Pasteur revisited: An unexpected finding in Bacillus anthracis vaccine strains

KEYWORDS:

• anthrax
• immunity
• plasmids
• toxins
• vaccine

This article refers to:

Identification of the pXO1 plasmid in attenuated Bacillus anthracis vaccine strains

Bacillus anthracis is a spore-forming, gram-positive bacterium that is the causative agent of anthrax. This bacterium is normally found in environmental samples such as soil. B. anthracis causes 3 forms of anthrax, cutaneous, gastrointestinal and inhalation-derived infections.¹ Anthrax has been well characterized throughout history; documented cases date back prior to 450 BCE.² Vaccine development for anthrax began in the 1880s with attenuated strains developed by Pasteur in France and Greenfield in London.³ From there, many vaccine formulations have been developed with the aim of improving efficacy while decreasing side effects in vaccine recipients. The current leading vaccines are alum-adsorbed preparations from non-encapsulated attenuated bacteria, BioThrax in the United States or the similar AVP vaccine in the United Kingdom. These vaccines provide protection from all forms of anthrax but require many vaccine boosters (e.g. annual re-immunization is recommended in the US). Live attenuated B. anthracis vaccines are no longer used for human immunizations.^1,4 Attenuated strains of B. anthracis have been shown to provide immunological protection from virulent strains, but only if the bacterial toxins are present.

The main virulence factors of B. anthracis, the anthrax toxins and the bacterial capsule, are encoded on genes located in 2 extrachromasomal plasmids, termed pXO1 and pXO2. Plasmid-encoded virulence factors are commonly found in spore-forming bacteria, such as some other Bacillus species and certain types of pathogenic Clostridium.⁵ The relatively large plasmids of B. anthracis are fairly well characterized. The pXO1 plasmid is approximately 182 kb in size and carries the genes for lethal factor (LF), edema factor (EF), and protective antigen (PA) at 3 discrete locations.⁶ Once the resulting polypeptides are expressed, they combine to form 2 binary toxins known as lethal toxin and edema toxin, which are responsible for the extensive tissue destruction in anthrax.¹ In addition to its role in toxin formation, PA is also recognized as a potent immunogen, critical to the development of a protective immune response.³ The pXO2 plasmid is roughly 96 kb and carries the genes responsible for expression and formation of the poly-γ-D-glutamic acid (PGA) capsule.⁶ The unique capsule of B. anthracis is important for bacterial survival in the host, and is emerging as a critical biomarker of infection.^7,8 However, the presence of the capsule alone is not sufficient to elicit a neutralizing immune response.⁹

In this issue of Virulence, Xudong Liang and colleagues¹⁰ describe the identification of pXO1 in attenuated strains of the bacterium previously thought to be devoid of the plasmid. The pXO1 plasmid was believed to be lost in these strains following incubation of cultures at 42-43^OC for 10-20 days, a method of curing originally described by Pasteur.^11,12 In addition, they show that PA (a key protein encoded on pXO1) is expressed in each of the vaccine strains that were tested (Pasteur II from France, Qiankefusiji II from Russia, and Rentian II from Japan). This is surprising, as these strains of B. anthracis are considered safe due to the presumed lack of plasmid-encoded toxin components. Although not specifically addressed in the current study, it would be of interest to know whether or not the other toxin sub-components (LF and/or EF) are expressed at low but detectable levels in attenuated strains of the bacterium. One previous study¹² examined 2 different Pasteur-type vaccine strains for evidence of pXO1 plasmid DNA or expression of toxins. Neither was detected in that study, lending support to the idea that the bacteria had been cured of the plasmid by the Pasteur method. However, as the authors of the current paper have pointed out, this was in an era prior to the advent and refinement of more modern molecular biology techniques, and the resulting increase in detection sensitivity. When the authors of the current study attempted to eliminate pXO1 by high temperature treatment, the results clearly showed that the plasmid DNA remained, although at a reduced level than before, indicating that temperature stress may reduce total copy number of pXO1, but not completely eradicate the plasmid.

The authors point out that the presence of pXO1 at low copy levels in weakened strains may cause the bacterium to retain the ability to generate a fairly robust immune response. The authors also note the possibility of some risk involved in using these strains in the field as live (attenuated) vaccines. Although the use of these types of vaccines for anthrax in humans have fallen out of favor in most of the world, safety considerations are still important (e.g., in veterinary use). This study underscores the need for continued development of next generation anthrax vaccines that rely on well-characterized, purified B. anthracis antigens (including recombinant PA [rPA] and rPA-PGA conjugates)¹³ for use in humans and livestock. As recent events in the US have illustrated,^14,15 B. anthracis and anthrax remain as important topics for our consideration, much as they have been for many centuries.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.