Gossypium barbadense L. variety 7124 (resistant to Verticillium dahliae Kleb.) and G. hirsutum L. variety Ejing-1 (susceptible) were used in this study. Their seeds were first sterilized with 75% ethanol, and then rinsed with distilled water. After sterilization, the seeds were put in the pasteurized sands for germination and growth in the 15 cm × 10 cm × 10 cm plastic pots. The seedlings had grown at 25˚C - 28˚C and watered with Hoagland solution for 10 - 15 days. When plants had 2 - 3 leaves, the seedlings were subjected into various stresses.

A kind of defoliating V. dahliae isolate from infected cotton plants was used as the pathogen in this experiment. A single spore from the potato-dextrose agar (PDA) plate was inoculated into the Czapek Broth (NaNO₃, 0.3% w/v; MgSO₄, 0.1% w/v; KH₂PO₄, 0.1% w/v; FeSO₄, 0.0002% w/v; KCl, 0.1% w/v; Sucrose, 3% w/v; pH 6.0) to incubate for 5 - 8 days until the concentration of spores reached about 1.0 × 10⁷ spores/ml. The suspension liquid was adjusted to 1 × 10⁵ spores/ml with sterile distilled water to be used in inoculation experiment afterwards referring to the report [2]. The seedlings were infected with V. dahliae by root-dip inoculation into the suspension solution of fungal conidia for 5 min, and returned to their original pots for interval harvesting infected roots.

For the salt treatment, the plants were stressed by the Hoagland solution with addition NaCl of 100 mM. High temperature stress was imposed by transferring plants into 45˚C artificial atmospheric bank. For aerobic stresses, the whole plants were submerged into Hogaland solution. Plant growth regulator ABA was first dissolved into 50% ethanol and then diluted into the solution to a final concentration of 10 μM to spray cotton leaves and to submerge their roots. For ethylene treatment, plants were placed in a 10-litre jar with 200 mg/L 40% Ethephon spray.

The roots of the plants were harvested in 0, 1, 3, 7, 12, 24, 48, 72 hours after different stress treatments. All the roots were used to extract total RNA immediately.

Total RNA was isolated from above roots following the method described by the brochure of Trizol kit (GIBCOBRL, CA, USA). PolyA⁺ RNA was purified by oligo- (dT)_n cellulose affinity chromatogy (Invitrogen, CA, USA). Double stranded cDNAs were then prepared from polyA⁺ RNA according to the manufacturers’ recommendations (Takara, Liaolin, China). Genomic DNA was isolated with CTAB method according to the method of Sambrook et al. [24].

G. babardense genomic DNA (10 μg) was digested with EcoRI, BamHI and PstI endonucleases. Fragments were separated with 0.8% agarose gel and blotted onto HybondN⁺ membrane (Amersham, England, UK). The probe was prepared with a segment of GbPR10-1 coding segment (92 - 570 nt). The Gene Images random primer labeling module and Gene Images CDP-Star detection module were used for probe labeling, hybridization, and detection procedures (GIBCO-BRL, CA, USA).

Total RNA (1 μg) after digesting genomic DNA with DNaseI was added into a tube containing first-strand cDNA synthesis and PCR reaction buffer mixture according to the brochure of Takara RT-PCR kit (Takara, Dalian, China). The primers used in the RT-PCR were PR10-1(5’-C¹⁰³GA GTT ATG AGT TTG AGG TAA-3’) and PR10-2 (5’-G²³⁹AC TAG CAT CAC CTT CGA GCT-3’), which could amplify 140 bp fragment of different PR10-1 genes in both sea-island and upland cotton varieties. After the first cDNA strand was synthsized, the PCR reactions went on quickly without pulling out tubes or adding any components. The PCR reaction program was 94˚C, 30 seconds; 58˚C, 30 seconds; 72˚C, 30 seconds; 25 cycles. PCR products (10 μl) were then electrophorsized in 1.2% agarose gel with 10 μg/ml ethidium bromide. The quantity of products was analyzed with gene analysis software package in the White/Ultraviolet Trans-illuminator (UVP^®, CA, USA). The results were then used for determining the expression levels of PR10 gene in different intervals.

The template is used to format your paper and style the text. All margins, column widths, line spaces, and text fonts are prescribed; please do not alter them. You may note peculiarities. For example, the head margin in this template measures proportionately more than is customary. This measurement and others are deliberate, using specifications that anticipate your paper as one part of the entire proceedings, and not as an independent document. Please do not revise any of the current designations.

The plasmid DNA of the positive clones were isolated and sequenced by Sangon Company in China. Sequence similarity was analyzed with BLAST based on GenBank (http://www.ncbi.nlm.nih.gov). The amino acid sequence encoded by GbPR10-1 and its homological proteins were aligned using the ClustalX programs.

PR10 family proteins were screened using tblastX on GenBank (http://www.ncbi.nlm.nih.gov). When the similarity between two sequences is more than 99.5%, these two sequences were considered as the same sequence. All of nucleotide sequences (EST and CDS from G. babardense and G. hirsutum) which probably encode PR10 proteins were then assigned into PR10-1, PR10-2 etc. These PR10 protein sequence were then used in bioinformatics analysis afterwards. Phylogenetic tree was constructed by using the neighbor-joining method [25,26].

The ORF of GbPR10-1 gene was amplified from pMD18- T vector (Takara, Japan) plasmid which contained the full-length cDNA of GbPR10-1 gene. The PCR products digested by NotI/SalI were sub-cloned into the plasmid pET32-a. The plasmid of a positive clone was then transformed into the E. coli strain BL21 cell to get the recombinant expression vector. Isopropyl-1-thio-b-D-gactopyranoside (IPTG, 4 mM in the end) was added to the LB solution with Ampecillin (50 mg/L) when OD600 of the cultures reached 0.6. The cells were further cultured for 3 h and harvested by centrifugation at 4000 rpm 4˚C for 10 min.

The pellets were resolved in 50 mM Tris-HCl (pH 7.5) buffer containing 0.5 M NaCl, and 1 mM PMSF and lysised by sonication. Following centrifugation at 15,000 rpm 4˚C for 30 min, the supernatant was loaded onto SDS-PAGE gel to identify whether GbPR10-1 protein expressed highly or not in vitro. The supernatant of highly expressed clone was then loaded into a His-tag Sepharose-Fast-Flow gel column (Amersham-Pharmacia Biotech, USA) and washed with a gradient of 100 mM to 1M solution to collect recombinant protein.

RNase activity of the purified negative and recombinant GbPR10-1 protein was carried out at 37˚C. The reaction mixture contained 6 μg cotton total RNA and 4 μg proteins in 10 mM imidazole, 5 mM NaCl, and 10 mM Tris-HCl (pH 7.5) buffer. After 15 and 30 minutes incubation, the proteins were removed from the reaction mixtures by extraction with phenol-chloroform. The control mixture described above without GbPR10-1 protein was also incubated for 15 and 30 minutes at room temperature. The experimental results were observed on 1.2% agarose gel to determine the enzyme activity of protein.

Our previous report showed that an EST (Gb125, accession number CB066575) highly similar to other PR10 gene was strongly induced after V. dahliae inoculation [21]. This EST was than used as the probe to screen a sea-island variety 7124 root cDNA library. Seven clones were isolated in which two clones were full-length cDNA after two-round screening. The cDNA clone (Figure 1) contained an insert of 745 nt (nucleotide) with a complete coding frame of 477 bp from 91 nt to 568 nt. The start codon ATG site and a stop codon TAA site were recognized after comparison with other PR10 genes and confirmed by other homologous sequences in GenBank. Within 3’-untranslated region one polyadenylation

signal AATAAA was found at 68 bp upstream from the poly-(A)_n tail. Blast search showed that this gene was 98%, 86%, and 76% identical to GaPR10 (Accession NO. AF416652.1), GhPR10 (Accession NO. AF305065.1) and Betv1-sc3 (Accession NO. AF2122374C) at nucleotide level respectively, in which Betv1-sc3 was a typical PR10 protein. According to rule of gene nomenclature, this gene was then called as GbPR10-1 (Figure 2).

The deduced GbPR10-1 protein was 159 aa in length with a molecular weight of 17.9 kDa, which is a highly hydrophilic acid protein with predicted pI 4.95. The comparison of amino acid sequences revealed that GbPR10-1 shared 98%, 79%, 51% and 52% homology to G. arboum, G. hirsutum, Corylus avellana and Betula pendula PR10 proteins respectively. The conserved sequence K-A-X-EX-Y-L was also found in C-terminal from 145 aa to 151 aa (Figure 1). Several amino acids influencing PR10 protein RNase activity were detected in Gly⁵¹, Lys⁵⁵, and Glu⁹⁶, but the conserved RNase binding site P-loop structure G-X-G-G-X-G in Betv1 protein was not found in GbPR10-1 protein [27]. This motif is known as “P-loop” (phosphate-binding loop) and is frequently found with variations in protein kinases as well as in nucleotidebinding proteins [28]. Therefore, it indicated that RNase activity of GbPR10-1 protein was influenced by the change of this motif.

Genomic walking showed that GbPR10-1 gene has no intron in the coding sequence. Southern analysis (Figure 3) indicated that there are two copies in the EcoRI lane, and two copies in the BamHI digestion lanes. Endonucleases restriction analysis showed that only one BamHI digestion site in the probe region (from nucleotide 300 to nucleotide 700), indicating there is one copy GbPR10-1 gene in the G. babardense genome.

Recent data showed that different PR10 protein member is involved in controlling the growth or pathogens stimuli. In order to analysis relationship among different members in cotton, we screened GeneBank using tblastx and found that there are 9 different members in upland cotton. Phylogenetic tree of these putative PR-10 proteins from G. barbadense and G. hirsutum showed that all of these PR10 proteins could be classified into three subfamilies (Figure 4). Subfamily 2 was composed of PR10-2, PR10- 3, PR10-5 and PR10-6 proteins. Subfamily 3 was composed of PR10-4, PR10-7, PR10-8 and PR10-9 proteins. PR10-1, which showed the lowest similarity with other PR10 proteins, was assigned into subfamily 1. Expression pattern of subfamily 3 showed that these genes were highly expressed in developing cotton fiber (data not shown). The Expression patterns of PR10 genes in subfamily 2 were not typical. Subfamily 1 is composed of only one PR10 protein (Figures 4 and 5), indicating that the function of GbPR10-1 gene needs to be analyzed in detail.

To examine the biotic and abiotic stresses on GbPR10-1 gene expression in the cotton plant, we analyzed the expression change using RT-PCR strategy with total RNA from the roots of plants subjected to different stress conditions. Plants grown in pots until two weeks were treated with salt, high temperature, ABA, anaerobic and pathogens attack. In all experiments, two primers PR10-1 and PR10-2 were used to amplify the coding region of PR10 and they displayed different patterns of expression under different conditions (Figure 6).

The expression pattern showed that GbPR10-1 gene transcripts did not decline after the fungus treatment; it subsequently increased and reached the maximum levels at 7 hours then decreased gradually to the normal level at 72 h. Differently, the expression of GhPR10 firstly declined after pathogens inoculation, return to normal condition in the 48 h after inoculation. For G. hirsutum variety Ejing-1, it was too sensitive to endure long-time pathogens attack and plants were wilt in 12 h post-inoculation and died after 72 hours treatment, so the total RNA were not extracted till 12 h after the pathogen treatment. However, the GhPR10 gene expression change was still clearly detected. The highest expression level of GhPR10 transcripts after inoculation was not up-regulated, and its expression levels under different intervals were much lower than that of GbPR10-1 in 7124.

Meanwhile, 1 µg of total RNA from G. barbadense with 100 mM NaCl was also adopted to analyze GbPR10- 1 gene expression. Compared to untreated plants, the expression levels of the GbPR10-1 had increased about three-fold and reached maximum levels till 24 h and this expression level did not decrease in 72 h after inoculation.

(A)                                                  (B)

Figure 6. Expression profiles of PR10 gene under abiotic and biotic stresses in cotton. Total RNA (1 µg/lane) was isolated at 0, 1 h, 3 h, 7 h, 12 h, 24 h, 48 h, 72 h after exposure to different stresses respectively, and subjected to one-step RT-PCR amplification (upper panel). The entire experiments were repeated twice using total RNA isolated from the roots of cotton as templates. Ubiquitin cDNA was used to normalize the amount of templates added in PCR reactions (lower panel). (A) GbPR10-1 gene expression profile under biotic stress including heat, salt, submergenece, and ABA; (B) Comparison of PR10 gene expression in sea-island and upland cotton under V. dahliae attack.

Upon exposure of plants to ABA solution, the transcript level of GbPR10-1 remained relatively constant in root of cotton. That means that GbPR10-1 gene does not respond to ABA induction.

The seedling was too young to endure long-time heat treatment and plants were wilted after 24 h high temperature and then died, so the total RNA were extracted till 24 h after the heat treatment. During the high-temperature stress (45˚C), the GbPR10-1 gene transcripts of G. barbadense were not declined in the first 7 h and subsequently increased and reached the highest level at 24 h.

GbPR10-1 fusion protein was expressed in E. coli after induction with IPTG for 4 hr. The result showed that recombinant GbPR10-1 is about 34 kDa (Figure 7) in the lane 3 which was consist with the putative size of the fusion protein. This fusion protein isolated by an affinity column was incubated with the V. dehaliae RNA, and it could cleave the total RNA into fragments slowly (Figure 8(a)). After incubation with V. dahliae on the plate for 4 days, GbPR10-1 show very low inhibition activity on hyphal growth of V. dahliae but not for its spore germination (Figure 8(b)).