Diphtheria, the classical disease caused by C. diphtheriae is an acute communicable infection of the upper respiratory tract that can be fatal [1,2]. Although the global incidence of diphtheria has been reduced dramatically as a result of increased vaccine coverage, infections by different biotypes of C. diphtheriae continue to be reported in many countries [3,4]. The diphtheria toxin (DT) is a polypeptide with a molecular weight of about 58,000 Daltons that is cleaved into two fragments to become active. The fragment A (DTA) with 193 amino acids is an enzyme that catalytically inactivates the ribosomal elongation factor-2 of eukaryotic cells while the 342 amino acids long fragment B (DTB) recognizes and binds to specific receptors on the surface of most eukaryotic cells, facilitating the translocation of the fragment A across the membrane [5]. Despite the existence of a very good toxoid vaccine, diphtheria has not been eradicated yet. On the other hand, the current toxoid based vaccine presents risks to vaccinees associated to the production process. These risks include incomplete inactivation and the presence of reactor and unknown proteins generated during the inactivation process. For this reason the immunization against diphtheria has to be done with three doses to minimize adverse effects that could take place with a full single dose [6,7]. Adverse side reactions of that kind could be easily avoided by purification of the current vaccine. However, the purification of the diphtheria toxoid would increase the cost of production considerably.

The Bacillus of Calmette-Guérin (BCG), a live attenuated substrain of M. bovis obtained almost a century ago, is currently the only vaccine licensed against TB world wide. The recombinant BCG approach developed on top of it emerged in the late 80’s as a creative technology for in vivo delivery of protective antigens and brought a new hope for the development of low cost multivalent vaccines [8,9]. In this context, a genetically modified BCG strain expressing a relevant diphtheria antigen could be a safer alternative to the current classic vaccine used against diphtheria in places where BCG is mandatory. Nevertheless, the stability of the expression of heterologous antigens in BCG is still a major issue and perhaps one of the most important obstacles to the use of this technology in vaccine development. In fact there are several reports in the literature showing that genetic rearrangements lead to the disruption of the expression of heterologous genes or promoters in mycobacteria [10-14]. In this paper, we report our first results over the expression of the C. diphtheriae gene dtb_PW8 by the M. bovis BCG substrain Moreau used in Brazil to produce the vaccine against TB, modified by the transformation with distinct plasmid vectors, pUS973 and pUS977, to generate two new recombinant strains in which the expression of the target gene is driven by a strong (hsp60) or a weak (P_AN) mycobacterium promoter. A comparison of the structural stability of plasmid constructs used to transform BCGs with constructs rescued after transformation and growth of the two rBCGs in culture medium are presented.

A 1051 bp long DNA fragment, named dtb_PW8, necessary to build the expression vectors used in this study to modify BCG was successfully amplified by PCR from the genome of the C. diphtheriae PW8 vaccine strain, using specific primers designed to do it as referred above. The identity of the dtb_PW8 gene was confirmed by DNA sequencing and analysis of similarity against the sequence of DT deposited in the GenBank. The gene dtb was further cloned into the extrachromossomal mycobacterial expression vectors pUS973 and pUS977 generating constructs pUS973dtb_PW8 and pUS977dtb_PW8, respectively, and these were used to transform BCG. The two new rBCG strains resulting from transformation were evaluated in terms of their capacity to express rDTB_{PW8 in vitro}. The only difference between vectors pUS973 and pUS977 is the promoter sequence used to drive the expression of the foreign gene. The pUS973 vector carries the strong constitutive hsp60 promoter from M. tuberculosis, while pUS977 carries the P_AN promoter from M. paratuberculosis. The expression of DTB in rBCG was evaluated by immunoblotting assays using polyclonal antibodies against the diphtheria toxoid. As predicted, a band of the expected size for DTB, 40 kDa, was recognized in lysates of rBCGpUS973dtb_PW8 or rBCGpUS977dtb_PW8 subjected to immunoblot analysis. The level of expression of rDTB_PW8 observed in BCG transformed with pUS973dtb_PW8 was similar to BCG transformed with pUS977dtb_PW8 (Figures 1(a) and (b)). Mycobacteria and corynebacteria have a common ancestry and may as well share antigens. This fact possibly contributed to the fact that the use of polyclonal antibodies in the immunoblotting assays led to the visualization of non-specific rest bands in both non-transformed and transformed BCG. In order to verify the structural stability of the recombinant plasmids, selected BCG transformants were subjected to two successive passages in 7H9 Middlelbrook medium supplemented with kanamycin. Subsequently, plasmid DNA was recovered by electroduction into E. coli [20]. Several colonies were selected and the purified plasmid DNA was digested with XbaI. Electrophoresis of the digested DNA revealed that most BCG clones transformed with pUS973dtb_PW8 had undergone mutations (Figure 2(a)), whereas none of the pUS977dtb_PW8 clones showed alteration in the structure of the plasmid (Figure 2(b)). In fact, a few constructs from rBCGpUS973dtb_PW8 (Figure 2(a), lanes 2-5) lost the XbaI site and part of the plasmid DNA since they exhibited a lower molecular mass when compared to the original vector (lanes 8-10) used as control. Also, other constructs recovered from clones of the same rBCGpUS973dtb_PW8 strain apparently lost part of their

DNA in spite of having released the insert (lanes 6 and 7). Indeed, the instability of constructs based on the hsp60 promoter is not new. It has been shown that mutations may occur as a result of modifications of the promoter or insert sequences [13]. What is new is that it also occurs with the dtb_PW8 gene from C. diphththeriae.

The genetic modification of the BCG vaccine as an approach to the development of new multivalent vaccines has been actively pursued ever since it was proposed in 1989 [8] and has been the focus of our research over the last years [21-28]. BCG is a powerful candidate to a multivalent vaccine platform. We successfully constructed a new and stable rBCG strain able to express the non-toxic fragment B of the C. diphthteria toxin by genetically modifying the Moreau substrain of BCG used in the production of the vaccine against TB in Brazil. Although a mutant of the DT, CRM197, has been expressed in BCG using the promoter pBlaf [29], this is the first clear demonstration that the safer B fragment of DT can be expressed and well tolerated by BCG. These are definitive conditions for the development of vaccine prototypes based on this technology since not every antigen is necessarily expressed or tolerated by BCG. Our best results came out with the BCG transformed with the pUS977dtb_PW8 construct in which the expression of DTB is driven by the P_AN promoter, a regulatory sequence derived from M. paratuberculosis. The P_AN sequence, a weak mycobacterial promoter [18], has been successfully used by others for the expression of the α-galactosidade of E. coli [30], the LacZ of E. coli [31], the Gp63 surface antigen of Leishmania [32], the Nef antigen of Simian Immunodeficiency Virus [33], the Nef, Gag, Env antigens of Simian Immunodeficiency Virus [34], the S1 subunit of the Bordetella pertussis toxin [24], the hepatitis B surface antigens [35] and the LipL32 antigen of Leptospira interrogans [28]. In our study all rBCGpUS977dtb_PW8 clones stably expressed significant amounts of DTB while rBCGpUS973dtb_PW8 clones were unstable and, consequently, unsuitable for DTB expression. Both hsp60 and P_AN are constitutive mycobacterial promoters and have been widely used in recombinant BCG studies [18, 35,36]. The reason for the instability of rBCGs transformed with the pUS973 construct designed with the stronger hsp60 promoter is not yet clear and was not the objective of this study. However, it is known that the expression of heterologous antigens in recombinant BCG essentially imposes a metabolic burden with an unpredictable outcome and also, although not frequent in mycobacteria, the modification of the hsp60 promoter may result from a recombination event between host and plasmid sequences [11]. Consequently, the foreign gene expression driven by the hsp60 promoter is not always feasible in BCG [11,37]. The hsp60 promoter has a much stronger activity when compared to the P_AN promoter and this may result in a more stressful metabolic scenario for the host cell, which in turn may select point mutations or deletions in the hsp60 promoter region disrupting the expression of the heterologous protein. In fact, there may be a lethal level of synthesis of the foreign protein followed by selection of mutants with deletion in the promoter region immediately after induction of the hsp60 promoter in the transformed host [13]. Based on the preliminary data presented here we believe that the recombinant BCG expressing the rDTB_PW8 polypeptide under the P_AN promoter can be further exploited for the development of a vaccine prototype as an alternative for the current vaccine against diphtheria. We showed that P_AN promoter was expressed continuously in vitro during intracellular growth of M. bovis BCG, and basically all rBCGpUS977dtb_PW8 clones stably expressed the nontoxic fragment B of the diphtheria toxin. On the other hand, the pUS973, a similar expression vector based on the use of a heat shock protein promoter (hsp60) was highly unstable and inadequate for expression of the dtb_PW8 gene.

This study was carried out in partial fulfillment of the requirements of a PhD thesis for Dilzamar V. Nascimento, Faculdade de Ciências Médicas (PGCM), Universidade do Estado do Rio de Janeiro, Rio de Janeiro, RJ, Brazil. Research supported by Bio-Manguinhos/FIOCRUZ, PAPES II/FIOCRUZ, FAPERJ, CNPq, CAPES, Programa de Núcleo de Excelência (PRONEX/MCT/CNPq).