induced protection, but the latter lasted only 1–2 years. In addition, it appeared to be serovar-specific,
i.e., of the four C. trachomatis ocular isolates, the protection was effective only against the serovar
used in the vaccine. Furthermore, after re-exposure to Chlamydia, some of the immunized individuals
developed a hypersensitivity reaction. Although the cause of the hypersensitivity reaction has not yet
been elucidated, it is attributed to an antigenic component present in the whole Chlamydia cells used
for vaccination (Morrison et al. 1989). This led to exploring the option of developing a subunit vaccine.
A likely candidate for formulating such a vaccine was identified when the sequence of the
C. trachomatis major outer membrane protein (MOMP) was analyzed. This sequence indeed
comprises four variable domains (VDs) that are unique to each serovar (Stephens et al. 1987, 1998)
and, therefore, may account for the serovar-specific protection observed during the first vaccine trials
(de la Maza and Peterson 2002; Morrison and Caldwell 2002; Brunham and Rey-Ladino 2005; de la
Maza et al. 2017). MOMP belongs to a family of porins found in the outer membrane of Gram
À
bacteria, of which E. coli OmpF is a prototype, whose monomers have a molecular mass of $40 kDa
and which assemble into homotrimers (Nikaido 1992; Sun et al. 2007) (cf. Chap. 1, Fig. 1.9).
SDS-PAGE analyses of native MOMP (nMOMP) purified from C. muridarum using Zwittergent
3-14 (Z3-14) showed it to indeed be a homotrimer (Sun et al. 2007). A topological model of
C. muridarum MOMP (see below, Fig. 15.5A) proposes that each monomer comprises 16 antiparallel
β-strands that form a β-barrel structure spanning the outer membrane, while the VDs are exposed to the
cell surface (Rodriguez-Maranon et al. 2002). Unfortunately, attempts to elicit protection in several
animal models using recombinant MOMP (rMOMP), MOMP peptides, or DNA MOMP-based
vaccines yielded disappointing results (Su and Caldwell 1992; Pal et al. 1999; Dong-Ji et al. 2000),
suggesting that the native conformation of MOMP and/or posttranslational modification(s) is or are
necessary for eliciting an efficient protection. This led to attempts at vaccination using nMOMP.
A trimeric form of MOMP, likely to correspond to its native structure, was isolated in detergent
solution. Mice vaccinated with this preparation mounted a strong immune response that was protective
against genital and respiratory challenges (Pal et al. 2002, 2005; Sun et al. 2007, 2009; Farris et al.
2010). The same type of preparation was found to elicit protection in monkeys against an ocular
infection (Kari et al. 2009). Detergents, however, are far from ideal surfactants to formulate a vaccine,
because (i) they can have toxic effects (see e.g. Speijers et al. 1988, 1989; Blake and Wetzler 1995;
Castro et al. 1995) and (ii) they tend to destabilize MPs (Chap. 2). The latter phenomenon is
particularly to be avoided when the critical epitopes are conformational ones, as seems to be the
case for MOMP. This led to consider resorting to APols.
15.2.2 Developing an Amphipol-Based Vaccine
In Studies 15.2–15.4 (Table 15.1), Luis M. de la Maza and his colleagues examined (i) whether transfer
of nMOMP from Z3-14 to A8-35 would stabilize it and (ii) whether the efficiency of the vaccine would
improve if it was formulated with MOMP/A8-35 rather than MOMP/Z3-14 complexes. These two
expectations were vindicated (Tifrea et al. 2011, 2014; Feinstein et al. 2014).
The stability of the secondary structure of nMOMP in the presence of either Z3-14 or A8-35 was
examined by Melanie J. Cocco and her colleagues by monitoring changes in the far-UV CD spectrum
as a function of temperature (Fig. 15.2A). At 24
C, the CD spectra of the two types of samples were
indistinguishable. In keeping with earlier data (Cai et al. 2009), the CD spectrum of Z3-14-solubilized
nMOMP changed with temperature, indicating a loss of β-strands and the formation of α-helices, with
a transition around 52
C (Fig. 15.2B). In contrast, the secondary structure of A8-35-trapped nMOMP
was not visibly affected up to 78
C, the highest temperature tested (Fig. 15.2B). MP stabilization upon
664
15 Biomedical Applications
i.e., of the four C. trachomatis ocular isolates, the protection was effective only against the serovar
used in the vaccine. Furthermore, after re-exposure to Chlamydia, some of the immunized individuals
developed a hypersensitivity reaction. Although the cause of the hypersensitivity reaction has not yet
been elucidated, it is attributed to an antigenic component present in the whole Chlamydia cells used
for vaccination (Morrison et al. 1989). This led to exploring the option of developing a subunit vaccine.
A likely candidate for formulating such a vaccine was identified when the sequence of the
C. trachomatis major outer membrane protein (MOMP) was analyzed. This sequence indeed
comprises four variable domains (VDs) that are unique to each serovar (Stephens et al. 1987, 1998)
and, therefore, may account for the serovar-specific protection observed during the first vaccine trials
(de la Maza and Peterson 2002; Morrison and Caldwell 2002; Brunham and Rey-Ladino 2005; de la
Maza et al. 2017). MOMP belongs to a family of porins found in the outer membrane of Gram
À
bacteria, of which E. coli OmpF is a prototype, whose monomers have a molecular mass of $40 kDa
and which assemble into homotrimers (Nikaido 1992; Sun et al. 2007) (cf. Chap. 1, Fig. 1.9).
SDS-PAGE analyses of native MOMP (nMOMP) purified from C. muridarum using Zwittergent
3-14 (Z3-14) showed it to indeed be a homotrimer (Sun et al. 2007). A topological model of
C. muridarum MOMP (see below, Fig. 15.5A) proposes that each monomer comprises 16 antiparallel
β-strands that form a β-barrel structure spanning the outer membrane, while the VDs are exposed to the
cell surface (Rodriguez-Maranon et al. 2002). Unfortunately, attempts to elicit protection in several
animal models using recombinant MOMP (rMOMP), MOMP peptides, or DNA MOMP-based
vaccines yielded disappointing results (Su and Caldwell 1992; Pal et al. 1999; Dong-Ji et al. 2000),
suggesting that the native conformation of MOMP and/or posttranslational modification(s) is or are
necessary for eliciting an efficient protection. This led to attempts at vaccination using nMOMP.
A trimeric form of MOMP, likely to correspond to its native structure, was isolated in detergent
solution. Mice vaccinated with this preparation mounted a strong immune response that was protective
against genital and respiratory challenges (Pal et al. 2002, 2005; Sun et al. 2007, 2009; Farris et al.
2010). The same type of preparation was found to elicit protection in monkeys against an ocular
infection (Kari et al. 2009). Detergents, however, are far from ideal surfactants to formulate a vaccine,
because (i) they can have toxic effects (see e.g. Speijers et al. 1988, 1989; Blake and Wetzler 1995;
Castro et al. 1995) and (ii) they tend to destabilize MPs (Chap. 2). The latter phenomenon is
particularly to be avoided when the critical epitopes are conformational ones, as seems to be the
case for MOMP. This led to consider resorting to APols.
15.2.2 Developing an Amphipol-Based Vaccine
In Studies 15.2–15.4 (Table 15.1), Luis M. de la Maza and his colleagues examined (i) whether transfer
of nMOMP from Z3-14 to A8-35 would stabilize it and (ii) whether the efficiency of the vaccine would
improve if it was formulated with MOMP/A8-35 rather than MOMP/Z3-14 complexes. These two
expectations were vindicated (Tifrea et al. 2011, 2014; Feinstein et al. 2014).
The stability of the secondary structure of nMOMP in the presence of either Z3-14 or A8-35 was
examined by Melanie J. Cocco and her colleagues by monitoring changes in the far-UV CD spectrum
as a function of temperature (Fig. 15.2A). At 24
C, the CD spectra of the two types of samples were
indistinguishable. In keeping with earlier data (Cai et al. 2009), the CD spectrum of Z3-14-solubilized
nMOMP changed with temperature, indicating a loss of β-strands and the formation of α-helices, with
a transition around 52
C (Fig. 15.2B). In contrast, the secondary structure of A8-35-trapped nMOMP
was not visibly affected up to 78
C, the highest temperature tested (Fig. 15.2B). MP stabilization upon
664
15 Biomedical Applications