Parasporal Body Formation via Overexpression of the Cry10Aa Toxin of Bacillus thuringiensis subsp. israelensis,and Cry10Aa-Cyt1Aa Synergism

Bacillus thuringiensis subsp. israelensis is the most widely used microbial control agent against mosquitoes and blackflies. Its insecticidal success is based on an arsenal of toxins, such as Cry4A, Cry4B, Cry11A, and Cyt1A, harbored in the parasporal crystal of the bacterium. A fifth toxin, Cry10Aa, is synthesized at very low levels; previous attempts to clone and express Cry10Aa were limited, and no parasporal body was formed. By using a new strategy, the whole Cry10A operon was cloned in the pSTAB vector, where both open reading frames ORF1 and ORF2 (and the gap between the two) were located, under the control of the cyt1A operon and the STAB-SD stabilizer sequence characteristic of this vector. Once the acrystalliferous mutant 4Q7 of B. thuringiensis subsp. israelensis was transformed with this construct, parasporal bodies were observed by phase-contrast microscopy and transmission electron microscopy. Discrete, ca. 0.9-μm amorphous parasporal bodies were observed in the mature sporangia, which were readily purified by gradient centrifugation once autolysis had occurred. Pure parasporal bodies showed two major bands of ca. 68 and 56 kDa on sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis. These bands were further characterized by N-terminal sequencing of tryptic fragments using matrix-assisted laser desorption ionization-time of flight mass spectrometry analysis, which identified both bands as the products of ORF1 and ORF2, respectively. Bioassays against fourth-instar larvae of Aedes aegypti of spore-crystal complex and pure crystals of Cry10Aa gave estimated 50% lethal concentrations of 2,061 ng/ml and 239 ng/ml, respectively. Additionally, synergism was clearly detected between Cry10A and Cyt1A, as the synergistic levels (potentiation rates) were estimated at 13.3 for the mixture of Cyt1A crystals and Cry10Aa spore-crystal complex and 12.6 for the combination of Cyt1A and Cry10Aa pure crystals.The subspecies Bacillus thuringiensis subsp. israelensis (serotype H-14) was discovered by Goldberg and Margalit in 1977 (11). To date, its insecticidal potential has not been overcome by any other bacterium (or any biological control agent) as an effective control measure against mosquito and blackfly larvae (8). Recently, its toxicity spectrum has been expanded to a coleopteran pest, the coffee berry borer (Hypothenemus hampei) (23), indicating that this strain may have potential versatility. Also, the so-called pBtoxis megaplasmid harbored in this strain, containing all the endotoxin-encoding genes found in its parasporal crystal, including cry4A, cry4B, cry10A, cry11A, and cyt1A, was recently sequenced (1). Among many other interesting aspects of this serotype, the occurrence of this mosquitocidal arsenal in one strain and their synergistic interaction make this bacterium scientifically and technologically attractive.The parasporal crystal of B. thuringiensis subsp. israelensis contains large amounts of Cry4A, Cry4B, Cry11A, and Cyt1A toxins (14), and consequently, most of the knowledge about the toxicity of this strain has been focused on these proteins, acting either as a complex (31) or tested separately (6). Although the cry10Aa gene was originally cloned in 1986 (known then as cryIVC) (30), to date, little is known about cry10Aa and the protein it encodes, mostly due to its very low levels of expression (10) in B. thuringiensis subsp. israelensis. Interestingly, cry10Aa is an operon as it includes two open reading frames (ORFs), previously reported as pBt047 and pBt048 (hereafter referred to only as ORF1 and ORF2, respectively), separated by a 48-bp untranslated gap (1). ORF1 contains the complete δ-endotoxin sequence (active toxin), with a coding capacity for a 78-kDa protein. Interestingly, ORF2 shows high identity with the coding sequence of the C-terminal half of Cry4-type proteins, with a coding capacity for a 56-kDa protein. Therefore, it is believed that a putative ancestral cry10Aa gene is similar in size to the cry4-type genes (ca. 4 kbp), but either a small sequence had been inserted in the middle of the coding sequence or site mutations produced end codons (two end codons flank the gap) in this region (1).Previous attempts to clone and express the cry10Aa gene included ORF1 and only part of ORF2 (7, 10, 30). This was a reasonable strategy, as most of the so-called “complete” protoxins are partially digested to become active toxins (δ-endotoxins) (28), and ORF1 included the complete sequence to code the Cry10Aa δ-endotoxin. However, in all these cases, the expression levels were very low, and no parasporal body was formed. Similar results were obtained when the promoter was changed and a stabilizing sequence was added to the construction (13). The low expression levels achieved in these cases led to conclusions that assumed low toxic levels of Cry10Aa when tested against mosquito larvae (30). In spite of the low toxicity of Cry10Aa found against mosquito larvae, a synergistic effect was reported between Cry10Aa and Cry4Ba toxins in Culex (7). Obtaining high levels of expression and crystallization of Cry10Aa are required to properly characterize and understand the toxic spectrum of this protein.In this report, we show the formation of parasporal bodies of Cry10Aa, achieved by cloning the whole Cry10Aa operon under the control of the cyt1A promoter and the STAB-SD sequence. We also show that Cry10Aa is as toxic as most of the other B. thuringiensis subsp. israelensis toxins acting separately, and in synergism with the Cyt1A toxin.