The role of a purine-specific nucleoside hydrolase in spore germination of Bacillus thuringiensis

Enzymes with nucleoside hydrolase (NH) activity are widely distributed in both prokaryotes and eukaryotes (Giabbai & Degano, 2004). Non-specific NHs, the most well-known of which is the inosine/uridine-preferring NH (IU-NH) from Crithidia fasciculata, hydrolyse both purine and pyrimidine nucleosides (Parkin et al., 1991). Purine-specific NHs (inosine/adenosine/guanosine-preferring NHs] display strict specificity (Parkin, 1996; Pelle et al., 1998; Versees et al., 2001). NHs are crucial in the salvage pathway of protozoan parasites which are deficient in the pathway for de novo synthesis of purine nucleotides (Gopaul et al., 1996). Non-parasitic NHs, for example the salivary purine nucleosidase of the mosquito Aedes aegypti, may be involved in the degradation of modified nucleosides found in tRNA (Giabbai & Degano, 2004; Ribeiro & Valenzuela, 2003), but apart from this, no clear role for non-parasitic NHs, including the NH identified from the exosporia of Bacillus anthracis and Bacillus cereus, has been reported so far (Giabbai & Degano, 2004; Redmond et al., 2004; Ribeiro & Valenzuela, 2003; Todd et al., 2003).

Bacillus thuringiensis, B. anthracis and B. cereus are members of the Gram-positive endospore-forming B. cereus group. They could be classified as one species on the basis of genetic evidence, even though they demonstrate widely different phenotypes and pathological effects (Helgason et al., 2000). As a natural pesticide, commercial formulations of B. thuringiensis containing separate entities of crystals and spores has become the leading biological insecticide used to control agricultural pests (Crickmore, 2006; Liu et al., 1998). B. anthracis is the cause of the acute and often lethal disease anthrax, and B. cereus is a ubiquitous soil bacterium and opportunistic human pathogen (Helgason et al., 2000).

Under conditions of nutrient depletion at high cell density, vegetative cells of Bacillus species can transform into spores by a process called sporulation. The spore is metabolically dormant, and resistant to heat, radiation, desiccation, pH extremes and toxic chemicals. The dormant spore also monitors its environment, and when conditions become favourable again, the spore germinates and is converted back into a new vegetative cell (Setlow, 2003). In nature, spores probably germinate in response to nutrient germinants which are generally single amino acids, sugars or purine nucleosides (Setlow, 2003). Among nutrient germinants, the combination of L-alanine and inosine is the best germinant for most Bacillus spores. Inosine is an independent germinant for B. cereus and B. thuringiensis spores, and an important co-germinant in B. thuringiensis, B. cereus and B. anthracis spore germination (Foerster & Foster, 1966; Hornstra et al., 2006).

The first event in nutrient-induced spore germination is probably the activation of the germinant receptors, which are located in the inner membrane of the spore (Hornstra et al., 2005). To activate the receptors, germinants must first penetrate the outer spore layers and access their corresponding receptors. The outermost layer of the B. cereus group spore is a loose-fitting, balloon-like structure known as an exosporium, which consists mainly of protein, polysaccharides, lipids and ash (Matz et al., 1970). Dozens of proteins, including a putative NH, were identified from the exosporia of B. cereus and B. anthracis spores (Redmond et al., 2004; Steichen et al., 2003; Todd et al., 2003). Although the role of some proteins has been studied, most functions remain to be elucidated (Boydston et al., 2006; Ramarao & Lereclus, 2005; Steichen et al., 2005; Yan et al., 2007).

Widespread distribution of NH raised the question of whether these genes indeed encode NH enzymes, and most importantly, what their role is, especially in non-parasitic organisms which could recycle nitrogenous bases via NP-catalysed phosphorolysis (Giabbai & Degano, 2004). In this paper, we have characterized one non-parasitic, purine-specific NH from B. thuringiensis and demonstrated its role in moderating inosine- or adenosine-induced spore germination.

Bacterial strains, plasmids and culture conditions.
All the strains and plasmids used in this study are listed in Table 1. B. thuringiensis was routinely grown in Luria–Bertani (LB) medium at 30 °C containing appropriate antibiotics (25 µg erythromycin ml^–1 or 10 µg chloramphenicol ml^–1). Casein hydrolysate/yeast-containing medium (CCY) was used for spore preparation (Stewart & Halvorson, 1953). For subcloning, Escherichia coli DH5α was grown at 37 °C in LB medium, containing ampicillin, erythromycin or chloramphenicol when necessary, for propagating plasmids.

Table 1. Bacterial strains and plasmids used in this study

DNA manipulations and transformation.
B. thuringiensis genomic DNA was isolated as described previously (Gonzalez et al., 1981). Primers P1 (5'-GCTCTAGAGAACCGATAATACCAGC-3', XbaI site is underlined), P2 (5'- GGGGTACCGAAGTCGCCAATAAATAG-3', KpnI site is underlined), P4 (5'-GGGGTACCGTCATAATAATCGTCTTCTTG-3', KpnI site is underlined) and P5 (5'-GCTCTAGACTAAATGATGAATTGGACC-3', XbaI site is underlined) were designed according to the conserved sequence of the B. cereus group. The DNA fragment containing iunH was amplified from genomic DNA of B. thuringiensis subsp. kurstaki CGMCC 1.1752 with primers P1 and P2. After digestion with XbaI and KpnI, the amplified DNA fragment was ligated into the corresponding sites of pKSV7 to generate pIC. Sequencing of the fragment was carried out by the SunBiotech Company (Beijing, PR China). Southern hybridization was carried out with probes labelled with a digoxigenin DNA labelling kit (Roche Biochemicals), according to the manufacturer's instructions. All PCRs were performed with Pfu DNA polymerase (TaKaRa) using standard conditions.

For B. thuringiensis electroporation, 80 µl cell suspensions in 40 % PEG were used per cuvette plus 1 µl (about 1 µg) plasmid DNA (dissolved in distilled water) at 11 kV cm^–1, 1000 Ω and 25 µF.

Construction of the iunH disruption mutant.
To construct an iunH disruption mutant, a DNA fragment corresponding to the upstream region of iunH (extending from positions –1003 to +223 with respect to the iunH translation initiation site) was amplified with primers P3 (5'-ACGCGTCGACGCTGTCATCGGTCTACTC-3', SalI site is underlined) and P4, and a DNA fragment corresponding to the downstream region of iunH (extending from positions –306 to +846 with respect to the iunH stop codon) was amplified with primers P5 and P6 (5'-CGGGATCCAATGATTGATTTTTATTATGAG-3', BamHI site is underlined) from B. thuringiensis. These PCR products were digested with SalI and BamHI, respectively. A 1256 bp DNA fragment containing the erythromycin resistance gene (erm) was amplified from pHT3101 with primers P7 (5'-ATAGGATCCAATAAGGGCGACACG-3', BamHI site is underlined) and P8 (5'-ACGCGTCGACCCCTTAGAAGCAAACT-3', SalI site is underlined), and digested with SalI and BamHI. The digested PCR fragments were purified, mixed in equal amounts and ligated with T4 DNA ligase. The ligation mixture was used as a template to amplify the complete tripartite DNA fragment with primers P4 and P5. Then, the amplified 3.6 kb DNA fragment was digested with XbaI and KpnI, and inserted into the corresponding sites of pKSV7 to give pID. Subsequently, pID was introduced into B. thuringiensis by electroporation, and cultured on LB plates containing erythromycin at 30 °C for 2 days. Transformants containing pID were selected and confirmed by plasmid isolation and digestion. One of the confirmed transformants was randomly selected and cultured overnight at 30 °C in LB broth without any antibiotic; then the culture was diluted and spread onto an LB plate containing erythromycin. After growing for about 10 h at 42 °C, colonies were replicated on LB plates containing chloramphenicol. Then, the replicated plates and the original plates were cultured at 30 °C. Chloramphenicol-sensitive and erythromycin-resistant strains were selected. Subsequently, the disruption mutant was confirmed by PCR analysis and Southern hybridization.

For complementation experiments, plasmid pIC, containing iunH and its putative promoter region, was transformed into the iunH disruption mutant by electroporation.

Spore preparation and germination analysis.
To prepare spores, B. thuringiensis was grown in CCY medium at 30 °C for 48 h. Spores were harvested and washed 5–10 times with cold distilled water. All spore preparations were free (>99 %) of vegetative and sporulating cells. For the germination assay, the heat-activated spores were diluted in germination buffer (10 mM NaCl, 10 mM Tris/HCl, pH 7.4). Small aliquots (1.2 ml) of the heat-activated spores at an OD₆₀₀ of 1 (about 1.5x10⁸ c.f.u. ml^–1) were supplemented with inosine (ranging from 0.01 mM to 10 mM), adenosine (0.1–2 mM) or guanosine (0.1–2 mM) separately, or 0.01 mM inosine with 1 mM L-alanine, and incubated at 37 °C. The decrease in OD₆₀₀ was monitored for up to 60 min for inosine-induced germination and 90 min for adenosine-induced germination. Phase darkening of germinated spores was observed by phase-contrast microscopy. The data obtained are means from triplicate experiments performed with three independent spore preparations.

Expression and purification of IunH.
To study the properties of IunH in vitro, it was necessary to obtain an adequate amount of IunH protein. Therefore, iunH was amplified from the genomic DNA of B. thuringiensis with the primers P9 (5'-GGACTTCCATATGAGAATAGTTAATAAGAAAA-3', NdeI site is underlined) and P10 (5'-CGGAATTCTTAAGGACAATCTGGCT-3', EcoRI site is underlined). The start codon (GTG) was replaced by ATG. The amplified fragment was digested with NdeI and EcoRI, and inserted into the corresponding sites of pET28a to generate a recombinant plasmid, pIE. Then, pIE was transformed into E. coli BL21(DE3) for high-level expression of iunH under the control of the T7 promoter. As the N terminus of IunH was designed to contain six consecutive histidines, the His6-tagged IunH was purified to homogeneity by Ni-NTA affinity chromatography. The concentration of the purified protein was determined by the method of Bradford using BSA as standard (Smith et al., 1985).

Enzyme assays.
Since IunH contains one consensus N-terminal {D, N}XDXXXDD aspartate cluster which is a fingerprint for NH enzymes, the enzymic activity of His6-tagged IunH was measured using inosine, adenosine, guanosine, uridine or cytidine as substrate. A volume (200 µl) of the reaction mixture, containing 50 mM HEPES (pH 7.3), the purified His6-tagged IunH and substrate, was incubated for 5 min at room temperature. Hydrolysis of the substrate (inosine, uridine or cytidine) was followed by continuous reading of A₂₈₀ on a Beckman DU-800 UV spectrophotometer. The conversion of a 1 mM solution of inosine, uridine or cytidine to products resulted in a change in A₂₈₀ of 0.92, 2.04 and 3.42, respectively, at pH 7.3 (Parkin et al., 1991). Hydrolysis of adenosine or guanosine was determined with the reducing sugar assay described by Parkin (1996).

The kinetic parameters K_m and V_max were determined at room temperature using a 1/v-1/[S] plot, where [S] is the concentration of inosine (between 0.2 and 2.5 mM). k_cat was derived from the equation k_cat=V_max/[E], where [E] is the concentration of IunH in the reaction mixture.

To check the effect of pH, IunH activity was measured in the following buffers: 100 mM potassium phosphate, 50 mM HEPES, 30 mM CHES and 30 mM MES, covering the pH range from 4 to 10. To assess the optimal temperature of the enzyme, the inosine hydrolase activity of IunH was assayed at temperatures ranging from 40 to 100 °C.

Inosine hydrolase activity of the intact spores and the vegetative cells.
Inosine hydrolase activity in the intact spores or the vegetative cells was determined spectrophotometrically using the difference in absorption between the nucleoside and the purine base (Parkin, 1996). All measurements were carried out at room temperature. Intact spores or vegetative cells at an OD₆₀₀ of about 1 were incubated with 0.5 mM inosine in a total volume of 900 µl of 50 mM HEPES (pH 7.3) for 0, 10, 20, 30 and 40 min, respectively. Spores or vegetative cells were removed by centrifugation. The supernatant was assayed for a change in absorbance at 280 nm, and the resultant pellets were dried at 42 °C for at least 24 h to determine their dry weight.

Cloning of an iunH-homologous gene from B. thuringiensis
BLAST searches of the B. cereus ATCC 14579 genomic sequence released by the DOE Joint Genome Institute () revealed five predicted proteins (GenBank accession numbers BC2331, BC2683, BC2889, BC3552 and BC5134) that are homologous to NH. The amino acid sequence of BC2889 is almost identical to the amino acid sequence of the putative NH purified from exosporia of B. cereus and B. anthracis (Redmond et al., 2004; Steichen et al., 2003; Todd et al., 2003). Based on the sequence of BC2889 and its homologues found in members of the B. cereus group, a 1.4 kb DNA fragment was amplified from B. thuringiensis genomic DNA by PCR. Sequencing showed that the DNA fragment contains an ORF, which was designated iunH (Fig. 1). iunH consists of 966 nt and encodes a protein containing 321 aa with a predicted molecular mass of 36.2 kDa. IunH contains one N-terminal {D, N}XDXXXDD aspartate cluster, which is a fingerprint for NH enzymes, and one conserved {V,I,L,M}HD{P,A,L} tetrapeptide sequence approximately 230 aa downstream from the N-terminal aspartate cluster (Giabbai & Degano, 2004) (Fig. 2).

(10K): Fig. 1. Disruption of the iunH gene in B. thuringiensis. The arrow represents iunH and its orientation, and the figure is to scale. The iunH gene was interrupted by an erythromycin-resistant cassette (erm) in the disruption mutant. The entire iunH gene with its promoter was inserted into pKSV7 to generate pIC.

(79K): Fig. 2. Amino acid sequence alignment of B. thuringiensis IunH with other NH proteins. Residue numbers are taken from the IunH sequence. (a) N-terminal region with the {D, N}XDXXXDD aspartate cluster conserved in all NH proteins. (b) Alignment of the region containing the {V,I,L,M}HD{P,A,L} tetrapeptide sequence approximately 230 aa downstream from the N-terminal aspartate cluster. These sequences were sourced as follows: IunH, B. thuringiensis (EU072023); IunH, B. cereus (YP084206); IunH, B. anthracis (NP845228); IunH, Leishmania major (P83851); IunH, Crithidia fasciculata (Q27546); URH1, Saccharomyces cerevisiae (AAG44107); YeiK, Escherichia coli (AAA60514); inosine/adenosine/guanosine-preferring NH (IAG-NH), Trypanosoma brucei brucei (XP843889); NH, Aedes aegypti (XP001651977); NH, Xenopus laevis (NP001079470); NH, Drosophila melanogaster (NP572912).

Disruption of iunH increases the rate of spore germination initiated by inosine or adenosine
To investigate the physiological role of IunH in B. thuringiensis, an iunH disruption mutant of B. thuringiensis was constructed by homologous recombination (Fig. 1). The iunH disruption mutant showed normal colony morphology, growth and sporulation. Spore germination was measured by monitoring the decrease in OD₆₀₀ of a spore suspension at 37 °C after the addition of either inosine or adenosine at the desired concentrations to the heat-activated spores. The dependence of spore germination on inosine concentration was measured in the wild-type strain, the iunH disruption mutant and the complemented strain. The germination rates of all spores were increased when the concentration of inosine was increased from 0.01 to 1 mM. As the germination rate was low when spores were induced by 0.01 mM inosine, no significant difference could be observed among the three strains (Fig. 3a). When 0.1 mM inosine was used, the OD₆₀₀ of the spore suspension of the iunH disruption mutant decreased by about 45 % during germination in contrast to a decrease of about 20 % in the wild-type spore suspension. As expected, the OD₆₀₀ of the complemented strain spores decreased by less than 10 %, which is even less than that of the wild-type spores (Fig. 3b). Phase-contrast microscopy showed that all the iunH mutant spores became completely phase-dark after incubation with 0.1 mM inosine for 30 min, while only some of the wild-type spores and almost none of the complemented strain spores turned phase-dark under these conditions (Fig. 4). When the concentration of inosine was increased to 1 mM, the germination rates of both the iunH disruption mutant and the wild-type spores increased significantly. However, the germination rate of complemented strain spores increased only slightly (Fig. 3c) and complemented strain spores could only germinate completely at a final concentration of 10 mM inosine (data not shown). L-Alanine (1 mM) could not induce spore germination, but concomitant addition of 0.01 mM inosine could stimulate the germination in the wild-type strain and in the iunH disruption mutant, while the germination rate of complemented strain spores did not increase (data not shown).

(16K): Fig. 3. Effect of iunH on spore germination initiated by inosine in B. thuringiensis. Spores of the wild-type (), the iunH disruption mutant () or the complemented strain () were heat-activated and subsequently incubated in germination buffer supplemented with 0.01 (a), 0.1 (b) or 1 mM (c) inosine. The decrease in OD600 was measured periodically and plotted as a percentage of the initial OD600 versus time.

(93K): Fig. 4. Nucleoside-induced, germination-associated, phase bright–phase dark transition efficiency in B. thuringiensis. All the iunH mutant spores became completely phase-dark after incubation with 0.1 mM inosine for 30 min, while only some of the wild-type spores and almost none of the complemented strain spores turned phase-dark under these conditions.

In addition to inosine, adenosine has been found to be required for spore germination of several Bacillus species (Hills, 1949; Lawrence, 1955; Hornstra et al., 2006). Adenosine-induced germination of wild-type spores, the iunH mutant spores and the complemented strain spores was assayed in B. thuringiensis. As with inosine, the germination rate increased with increasing concentrations of adenosine. When 0.5 mM adenosine was used, the OD₆₀₀ of the iunH disruption mutant spore suspension decreased by about 50 % during germination in contrast to a decrease of about 30 % in the wild-type spore suspension. However, the OD₆₀₀ of the complemented strain spore suspension decreased by only about 10 % (Fig. 5). Only a small percentage of complemented strain spores could germinate (date not shown) even at the highest concentration of adenosine used (about 3 mM, due to the low solubility of adenosine). In addition, guanosine, which is also a substrate of IunH, barely induced spore germination. Therefore, iunH can moderate the inosine- or adenosine-induced spore germination rate in B. thuringiensis.

(11K): Fig. 5. Effect of iunH on spore germination initiated by 0.5 mM adenosine in B. thuringiensis. Spores of the wild-type (), the iunH disruption mutant () or the complemented strain () were heat-activated and subsequently incubated in germination buffer supplemented with 0.5 mM adenosine. The decrease in OD600 was measured periodically and plotted as a percentage of the initial OD600 versus time.

Substrate specificity of IunH
To characterize its function and substrate specificity, iunH was overexpressed in E. coli BL21(DE3). Although the bulk of the expressed protein was in inclusion bodies, a clear band of IunH was observed in the supernatant of the extracts of E. coli BL21(DE3)/pIE (Fig. 6). His6-tagged IunH was purified to homogeneity by Ni-NTA affinity chromatography and the enzyme activity of IunH was measured with inosine, adenosine, guanosine, uridine or cytidine as substrate. As shown in Table 2, IunH can hydrolyse inosine, adenosine and guanosine with the following activity order: inosine >adenosine >guanosine. The K_m and K_cat values with adenosine or guanosine as substrate were not calculated due to the low sensitivity of the method employed for assaying reducing sugars. Hydrolase activity could not be detected when uridine or cytidine was used as a substrate. Thus, IunH belongs to a class of purine-specific NHs (inosine/adenosine/guanosine-preferring NH), based on its substrate specificity.

(69K): Fig. 6. SDS-PAGE of IunH expressed in E. coli and its purification. Lanes: M, standard molecular mass markers; 1, total protein of IPTG-uninduced BL21(DE3) containing pIE; 2, total protein of IPTG-induced BL21(DE3) containing pIE; 3, supernatant of IPTG-induced BL21(DE3) containing pIE; 4, purified IunH protein from Ni-NTA affinity chromatography. Samples were separated on an 8 % SDS-PAGE gel and stained with Coomassie brilliant blue R-250. The arrow indicates the position of IunH.

Table 2. Specific activity of IunH with different substrates

Kinetic characterization of the recombinant IunH
Kinetic studies were used to substantiate the catalytic properties of the expressed recombinant IunH. In standard reaction buffer (50 mM HEPES, pH 7.3) with inosine as substrate, the specific activity of the inosine hydrolase of IunH was 57.2±2.56 µmol min^–1 (mg protein)^–1. The kinetic parameters of IunH were K_m=399±115 µM, k_cat=48.9±8.5 s^–1 and k_cat/K_m=1.23x10⁵ M^–1 s^–1.

The optimal pH and temperature for IunH
The effect of pH on the inosine hydrolase activity of IunH was examined at room temperature with 3 mM inosine as substrate. Inosine hydrolase activity of the purified recombinant IunH could be detected over a wide range of pH values, the highest activity being obtained at pH 6 (Fig. 7a). The activity was stable over a wide temperature range, the highest activity being at 80 °C with high activity being maintained up to 100 °C (Fig. 7b). This result indicates that IunH is a highly heat-stable enzyme.

(12K): Fig. 7. Activity of purified IunH as a function of pH (a) and temperature (b). The inosine hydrolase activity of IunH was measured at room temperature in 100 mM potassium phosphate, 50 mM HEPES, 30 mM CHES or 30 mM MES, covering the pH range from 4 to 10. To assess the optimal temperature of the enzyme, the inosine hydrolase activity of IunH was assayed at temperatures ranging from 40 to 100 °C. The activity of IunH at pH 6.0 or at 40 °C was defined as 100 %.

Activity of inosine hydrolase in dormant spores
Sequence comparisons have revealed that B. thuringiensis IunH has a high level of similarity with its homologues purified from the exosporia of B. cereus and B. anthracis (Redmond et al., 2004; Steichen et al., 2003; Todd et al., 2003). IunH is probably located in the exosporium of B. thuringiensis spores and, therefore, inosine hydrolase activity in intact spores was measured. The specific activity of inosine hydrolase in wild-type spores of B. thuringiensis was 2.89±0.23x10^–2 µmol min^–1 (mg dry wt)^–1, while the iunH mutant spores exhibited no detectable activity. The specific activity in the complemented strain spores was 8.40±0.69x10^–2 µmol min^–1 (mg dry wt)^–1, which is almost threefold higher than observed in the wild-type spores (Fig. 8). The higher activity of IunH in complemented strain spores may be due to the introduction of multiple copies of iunH by pKSV7 (Smith & Youngman, 1992). Although there was considerable IunH activity in the spores, no activity was detected in intact vegetative cells from the wild-type, iunH disruption mutant and complemented strains (data not shown).

(11K): Fig. 8. Activity of inosine hydrolase in whole, dormant spores. Whole spores of the wild-type (), the iunH disruption mutant (•) or the complemented strain () were incubated with 0.5 mM inosine in a total volume of 900 µl 50 mM HEPES (pH 7.3) for 0, 10, 20, 30 and 40 min, respectively. Spores were removed by centrifugation and the supernatant was assayed for the change in OD280. Spore preparations whose optical density was examined in advance, were centrifuged and the deposits were dried at 42 °C for at least 24 h to determine their dry weight.

Inosine-initiated spore germination starts with the interaction of inosine with its corresponding receptors, such as GerI, GerQ and GerR. Then the spores undergo the release of dipicolinic acid and cations, hydrolysis of the peptidoglycan cortex, germ-cell-wall expansion and finally resumption of vegetative growth (Hornstra et al., 2006; Setlow, 2003). This process has been described in detail, but how it is regulated is still poorly understood. Here, we provide evidence that the amount of inosine or adenosine acting as germinant can be modulated by the activity of the inosine hydrolase IunH, which converts inosine to hypoxanthine and ribose, and adenosine to adenine and ribose. This allows for modulation of the inosine- or adenosine-induced germination efficiency. Disruption of iunH indeed resulted in an increase in the inosine- or adenosine-induced germination rate, whereas overexpression in the complemented strain resulted in a significant decrease of its germination-triggering capacity.

Unlike the NH from C. fasciculata, whose activity decreases rapidly at pH values below 7 (Parkin et al., 1991), B. thuringiensis IunH showed highest inosine hydrolase activity at pH 6. Given the fact that the spores can germinate in the alkaline environment of the insect gut (Schnepf et al., 1998), the relatively low activity of IunH at alkaline pH values may increase the sensitivity of spores to the germinant. In addition, the heat stability of IunH enables it to function even after the spores have been exposed to high temperatures, which is especially important when dealing with spores in feeds and foods.

The in vivo inosine hydrolase activity indicates that the spores are able to monitor inosine levels in their environment. The difference in inosine hydrolase activity of wild-type, disruption mutant and complemented strain spores has confirmed the role of the inosine hydrolase function of IunH in modulating inosine-induced germination. Five ORFs encoding putative inosine-preferring NHs have been found in the genomes of B. cereus strain ATCC 14579 and four ORFs have been found in B. cereus strain ATCC 10987 (Ivanova et al., 2003; Rasko et al., 2004). This number is similar to that found in the genome of B. thuringiensis (Challacombe et al., 2007). The disruption of iunH eliminates inosine hydrolase activity in spores and promotes inosine-initiated germination of B. thuringiensis. This indicates that IunH is the most prominent NH in the exosporium, moderating nucleoside-induced germination capacity in B. thuringiensis.

We are grateful to Professor Dafang Huang (Biotech Institute, CAAS, China) for the gift of plasmid pHT3101. We are also grateful to Dr Paul Babitzke (Penn State University, PA, USA) for plasmid pKSV7. This work was supported by grants from the National Natural Science Foundation of China (Grant Nos. 30430010) and the National Basic research Program of China.

Edited by: T. Abee