EXPRESSION OF THE SIGMA 35 AND CRY 2 AB GENES INVOLVED IN Bacillus thuringiensis VIRULENCE

There are several genes involved in Bacillus thuringiensis sporulation. The regulation and expression of these genes results in an upregulation in Cry protein production, and this is responsible for the death of insect larvae infected by Bacillus thuringiensis. Gene expression was monitored in Bacillus thuringiensis during three developmental phases. DNA macroarrays were constructed for selected genes whose sequences are available in the GenBank database. These genes were hybridized to cDNA sequences from B. thuringiensis var. kurstaki HD-1. cDNA probes were synthesized by reverse transcription from B. thuringiensis RNA templates extracted during the exponential (log) growth, stationary and sporulation phases, and labeled with PadCTP. Two genes were differentially expressed levels during the different developmental phases. One of these genes is related to sigma factor (sigma), and the other is a cry gene (cry2Ab). There were differences between the differential levels of expression of various genes and among the expression detected for different combinations of the sigma factor and cry2Ab genes. The maximum difference in expression was observed for the gene encoding sigma factor in the log phase, which was also expressed at a high level during the sporulation phase. The cry2Ab gene was only expressed at a high level in the log phase, but at very low levels in the other phases when compared to the sigma.


INTRODUCTION
Many genes are involved in sporulation mechanisms.It is estimated that about 800 genes are active in this phase in Bacillus subtilis, and this is in addi-tion to the pleiotropic effects of many genes and other cell functions that determine the production of enzymes and antimicrobial substances.Sporulation is the result of the successive actions of several genes that act in cascade (Balassa, 1969).During sporulation, the regulation of cry genes results in upregulation of the Cry proteins, which kill insect larvae (Lereclus et al., 2000).These proteins also allow the germination of bacterial spores and multiplication of bacterial cells in larval intestines.The production of extracellular factors in response to nutrient deficiency allow bacterial cells to wound and invade host tissues, accessing alternative nutrient sources and providing favorable conditions for a new multiplication or exponential growth cycle.
The expression of cry genes is regulated by two mechanisms.One of them, which regulates most of the cry genes, depends on specific sigma factors of the sporulation phase.The other mechanism does not depend on sporulation, and regulates factors that are typical of the stationary phase, e.g., cry 3 (Valadares-Inglis et al., 1998).
The selection of B. thuringiensis isolates with high virulence is a very important step for the development of biopesticides (Polanczyk & Alves, 2005).There have been some reports that characterized B. thuringiensis according to cry gene content (Bravo et al., 1998;Ferrandis et al., 1999;Uribe et al., 2003, Vilas-Bôas & Lemos, 2004), but the expression level of these genes, or even their product precursors, have still not been completely elucidated.Therefore, the search for the phases in which the cry genes and their precursors are expressed, as well as their expression levels in a standard strain such as HD-1, becomes of fundamental importance for the selection of other B. thuringiensis strains that may be used for insect control.
In this study, the differential expression of genes from B. thuringiensis kurstaki HD-1 related to virulence, sporulation and crystal protein formation was monitored in order to determine the genes expressed at the logarithmic, stationary and sporulation phases of the development cycle of this bacterium.

MATERIAL AND METHODS
Bacterial strain -The B. thuringiensis var.kurstaki HD-1 strain used was obtained from the 'Bacillus Genetic Stock Center' and is kept at the Laboratory of Bacterial Genetics and Applied Biotechnology, FCAV -UNESP/ Jaboticabal, São Paulo, Brazil.
Gene selection for DNA array production -For DNA array production, B. thuringiensis genes were selected, cloned, sequenced and annotated based on information from the GenBank database.The genes selected (Table 1) are related to virulence factors (expressed during the stationary phase), sporulation mechanisms and formation of protein crystals (expressed during the stationary and sporulation phases).DNA arrays were produced at the 'Brazilian Clone Collection Center' (BCCCenter).
Extraction and quantification of genomic DNA -B.thuringiensis var.kurstaki (HD1) cells were grown in a semi-solid medium (Nutrient Agar) (Gordon et al., 1973), from 'BIOBRÁS' and incubated at 30ºC for approximately 18 h.Colonies isolated from each line were inoculated in 50 mL of Brain Heart Infusion medium, also from 'BIOBRÁS' and were grown with agitation at 200 rpm at 30ºC for 4 h and 30 min.
Genomic DNA was extracted as described by Marmur (1961), and after ethanol precipitation, the DNA samples were resuspended in TE buffer (10 mM Tris; EDTA 1 mM, pH 8.0) and stored at -20ºC.DNA was quantified from samples of each strain using a spectrophotometer (BECKMAN, model DU-640B), and DNA quality was determined by electrophoresis in a 0.8% agarose gel with TBE buffer (89 mM Tris; 89 mM boric acid and 2.5 mM EDTA, pH 8.2), at 90 V for 90 min.The DNA samples were diluted to 10 ng µL -1 for analysis.
Design of oligonucleotide primers -Pairs of oligonucleotide primers were designed using the Gene Runner 3.0 software (Hastings Software, Inc.) and sequences obtained from GenBank (Table 1).The selected genes were amplified by PCR (DeRisi et al., 1997) using genomic DNA as a template, and the PCR products were designed to be used in macroarrays for the evaluation of gene expression at different developmental stages of B. thuringiensis var.kurstaki (HD1).Oligonucleotides were synthesized by MWG (The Genomic Company) and primers were used at a concentration of 1 pmol mL -1 in the PCRs.
The amplification products were then resuspended in 50% DMSO at concentrations of 2.5 µg µL -1 and 100 µg µL -1 , and transferred to a 384-well plate, then applied to nylon membranes at the Brazilian Clone Collection Center (BCCCenter).
Growth phases were defined through successive microscopic analyses of the cultures and by spectrophotometer readings at 600 nm.
The strain was cultured in Nutrient Agar (BIOBRÁS) (Gordon et al., 1973) and incubated at 30ºC for approximately 18 h.Colonies isolated from the strains were inoculated in 50 mL of nutritious broth medium (BIOBRÁS) and cultured with shaking at 200 rpm at 30ºC for 12 h (DO 0.4) to prepare the preinoculum.Inocula for the three growth phases were obtained by culturing 1 mL of the pre-inoculum in 50 mL of medium.Cells were cultured for 4h30 min (DO 0.1), 6h30 min (DO 0.2) and 9 h (DO 0.3) to obtain log phase, stationary phase and sporulation phase samples, respectively.
Samples were collected at each of the three phases of development and total RNA was immediately extracted.
Production of cDNA probes for macroarray hybridization -Complementary DNA was synthesized by reverse transcription of the RNA extracted from B. thuringiensis var.kurstaki HD-1 samples from different developmental phases.
Thirty micrograms of total RNA (in 6 µL) was added to 156 pmols of Random Primer Pd(N) 6 (Amersham Pharmacia Biotec Inc.).Samples were incubated at 75ºC for 10 min to allow primers to anneal, and then transferred to ice bath.Five microliters of First Strand Buffer (5X, BRL) (Invitrogen), 2.5 µL of DTT (100 mM), 2.0 µL of RNAguard (Invitrogen), 2.5 µL of ATG (10 mM each) and 5 µL of 33 P αdCTP (50 µCi) were then added.The mix was then heated to 42ºC and 1.25 µL of Superscript II (Invitrogen, 200 U µL -1 ) was added.Samples were gently stirred and incubated at 37ºC for 20 min.Then, 1.25 µL of dCTP (10 mM) was added to the mix and the mix was incubated at 37ºC for approximately 2 h.Samples were then denatured at 94ºC for 5 min and incubated at 37ºC for 15 min after adding 1.4 µL NaOH (5M).After incubation, 1.8 µL HCl (3.94M) and 7.0 µL of Tris-HCl (1M) pH 7.5 were added.The total volume of the probe was measured in a scintillator and purified using a G-50 column (Amersham Pharmacia Biotec Inc.).After purification, the probe volume was measured again.The incorporation efficiency of purified probe should range from 30 to 50%, compared to the unpurified version.

Macroarray hybridization
Pre-hybridization -Membranes were pre-hybridized according to a Southern protocol (Sambrook & Russel, 2001) for 4 h at 42ºC.Hybridization cylinders containing 10 ml of solution were used, and each cylinder contained only one membrane.

Hybridization -
The hybridization solution was prepared as recommended for a Southern protocol (Sambrook & Russel, 2001) and kept at 42ºC until use.After purification and denaturing, the probe was added to a 50 mL tube containing the hybridization solution for 42 h at 42ºC.Membrane washing -After hybridization, probes were discarded and membranes were washed under high stringency conditions (Desprez et al., 1998).Membranes were then sealed and exposed to a phosphoimaging screen for approximately 72 h.Detection and measurement of hybridization signals -Hybridization signals were obtained from 30 genes from the three development phases of this bacterium (treatments).Each gene was spotted three times onto each membrane, and two different membranes were used, with a completely randomized design, and therefore there were six repetitions of each treatment.
After the membranes were exposed, the phosphoimaging screens were analyzed with a Cyclone Phosphor-Imager (Packard Instruments), and hybridization signals were measured using the Optiquant software, and expressed in DLU (Digital Light Units).
The probe hybridization data were normalized by subtracting background-related values from regions of the image that did not correspond to DNA spots.The genes expression levels were subject to analysis of variance (ANOVA), being their means compared by a Tukey test (p ≤ 0.05).These analyses were performed with the SAS statistical program (SAS Institute, 1995).

RESULTS AND DISCUSSION
Gene expression analysis -PCR products were analyzed by electrophoresis in an agarose gel in order to determine their size and purity.This procedure is considered one of the most important steps in large-scale gene expression studies (Deyholos & Galbraith, 2001).It confirms that a unique amplicon is present and provides an estimate of the concentration of the amplified product.
Three developmental phases of B. thuringiensis kurstaki HD-1 were analyzed in order to investigate the expression of 30 genes.The sequences of these genes can be found in GenBank, which also includes recent citations of other sequences obtained by well-known researchers, allowing specific primers to be designed for the detection of these genes.The genes that were chosen for the present study play important biological roles in this bacterium and encode the Cry proteins.These proteins mediate highly specific toxic activities, allowing larvae from insect pest species to be controlled by this bacterium.The other genes evaluated in this study are involved in basic metabolic processes, and act to initiate the process that ultimately leads to cry1 gene expression.Our results indicate that two important genes related to physiological processes that take place before sporulation are differentially expressed at phases in bacterial development.
Two B. thuringiensis genes exhibited significant variations in their levels of expression (Figure 1); one gene was related to sigma factors (sigma 35 ), and the other was related to the cry gene (cry2Ab).The sigma factor is expressed during all three phases ex-amined, laying an important role in transcription.RNA polymerase begins transcription by recognizing a specific region of DNA, the promoter.Promoter recognition by RNA polymerase is mediated by a sigma factor, which attaches to the enzyme, and identifies promoter sites.The action of several types of promoters and sigma factors regulate cell functions, allowing a balance of physiological activities to be maintained.Therefore, the observation that sigma 35 factor is expressed at all the development stages in this study confirms its direct involvement in the cry1Aa promoter site recognition of B. thuringiensis, as reported by Adams et al. (1991).
The sigma 35 holoenzyme is also related to other cry genes, as observed by Tounsi & Jaoua (2002), who suggested that the transcription of a cry1Ia promoter in B. thuringiensis var.kurstaki is mediated by this holoenzyme.They described this enzyme as an RNA polymerase specific to the sporulation phase associated with the BtI promoter.However, the sigma 38 -dependent BtII is apparently not involved in cry1Ia transcription.Despite the fact that both promoters are active during the medium and late sporulation stages of B. thuringiensis var.kurstaki HD-1-Dipel, sigma 35 was expressed in the sporulation, vegetative and stationary phases in this study, but was expressed at lower levels in the stationary phases (Figure 1).
The high expression of the sigma 35 factor from the log phase onwards suggests that the activation of cry genes begins in this phase.This is because these genes are related to sporulation, which depends on the activation of several genes in cascade, including the sigma 35 factor.This hypothesis is also supported by the fact that transcription from the amyE promoter increased substantially as cell growth reached the log phase.The gene that encodes sigma 35 is a significant promoter, when one observes cry genes expression, indicating that the presence of other promoters might change the expression in each phase in which the gene is expressed.Expression of cry1Ac begins in the B. thuringiensis log phase (Chak et al., 1994).cry1Ac expression in B. thuringiensis kurstaki Tt14 is regulated by the amyE promoter, which is dependent on a specific sigma factor (sigma 43 ) (Yang et al., 2003).Other studies have shown that cry gene expression is dependent on sigma 28 or sigma 35 factors (Nicholson et al., 1987).
The cry2Ab gene was also expressed in the log phase in B. thuringiensis kurstaki HD-1, although at much lower levels than the gene encoding sigma 35 .Despite the fact that cry2Ab has been described as Diptera-specific, it has also been reported to have effects upon Lepidoptera (Schnepf et al., 1998).Its expression in the log phase, together with the gene encoding sigma 35 factor, suggests that this holoenzyme is not responsible for recognition of the cry2Ab promoter site.This explains why it was not possible to detect the expression of the cry1 genes at this stage of development.The sigma 35 factor is responsible for promoter site recognition; there would be no expression of these genes before expression of the gene encoding the sigma factor.In the log phase, the cry2Ab gene is transcribed using an alternative promoter unrelated to sigma 35 .This unknown alternative promoter may also activate cry2Ab expression at earlier developmental phases, before the cry1A genes are expressed.
There was a difference in the levels of expression of the sigma factor and cry2Ab (Tukey test, p < 0.01) (Tables 2 and 3).The highest mean of expression was observed in the log phase for the gene encoding sigma 35 , and the second highest expression was during the sporulation phase (Table 3).Cry2Ab was only expressed at significant levels in the log phase, but the expression of this gene was still low when compared with the sigma-encoding gene during all developmental phases examined (Tables 2 and 3).
The expression of cry2Ab is highly relevant even at low levels, since this gene has previously been studied using transgenic varieties of cotton, maize and tomato, which simultaneously express cry1Ac and cry2Ab, the so-called 'second generation' of commer- cial transgenic genotypes.A combination of cry genes is being used to minimize selection pressures and to avoid Bt-resistant insect pests.The use of cry2Ab allows monitoring programs and preventive resistance control measures to be implemented, because crossresistance with the cry2Ab and cry1Ac combination is much lower than when using the combination of cry1Ac and cry1Aa, as observed for the pink bollworm (Tabashnik et al., 2002) and cry1Ac and cry1Aa reported in Heliotis virescens with transgenic Bt maize (Fuentes et al., 2003).
In the log phase, only the gene encoding sigma 35 was expressed at high levels.This result contrasts to the results from a previous study that reported that both cry1Aa and sigma 35 are expressed in the log phase (Schnepf et al., 1998).However, this study reporting simultaneous expression of cry1Aa and sigma 35 used an in vivo fusion of the cry1Aa'-'lacZ genes, whereas gene expression was analyzed in the natural state without gene fusion in our study, which may explain the differences in the results between these two studies.
In addition to sigma 35 , we also attempted to analyze the expression of the gene that encodes sigma 28 using a specific set of primers.However, no sigma 28 expression was observed in any of the developmental phases, thus suggesting that sigma 28 is not involved in the recognition of the cry promoters studied here.
Based on the finding that no other genes were significantly expressed during the developmental stages studied in these experiments, we suggest that new studies should consider the development phases defined in this work and even divide them into shorter time periods in order to attempt to explain the lack of expression of genes that are supposedly related to sporulation and virulence factors in B. thuringiensis var.kurstaki HD1-Dipel.It is also important to mention that the absence of expression of other genes using macroarray techniques may be simply due to the lack of information about the time intervals between the activations of the cry genes.For example, in B. subtilis, seven stages have been defined during sporulation, and sporulation is mediated by activation of genes through a signaling cascade.About 800 genes are involved in sporulation in B. subtilis and other species from this genus (Mandelstam, 1969).

CONCLUSION
During the log developmental phase of the B. thuringiensis kurstaki-HD-1 bacterium, the genes encoding sigma 35 and cry2Ab were expressed at the highest levels.

Table 1 -
Oligonucleotide primers used in this study.

Table 2 -
Variance analysis (ANOVA) for expression genes (DLU) relative to exponential, logarithmic and stationary developmental phases.

Table 3 -
Comparison between sigma 35 and cry2AB gene expression means at exponential, logarithmic and stationary developmental phases.Means followed by different letters in the column are different (Tukey test, p < 0.01).