Adjuvant properties of a simplified C32 monomycolyl glycerol analogue

Article abstract of DOI:10.1016/j.bmcl.2009.02.027

A simplified C₃₂ monomycolyl glycerol (MMG) analogue demonstrated enhanced immunostimulatory activity in a dioctadecyl ammonium bromide (DDA)/Ag85B-ESAT-6 formulation. Elevated levels of IFN-γ and IL-6 were produced in spleen cells from mice immunised with a C₃₂ MMG analogue comparable activity to the potent Th1 adjuvant, trehalose 6,6′-di-behenate (TDB).

Full text of DOI:10.1016/j.bmcl.2009.02.027

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2029–2032
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Adjuvant properties of a simplified C32 monomycolyl glycerol analogue
Veemal Bhowruth a, David E. Minnikin a, Else Marie Agger b, Peter Andersen b, Vincent W. Bramwell c,
Yvonne Perrie c, Gurdyal S. Besra a,
*
a School of Biosciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
b Department of Infectious Disease Immunology, Adjuvant Research, Statens Serum Institut, Copenhagen 2300, Denmark
c School of Life and Health Sciences Aston University, Aston Triangle, Birmingham B4 7ET, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A simplified C32 monomycolyl glycerol (MMG) analogue demonstrated enhanced immunostimulatory
activity in a dioctadecyl ammonium bromide (DDA)/Ag85B-ESAT-6 formulation. Elevated levels of IFN-
Received 19 December 2008
Revised 5 February 2009
Accepted 6 February 2009
Available online 11 February 2009
c
and IL-6 were produced in spleen cells from mice immunised with a C32 MMG analogue comparable
activity to the potent Th1 adjuvant, trehalose 6,60-di-behenate (TDB).
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Monomycolyl glycerol
Adjuvant
Vaccine
Tuberculosis
Ag85B-ESAT-6
Tuberculosis (TB) in man has for many years been combated by
vaccination with an attenuated strain of Mycobacterium bovis—the
M. bovis BCG vaccine developed more than 80 years ago. However,
although more than 3 billion doses of BCG have been administered,
it does not always provide satisfactory resistance in every popula-
tion. Modern approaches use well-defined vaccine formulations to
minimize side-reactions.1 However, in order to induce a sufficient
protective immune response, adjuvant potentiating agents may be
required. An adjuvant is any substance that serves to direct, accel-
erate, prolong and/or enhance the specific immune response of a
vaccine. The development of a specific kind of immune response
(humoral or cell-mediated) can be determined by the choice of
adjuvant. For example, protective immunity against intracellular
pathogens like Mycobacterium tuberculosis requires a cell-mediated
immune response, and a suitable adjuvant for a subunit vaccine di-
rected against TB, should enhance a Th1 response.2
analogues thereof) has been included in various adjuvant formula-
tions.6–8 Rosenkrands et al.9 investigated the immunostimulatory
capacity of a total lipid extract of M. bovis BCG Copenhagen. The
tuberculosis subunit vaccine candidate, Ag85B-ESAT-6 (H1) was
administrated in combination with BCG lipids in cationic lipo-
somes formed of either dimethyl dioctadecyl ammonium bromide
(DDA), N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethyl ammonium
chloride (DOTAP) or cholesteryl 3b-N-(dimethylaminoethyl) carba-
mate hydrochloride (DC-Chol); neutral liposomes were formed by
1,2-dioleoyl-sn-glycerol-3-phosphoethanolamine (DOPE) DOPE-
phosphatidylcholine (PC) and anionic liposomes formed by
DOPE-PC-phosphatidylglycerol (PG). Immune responses were
monitored by in vitro restimulation of peripheral blood mononu-
clear cells purified 1 week after the last immunisation. Rosenk-
rands et al.9 showed that apolar BCG lipids, together with DDA,
elicited the most pronounced levels of IFN-
c released. DDA-BCG
Mycobacteria have long been known to exert a number of
immunomodulatory effects and are a good source of adjuvant
preparations. Freund’s complete adjuvant, consisting of a paraffin
oil emulsion and heat killed mycobacteria,3 is best-known, but live
M. bovis BCG has also been used as an immunotherapeutic agent.4,5
Similarly, purified components of mycobacteria have been shown
to have immunostimulatory activity. Wax D (a complex of peptido-
glycan, arabinogalactan and mycolic acids) was found to possess
strong adjuvant activity and trehalose 6,6-dimycolate (or synthetic
lipids only gave minimal levels of IL-5. Analysis of antigen-specific
antibodies demonstrated an efficient induction of both IgG1 and
IgG2a antibodies by the combinations of DDA-BCG lipids. The
IgG1 titre was 2.6-fold higher than that when the antigen was
administrated in alum. The ratio of IgG1:IgG2a was increased with
the DDA-BCG lipids compared to the alum control.
Further studies have shown that in subcutaneous immunisation
of mice with 10 lg of the total, polar or apolar lipids incorporated
into DDA liposomes and administrated in conjunction with the
recombinant fusion protein Ag85B-ESAT-6 (H1), the apolar fraction
was found to induce IFN-
c release of almost four times that
* Corresponding author. Tel.: +44 0121 4158125.
E-mail address: g.s.besra@bham.ac.uk (G.S. Besra).
recorded with polar lipids.10 The apolar components were
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.02.027

2030
V. Bhowruth et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2029–2032
separated according to the procedures outlined by Dobson et al.,11
to give four principal lipid fractions, phthiocerol dimycocerosates
(PDIMs), triacylglycerols (TAGs), phenolic glycolipids (PGLs) and
C70–90 monomycolyl glycerol (C70–90 MMG). C70–90 MMG induced
was achieved using dicyclohexylcarbodiimide (DCC) and 4-pyrroli-
dinopyridine (PYP).
The 1,2-isopropylidiene-protected glycerol (4a–d) preferen-
tially coupled with the carboxyl terminus of the unprotected my-
colic acid, however a doubly coupled mycolate with a glycerol
unit was identified as a minor side-product (<5%) by mass spec-
trometry. The TLC Rf values of the product and side-product were
0.20 and 0.40, respectively (hexane/ethyl acetate 95:5) and were
separated by column chromatography. Deprotection of the isopro-
pylidiene was achieved with trifluoroacetic acid (TFA)/tetrahydro-
furan (THF)/water (8:17:3) generating erythro and threo C32 sn and
rac glycerol coupled products (5a–d). NMR and mass spectral anal-
ysis of 5a is outlined in Andersen et al.10
extremely high levels of all three cytokines IL-12, TNF-
a and IL-6
with values of >250 pg/ml, 3500 pg/ml and 5000 pg/ml, respec-
tively. Herein, we report the synthesis of four structural isomers
of a simplified C32 MMG analogue and immunological evaluation.
In order to generate a series of MMG analogues, the core structures
were broken down into two entities, the lipid tail and the glycerol
unit. The rac or sn isopropylidiene-protected glycerol units were
purchased from Sigma Aldrich. The C32 corynomycolate isomers
were synthesised according to Datta et al.12 Briefly, the C32 coryno-
mycolic acid was synthesised by sodium hydride promoted self-
Claisen condensation of methyl palmitate (1) followed by sodium
borohydride reduction of the resulting keto ester to give the diaste-
reomeric hydroxyesters (2a) and (2b) (Scheme 1). Flash column
chromatography of the hydroxyesters mixture gave first 2b, fol-
lowed by 2a (TLC Rf 0.55 and 0.40 in chloroform, respectively).
Hydrolysis with 5% KOH in 1:1 biphasic butanol:water released
the racemic corynomycolic acid (3a) and its diastereoisomer (3b).
Coupling of the glycerol unit with the mycolic acids (3a and 3b)
Ten groups of three BALB/c mice received three doses of 2
lg H1
with 250 g DDA and 50
l
l
g of trehalose 6,60-di-behenate (TDB), a
potent Th1 adjuvant or quantities of lipid adjuvant as described in
Table 1 at days 0, 14 and 28.13 Blood samples were taken via the
tail vein at regular intervals to monitor H1 specific antibody and
antibody isotype titres using standard enzyme linked immunosor-
bent assay (ELISA) techniques. The groups were as outlined in
Table 1. Analysis of H1 specific cell proliferation was carried out
using 100 ll volumes of spleen cells from individual mice as
CH3(CH 2
CH3(CH 2
)
)
OH
CH3(CH 2
)
OH
O
(a)
14
14
13
CH3(CH2
)
14
+
O
O
OMe
CH3(CH 2
)
13
OMe
(2a)
OMe
(2b)
(1)
(b)
(b)
CH3(CH2
CH3(CH2
)
OH
CH3(CH2
CH3(CH2
)
OH
OH
14
14
13
+
O
O
)
)
13
OH
(3a)
(3b)
(c)
(c)
(c)
(c)
CH3(CH 2
CH3(CH 2
)
)
OH
O
14
CH3(CH 2
CH3(CH 2
)
)
OH
O
CH3(CH 2
CH3(CH 2
)
)
OH
O
CH3(CH 2
)
)
OH
14
13
14
13
14
13
O
O
H
O
O
13
CH3(CH 2
O
H
O
O
O
O
H
H
O
O
H
O
O
H
H
H
H
H
H
H
(4a)
(4b)
(4c)
(4d)
(d)
(d)
(d)
(d)
CH3(CH2
CH3(CH 2
)
)
OH
O
CH3(CH 2
CH3(CH 2
)
)
OH
CH3(CH 2
CH3(CH 2
)
)
OH
O
CH3(CH2
CH3(CH2
)
)
OH
O
14
13
14
13
14
13
14
13
O
H
O
H
O
H
O
H
O
H
H
OH
OH
H
OH
OH
H
H
OH
OH
H
H
OH
OH
H
(5a)
(5c)
(5b)
(5d)
Scheme 1. Reagents and conditions: (a) (i) NaH, dry xylene, reflux, 5 h (82 %); (ii) NaBH4, 3 h, rt (95 %); (b) 5 % KOH, 1:1 butanol/water, reflux (90%); (c) DCC, PYP, L-a-b-
isopropylidene glycerol (45 %); (d) 8:17:3 (TFA/THF/H2O), rt, overnight (49%).

V. Bhowruth et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2029–2032
2031
and IL-2 using DuoSetÒ capture ELISA kits (R&D Systems Europe,
Oxfordshire, UK) according to the manufacturer’s instructions.
From the adjuvants tested, it could be hypothesized that the syn-
thetic analogue that corresponds with the natural MMG might be
of most interest in mimicking the adjuvant effect of this mycobac-
terial agent. Interestingly, it could be seen that this synthetic agent
was more effective than the optimized dose of the TDB adjuvant in
the elicitation of H1 specific spleen cell proliferation following
restimulation of spleen cells from immunised mice with antigen
in vitro (Fig. 1). However, this was only seen when the dose of
the adjuvant was 100-fold reduced in comparison to the dose of
Table 1
Dosage groups used to analyse C32 mycolic acid variants
Group
Dose
1
2
3
4
5
6
7
8
9
MMG [50]
5aa [50]
5ca [50]
5ba [50]b
5da [50]
5aa [5]b
5aa [0.5]b
Incomplete Freund’s (ICF)(without DDA)
TDB [50]
Naïve control
10
TDB. The dose of TDB has previously been optimized to 50
250 g DDA in order to provide the best adjuvant activity. It is clear
from these results that from the adjuvants tested, the most effec-
tive dose or concentration of the MMG analogue is 0.5 g with
250 g DDA. H1 specific cytokine production showed that compar-
atively more IFN- and IL-6 was produced by spleen cells from
lg in
l
Key: [microgrammes per dose of lipid adjuvant].
a
GLYCEROL C32 (5a–d).
Expected immunoactive agent.
b
l
l
previously described.14 These were seeded onto 100
media in sterile 96-well suspension culture plates containing var-
ll volumes of
c
mice immunised with the low dose synthetic adjuvant correspond-
ing to the natural MMG structure (Fig. 2).
ious concentrations of H1 antigen. After 72 h incubation, 0.5 Ci of
l
[3H]thymidine (Amersham, UK) was added to each of the microcul-
tures, and the incubation continued for a further 24 h. The well
contents were harvested onto glass filter mats using a cell har-
vester (Titertek). The discs representing each well were punched
from the filter mats into 5 ml volumes of scintillation fluid (Opti-
phase Hisafe, Fisher, UK) to measure incorporation of [3H]thymi-
dine into the cultured cells using standard scintillation counting
procedures. Concomitant H1 specific cytokine production was ana-
lysed by incubation of similar spleen cell microcultures with a con-
From the other cytokines assessed, most groups generated ele-
vated IL-2 indicative of a T-cell driven component to the adjuvant ef-
fect of these analogues (data not shown). Figure 3 shows that H1
specific IgG2a titres were also high for the low dose analogue. All
of the formulated analogues were able to engender enhanced im-
mune responses at the 50 lg dose (Figs. 1 and 3) and these were
comparable to the potent Th1 adjuvant TDB. This observation was
corroborated by isotype specific antibody responses detected in
blood samples taken throughout this study (IgG1 and total IgG not
shown). The immunogenicity studies in mice showed that a glycerol
based non-polar mycobacterial lipid analogue adjuvant formulated
in DDA vesicles can give IgG2a mediated anti H1 antibody titres at
least as high as DDA/TDB with 100-fold less of the adjuvant moiety
present compared to TDB (P = 0.867 after the third dose).
The IgG2a antibody titre delineates a Th1 type response in
BALB/c mice, thought to correlate strongly with protective immune
responses. The BALB/c model is usually predisposed to Th2 type
responses and therefore the ability to engender a Th1 type
response here, supported by the antibody subtype and cytokine
data, is worthy of note. Whilst it remains to be seen whether
alternative synthetic analogues behave in the same way as 5a that
is representative of the mycobacterial MMG in their activity at very
low dose, the data collated from this study strongly support the use
of synthetic MMG as a potent adjuvant with excellent potential.
These results also show the potential for DDA vesicles as a platform
for the delivery of lipid based co-adjuvants in order to elicit poten-
tially effective Th1 biased immune responses against H1 and other
protein subunit vaccines. This study also emphasizes the impor-
tance of optimization of adjuvant dose or adjuvant ratio in partic-
ulate delivery platforms containing co-adjuvants. The inverse
dose-response adjuvant activity of the 5a isomer was highly
centration of 5
these microcultures were assessed for IFN-
l
g/ml H1 antigen. Supernatants obtained from
and IL-6, IL-5, IL-10
c
H1 Specific Spleen Cell Proliferation
20
15
10
5
***
1
2
3
4
5
6
7
8
9
10
***
0
0
2.5
5
10
[H1] microgrammes / ml
Figure 1. Spleen cell proliferation following restimulation with H1 antigen. ***
Significantly greater than all other groups (n = 3, P < 0.05). Legend Key: [micro-
grammes per dose of lipid adjuvant]; *GLYCEROL C32 (5a–d); **Expected
immunoactive agent. Treatment groups are delineated in Table 1.
Figure 2. H1 specific cytokine production following restimulation of spleen cells with antigen. (A) Interferon gamma (IFN-c); (B) Interleukin-6 (IL-6). Horizontal bars
represent the average cytokine production. Treatment groups are delineated in Table 1.

2032
V. Bhowruth et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2029–2032
Anti-H1 specific IgG2a titres
References and notes
100000
10000
1000
100
1. Hoft, D. F. Lancet 2008, 372, 164.
2. Lindblad, E. B.; Elhay, M. J.; Silva, R.; Appelberg, R.; Andersen, P. Infect. Immun.
1997, 65, 623.
3. Freund, J. Bibl. Tuberc. 1956, 130.
4. Bast, R. C., Jr.; Zbar, B.; Borsos, T.; Rapp, H. J. N. Engl. J. Med. 1974, 290, 1458.
5. Bast, R. C., Jr.; Zbar, B.; Borsos, T.; Rapp, H. J. N. Engl. J. Med. 1974, 290, 1413.
6. Koike, Y.; Yoo, Y. C.; Mitobe, M.; Oka, T.; Okuma, K.; Tono-oka, S.; Azuma, I.
Vaccine 1998, 16, 1982.
7. Holten-Andersen, L.; Doherty, T. M.; Korsholm, K. S.; Andersen, P. Infect. Immun.
2004, 72, 1608.
8. McBride, B. W.; Mogg, A.; Telfer, J. L.; Lever, M. S.; Miller, J.; Turnbull, P. C.;
Baillie, L. Vaccine 1998, 16, 810.
10
1
9. Rosenkrands, I.; Agger, E. M.; Olsen, A. W.; Korsholm, K. S.; Andersen, C. S.;
Jensen, K. T.; Andersen, P. Infect. Immun. 2005, 73, 5817.
10. Andersen, C. S.; Agger, E. M.; Rosenkrands, I.; Gomes, J. M.; Bhowruth, V.;
Gibson, K. J.; Petersen, R. V.; Minnikin, D. E.; Besra, G. S.; Andersen, P. J.
Immunol. 2009, 182, 424.
11. Dobson, G.; Minnikin, D. E.; Minnikin, S. M.; Parlett, J. M.; Goodfellow, M.;
Ridell, M.; Magnusson, M. In Chemical Methods in Bacterial Systematics;
Goodfellow, M., Minnikin, D. E., Eds.; London: Academic Press, 1985; pp
237–265.
1
2
3
4
5
6
7
8
9
Treatment Group
Figure 3. Serum anti-H1 specific IgG2a antibody titres following immunisation
with the various formulations including the C32 MMG analogues. Horizontal bars
represent the average antibody titres. Treatment groups are delineated (Table 1).
12. Datta, A. K.; Takayama, K.; Nashed, M. A.; Anderson, L. Carbohydr. Res. 1991,
218, 95.
unexpected. Speculation as to how this occurs may require inves-
tigation into physicochemical aspects as well as possible biological
mechanisms of action—such as interaction with Toll-like receptors
and antigen presenting cells. Such possible interactions are re-
viewed elsewhere.15 Delineation of the exact mechanisms that
underpin this inverse dose-response adjuvant activity provides
the basis for interesting research from a biological as well as a
pharmaceutical perspective.
13. Vaccine formulation: The vaccine delivery system consisting of
dimethyldioctadecylammonium bromide (DDA, Avanti Polar Lipids,
Alabaster, AL, USA), the experimental monomycolyl glycerol analogue, or the
adjuvant
a,
a0-trehalose 6,60-dibehenate (TDB, Avanti Polar Lipids, Alabaster,
AL, USA), were dissolved in chloroform/methanol (9:1) and dried under
nitrogen in flat bottomed glass vial to give a thin lipid film. The film was
hydrated with 10 mM tris buffer (pH 7.4), containing sucrose in order to
facilitate an isotonic solution suitable for injection, above the liquid-crystalline
temperature (Tc) of the DDA and the suspension was allowed to stand for about
30 min above Tc. Subsequently, recombinant H1 was added to give the desired
concentration of delivery system, adjuvant and protein for the purposes of
administration outlined above.
Acknowledgements
14. Kirby, D. J.; Rosenkrands, I.; Agger, E. M.; Andersen, P.; Coombes, A. G. A.;
Perrie, Y. J. Drug Target 2008, 16, 543.
15. Perrie, Y.; Mohammed, A. R.; Kirby, D. J.; McNeil, S. E.; Bramwell, V. W. Int. J.
Pharm. 2008, 364, 272.
This work was supported by the European Commission (LSHP-
CT-2003-503367), Mr. James Bardrick and a Royal Society Wolfson
Research Merit Award.