragments, F1, G4, H5 and K6, isolated by the second method were almost identical to pap5-1, pap11, pap12 and pap26 genes, respectively, as named by Li and colleagues (2002). The full-length coding sequence of the identified pap genes predicted by AGI were double checked with EST and TC sequences available until year 2005 on TIGR database.
 Interestingly, E7 fragment was found to be identical to a segment of pap5-1 gene on chromosome 1 and a segment of a gene on chromosome 4 encoding an unknown protein carrying a zinc finger domain, designated as zf4. The full length trans-spliced product, called pap5-2, is composed of 13 exons carrying           a ~2120 bp open reading frame encoding an ~708 amino acid sequence. It is notable that there would be an early stop codon within the resulting mRNA of pap5-2 gene unless a frame shift close to trans-splicing site is presumed.
 The predicted protein sequence of the identified PAP genes carry all five blocks of conserved amino acid sequences encompassing seven metal-ligating residues as shown by Schenk and colleagues (2000). As well, fibronectin type III domain was found in the N-terminal parts of PAP5-1, PAP12, PAP18, and PAP26. Analysis with PSORT program collection indicated that all identified could be secretory proteins except for PAP9 and AP5-2. PAP9 sequence includes several hydrophobic regions that are common characteristics of membrane proteins. Search in Blocks and PROSITE databases showed the presence of C2H2 type zinc-finger domain and two nuclear targeting motifs within the N-terminal sequence of PAP5-2 both of which are indicative of being a nuclear protein.

Gene expression analysis: In order to illustrate and quantify the relative expression levels of the identified genes in various environmental conditions two approaches were taken:

1) Reverse-northern blot analysis: Reamplified cDNA fragments were arrayed on nylon membrane and hybridized to labelled cDNA molecules derived from total RNA extracts of Pi-fed and Pi-starved roots. It was shown that all seven cDNA clones were up-regulated considerably in response to long-term Pi starvation (Figure 4). After normalization of data against alpha-tubulin as constitutively expressed gene, relative levels of transcripts were calculated to be 2.5 to 4.6 times higher in Pi-starved roots in compare to Pi-fed ones (Table 3). It is noteworthy that it is not possible to differentiate the expression level of pap5-2 from pap5-1 gene and zf4 genes since both genes could be expressed.

2) Semi-quantitative RT-PCR: After reverse transcription, initial cDNA template level for each sample was adjusted with respect to alpha-tubulin expressed transcripts prior to semi-quantitative PCR amplifications. Primers were designed based on specific unique sequences of each gene to ensure distinctions among gene family members. In general, the relative expression levels measured by this technique were consistent with the results of reverse-northern blots (Table 3).
 Figure 5 illustrates comparisons among the expression patterns of the isolated pap genes in various environmental conditions. To calculate relative expressions in each condition, the band intensities were divided by that of the control plants which were grown in half-strength MS medium with sufficient amount of Pi (Table 3). Pap5-1 gene was highly induced by Pi deprivation condition while it was repressed by other abiotic and biotic stresses. The transcript levels of pap9 in all examined environmental conditions were higher than the control plants, except for the nitrogen starvation treatments. This was more pronounced in chitin-treated and Pi-starved plants. The expression of pap12 was only increased in response to Pi starvation and cold stress, though it was constitutively expressed in all the other environmental conditions at low levels. In comparison, the expression of pap18 gene was repressed in all environmental stresses except for the Pi-starved plants. The expression of pap26 gene was repressed in all environmental conditions other than no-Pi and high-salt treatments. Interestingly, pap5-2 transcripts were accumulated in these two conditions too. For this experiment, primers specific to 5¢-region of zf4 and 3¢-end of pap5-1 genes were used such that it is expected only trans-spliced products are amplified. We consistently obtained a 675-bp band when conducting RT-PCR with Pi-fed mRNA. However, Southern blotting of the RT-PCR products confirmed that this band is not related to pap5-2 while the others are (data not shown).


DISCUSSION

Altogether, 59 Apases with divergent protein sequences were identified in Arabidopsis plant that could be classified into at least five major groups (Lohrasebi T. and Malboobi M.A., unpublished data). While 29 PAP protein sequences aligned well, they are barely aligned with the members of other Apase families.
 The examination of gene expression patterns may provide some clues about the biological roles of PAP. Zhu and colleagues (2005) studied the expression profiles of A. thaliana pap genes in five organs using semi-quantitative RT-PCR method. Based on the clustered data, four groups of pap genes were recognized with respect to their expression in root, stem, leaf, flower and silique. Sixteen genes were constitutively expressed in all organs while others were expressed in one or more organs. Interestingly, all of the genes, except for pap5-1, were expressed in flowers. In another effort, Misson et al., (2005) reported differential expression of 11 out of 27 pap genes represented on ATH1 DNA chip in response to either short term or long term Pi starvation stress.
 Altogether, seven Pi-starvation induced cDNA fragments encoding PAPs, six secreted PAPs and one nuclear were isolated (Table 2). Both sequences of differentially-displayed DNA fragments and subsequent RT-PCR reactions with primers specific for the unique sequences verified the transcription of open reading frames predicted by Arabidopsis Genome Initiative (AGI). However, there was an error in recognition of the second exon-intron boundary of pap5-1. Other researchers have also reported such erroneous predictions in pap7 and pap8 coding sequences (Li et al., 2002). Despite using different sets of primers and RNA preparations, we were unable to amplify pap11 transcripts while Li et al (2002) have shown that the expression of this gene is highly induced after 3 days of Pi starvation treatment. Also, Zhu and colleagues (2005) did not detect the expression of pap5-1 in seven examined organs including roots of A. thaliana grown in pots, whereas both we (Figure 5) and Misson et al (2005) showed its expression in roots particularly when Pi was limited. These discrepancies could be due to growth conditions indicating complex patterns of Apase expression in distinct locations and in response to environmental conditions.
 Li et al (2002) used specific primers to amplify the coding region of individual PAPs. They found that the transcript levels of pap7, pap8, pap9, pap10 and pap13, were not affected after 1, 3 and 5 days growth in low phosphate medium while the expression of pap11 and pap12 were elevated. In our experience, the transcript levels of pap5, pap9, pap12, pap18 and pap26 were highly induced after 14 days of Pi starvation. The discrepancy in these results could be related to the duration of Pi-starvation treatments. Consistently, we found that the expression of pap12 increased proportionally after 6 hrs, 5, 8, 14 and 21 days of low phosphate treatment (data not shown).
 We have also shown that several environmental conditions such as no sulfur, no nitrogen, high salt and cold affected the expression of the pap genes differentially (Figure 5, Table 3). The induction of Apases of other plant species in response to high salt has been shown as well (delPozo et al., 1999; Mimura et al., 2003; Parida and Das 2004). The expression data stored in TAIR website (http://www.arabidopsis.org/; Yamada, et al., 2003) also include the effect of several environmental conditions (but not Pi starvation). Based on these data, osmotic stresses (e.g. high salt, ABA, drought and monitol treatments) induce the expression of pap5-1, pap12 and pap18. Exposure to UV-B increased the expression of pap5-1 in root while repressed it in the shoot. Responsiveness to cold was exceptionally high (seven times) for pap18 in one-hr treated plants grown on soil while no major effect was observed for long term-treated plants, nor for the other pap genes (Yamada et al., 2003).
 The emerging data in recent years have signified the involvement of acid phosphatases in response to biotic stresses too. Jakbek et al. (2002) identified an Apase gene that accumulates during the hypersensitive reaction in bean. Similarly, Petters et al. (2002) identified an Apase expressed locally in potato leaves in response to bacterial infiltration. This gene is highly homologous to a Pi starvation induced Apase in tomato. We have also found that pap9 expression was highly induced by chitin treatment as an elicitor (Figure 5).

Concluding Remarks

Why plants have so many Apases is still an interesting question that requires accumulative efforts for compilation of data describing the structural features, expression patterns, intermolecular relations, sub-cellular localizations and biochemical characteristics that lead to exploration of functions and roles of the Apases. Three independent Arabidopsis mutants defective in Apases activity have already been characterized. pho3 with reduced Apase activity phenotype is also defective in various Pi deficiency responses (Zakhleniuk et al., 2001). Reduced Apase activity on the root surface but not in whole cell extract of pup1 is indicative of restriction of Apase activites to certain locations (Trull and Deikman 1998). Tomscha et al. (2004) have introduced an Arabidopsis mutant, pup3, with 25 to 49 percent APase activity reduction mainly related to PAP26 and PAP12 isozymes. Based on the position of mutation locus, they suggested possible defect in pap26. As the transcript level of pap12 was not affected, they presumed posttranscriptional modifications in a ground of mutual relations among Apases could be the cause of observed phenotypes. The availability of T-DNA insertional mutant (Pan et al., 2003) and gene-specific silenced lines (Hilson et al., 2004) are other great resources for functional analysis of Apases. Overexpression of the isolated Apases genes, mutation phenotyping and complementation of the mutants would pave the way for understanding the distinct roles of Apases in Pi homeostasis in plants.

Acknowledgments

This research was funded by a grant from National Research Council of I.R. Iran. We would also like to acknowledge the support of National Institute for Genetic Engineering and Biotechnology of I.R. Iran.

Attwood TK, Croning MDR, Flowver DR, Lewis AP, Mabey JE, Scordis P, Selley JN, Wright W (2000). PRINTS-S: the database formerly known as PRINTS. Nucleic Acid Res. 28: 222-224.
Bateman A, Birney E, Durbin R, Eddy SR, Howe KL, Sonnhammer ELL (2000). The Pfam protein families database. Nucleic Acid Res. 28: 263-266.
Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ (1979). Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochem. 18: 5294-5299.
DelPozo JC, Allona I, Rubio V, Layva A, delaPena A, Aragoncillo C, Paz Ares J (1999). A type 5 acid phosphatase starvation and by some other types of phosphate mobilizing/oxidative stress conditions. Plant J. 19: 579-589.
Duff SMG, Sarath G, Plaxton WC (1994). The role of acid phosphatase in plant phosphorus metabolism. Physiol Plant. 90: 791-800.
Flanagan JU, Cassady AI, Schenk G, Guddat LW, Hume DA (2006). Identification and molecular modeling of a novel, plant-like, human purple acid phosphatase.
Gene. 377: 12-20.
Goldstain A, Baertlein D, McDaniel R (1988a). Phosphate starvation inducible metabolism in Lycopersicon esculentum: I. Excertion of acid phosphatase by tomato plants and suspension-cultured cells. Plant Physiol. 87: 716-720.
Hammond JP, Bennett MJ, Bowen HC, Broadley MR, Eastwood DC, May ST, Rahn C, Swarup R, Woolaway KE, White PJ (2003). Changes in gene expression in Arabidopsis shoots during phosphate starvation and the potential for developing smart plants. Plant Physiol. 132: 578-596. 
Hilson P, Allemeersch J, Altmann T, et al. (2004). Versatile gene-specific sequence tags for Arabidopsis functional genomics: transcript profiling and reverse genetics applications. Genome Res. 14: 2176-89.
Huang HB, Horiuchi A, Goldberg J, Greengard P, Nairn AC (1997). Site-directed mutagenesis of amino acid residues of protein phosphatase 1 involved in catalysis and inhibitor binding. Proc Natl Acad  Sci U.S.A. 94: 3530-5.
Jakobek JL, Lindgren PB (2002). Expression of a bean acid phosphatase cDNA is correlated with disease resistance. J Exp Bot. 53: 387-9.
Klabunde T, Starter N, Witzel H, Frohlich R, Krebs B (1996). Mechanism of Fe (III)-Zn (II) purple acid phosphatases based on crystal structures. J Mol Biol. 259: 737-748.
Letunic I, Goodstadt L, Dickens NJ, Doerks T, Schultz J, Mott R, Ciccarelli F, Copley RR, Ponting CP, Bork P (2002). Recent improvements to the SMART domain-based sequence annotation resource. Nucl Acids Res. 30: 242-244.
Li D, Zhu H, Liu X, Leggewie G, Udvardi M, Wang D (2002). Purple acid phosphatases of A. thaliana. Comparative analysis and differential regulation by phosphate deprivation. J Biol Chem. 277: 72-81.
Ling P, Roberts RM (1993). Uteroferrin and intracellular tartrate-resistant acid phosphatases are the products of the same gene. J Biol Chem. 268:6896-6902.
Malboobi MA, Lefebvre D (1997). A phosphate-starvation inducible beta-glucosidase gene (psr3.2) isolated from Arabidopsis thaliana is a member of a distinct subfamily of the BGA family. Plant Mol Biol. 34: 57-68.
Marshner H (1995). Mineral nutrition of higher plants, Edn 2. Boston: Academic Press.
Mertz P, Yu L, Sikkink R, Rusnak F (1997). Kinetic and spectroscopic analyses of mutants of a conserved histidine in the metallophosphatases calcineurin and lambda protein phosphatase. J  Biol Chem. 272: 21296-302.
Mimura T, Kurra-Hutta M, Tsujimura T, Ohnishi M, Miura M, Okazaki Y, Mimura M, Maeshima M, Washitani-Nemoto S (2003). Rapid increase of vacuolar volume in response to salt stress. Planta. 216: 397- 402.
Mohsenzadeh S, Malboobi MA, Razavi K, Farrahani Achtiani S (2005). Physiological and molecular responses of Aeluropus lagopoides (Poacea) to water deficit. Environ Exp Bot. 56: 314-322.
Misson J, Raghothama KG, Jain A, Jouhet J, et al. (2005). A genome-wide transcriptional analysis using Arabidopsis thaliana Affymetrix gene chips determined plant responses to phosphate deprivation. Proc Natl Acad  Sci USA. 102: 11934-11939.
Miyasaka SC, Habte M (2001). Plant mechanisms and mycorrizal symbiosis to increase phosphorous uptake efficiency. Commun Soil Sci Plant Anal. 32: 1101-1147.
Murashige T, Skoog F (1962). A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant. 15: 473-497.
Nakai, K, Horton P (1999). PSORT: a program for detecting sorting signals in proteins and predicting their subcellular localization. Trends Biochem Sci. 24: 34-36.
Pan X, Liu H, Clark J, Jones J, Bevan M, Stein L (2003). ATIDB: Arabidopsis thaliana insertion database. Nucl Acids Res. 31: 1245-1251.
Parida AK, Das AB (2004). Effects of NaCl stress on nitrogen and phosphorus metabolism in a true mangrove Bruguiera parviflora grown under hydroponic culture. J Plant Physiol. 161: 921- 928.
Petters J, Gobel C, Scheel D, Rosah L (2002). A pathogen- responsive cDNA from Potato encodes a protein with homology to a phosphate starvation induced phosphatase. Plant Cell Physiol. 43: 1049-1053.
Pietrokovski S, Henikoff JG, Henikoff S (1996). The Blocks database- a system for protein classification. Nucl Acid Res. 24: 197-200.
Robinson NJ, Procter CM, Connolly EL, Guerinot ML  (1999). A ferric-chelate reductase for ion uptake from soils. Nature. 397: 694-697.
Rodriguez H, Fraga R (1999). Phosphate solubilizing bacteria and their role in plant growth promotion. Biotech Advances. 17: 319-339.
Saleki R, Young PG, Lefebvre D (1993). Mutants of Arabidopsis thaliana Capable of Germination under Saline Conditions. Plant Physiol. 101: 839-845.
Sambrook J, Fritsch EF, Maniatis T (1998).  Molecular cloning. A laboratory manual. 2nd Ed. Cold Spring Harbor Laboratory Press, New York.
Schultz J, Copley RR, Doerks T, Ponting CP, Bork P (2000). A Web-based tool for the study of genetically mobile domains. Nucl Acid Res. 28: 231-234.
Shenck KG, Korsinczky ML, Hume DA, Hamilton S, Dejersey J (2000a). Purple acid phosphatases from bacteria: similarities to mammalian and plant enzymes. Gene. 255: 419-424.
Tester M, Smidt SE, Smidt FA (1987). The phenomenon of nonmycorrhizal plants. Can J Bot. 65: 419-431.
Tomscha JL, Trull MC, Deikman J, Lynch JP, Guiltinan MJ (2004). Phosphatase under-producer mutants have altered phosphorus relations. Plant Physiol. 135: 334-45.
Trull MC, Deikman J (1998). An Arabidopsis mutant missing one acid phosphatase isoform. Planta. 206: 544-550.
Tsyguelnaia I., Doolittle RF (1998). Presence of a fibronectin type III domain in a plant protein. J Mol Evol 46: 612-614. 
Uppenberg J, Lindqvist F, Svensson C, Ek-Rylander B, Andersson G (1999). Crystal structure of a mammalian purple acid phosphatase. J Mol Biol. 290: 201-11.
Wu P, Ma L, Hou X, Wang M, Wu Y, Liu F, Deng XW (2003). Phosphate starvation triggers distinct alterations of genome expression in Arabidopsis roots and leaves.
Plant Physiol. 132: 1260-1271.
Yamada K, Lim J, Dale JM, et al (2003). Empirical Analysis of the Transcriptional Activity in the Arabidopsis Genome. Science. 302: 842-849.
Zakhleniuk OV, Raines CA, Lioyd JC (2001).  pho3: a phosphorus-deficient mutant of Arabidopsis thaliana (L.) Heynh Planta. 212: 529-534.
Zhang J, Zhang Z, Brew K, Lee EY (1996). Mutational analysis of the catalytic subunit of muscle protein phosphatase-1. Biochem. 35: 6276-82.
Zhu H, Qian W, Lu X, Li D, Liu X, Liu K, Wang D (2005). Expression patterns of purple acid phosphatase genes in Arabidopsis organs and functional analysis of AtPAP23 predominantly transcribed in flower. Plant Mol Biol. 59:581-594.