Roles of α-methyl trans-cyclopropane groups in behavior of mixed mycolic acid monolayers

Roles of α-methyl trans-cyclopropane groups in behavior of mixed mycolic acid monolayers

Accepted Manuscript Roles of α-Methyl trans-Cyclopropane Groups in Behavior of Mixed Mycolic Acid Monolayers Masumi Villeneuve, Hiroki Noguchi PII: D...

NAN Sizes 0 Downloads 1 Views

Accepted Manuscript Roles of α-Methyl trans-Cyclopropane Groups in Behavior of Mixed Mycolic Acid Monolayers

Masumi Villeneuve, Hiroki Noguchi PII: DOI: Reference:

S0005-2736(18)30318-3 https://doi.org/10.1016/j.bbamem.2018.10.019 BBAMEM 82880

To appear in:

BBA - Biomembranes

Received date: Revised date: Accepted date:

1 June 2018 15 September 2018 16 October 2018

Please cite this article as: Masumi Villeneuve, Hiroki Noguchi , Roles of α-Methyl transCyclopropane Groups in Behavior of Mixed Mycolic Acid Monolayers. Bbamem (2018), https://doi.org/10.1016/j.bbamem.2018.10.019

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Roles of α-Methyl trans-Cyclopropane Groups in Behavior

Hiroki Noguchi†

SC

Masumi Villeneuve∗12

RI PT

of Mixed Mycolic Acid Monolayers

∗ Graduate School of Integrated Arts and Sciences

NU

Hiroshima University, 7-1, Kagamiyama 1-chome, Higashi-Hiroshima, 739-8521 Japan †Department of Chemistry, Faculty of Science

AC

CE

PT E

D

MA

Saitama University, 255 Shimo-okubo, Sakura-ku, Saitama, 338-8570 Japan

1 2

corresponding author email address: [email protected]

ACCEPTED MANUSCRIPT

Key words Mycobacterial keto mycolic acid; α-methyl trans-cyclopropane

AC

CE

PT E

D

MA

NU

SC

RI PT

group; conformational behavior; phase diagram; mixed Langmuir monolayer

1

ACCEPTED MANUSCRIPT

RI PT

Abstract Mixed Langmuir films of type 1 alpha- (α-) and keto-mycolic acids (MAs) were investigated to understand the roles of α-methyl trans-cyclopropane

SC

containing keto-MA in determining the physical and chemical properties of the monolayers. Surface pressure (π) vs. mean molecular area (A) isotherms were measured at constant mole fractions defined as the ratio of the keto-MA

NU

molarity to the total molarity of α-MA and keto-MA (Xketo ) at 25 ◦C and 37 ◦C. A and the elastic modulus (E) of the mixed monolayer were compared

MA

for different Xketo at fixed π values. In keto-MA rich monolayers, A values were much larger than values the combined areas of α-MA and keto-MA,

D

while the E values were close to those of solid keto-MA monolayers. A and E

PT E

were also plotted against the mole fraction of α-methyl trans-cyclopropane containing keto-MA, which showed that the α-methyl trans-cyclopropane group stabilized the W-form conformation of mycolic acids in monolayers,

CE

and rendered them solids. Furthermore, a comparison of the experimental results and the α-methyl trans-cyclopropane content in cell-wall MAs from various strains indicated that the ratio of trans-cyclopropane content was

AC

important in determining the nature of the mixed MA layer.

ACCEPTED MANUSCRIPT

1

Introduction

Mycolic acid (MA) (Fig. 1) from mycobacterial cell walls is a long-chain 2alkyl branched, 3-hydroxy fatty acid with two intra-chain groups [X] and

RI PT

[Y] in the major meromycolate chain. According to the distal intra-chain group [X] in the chain, MAs are classified into three major groups: a nonoxygenated α-MA [X = cis-cyclopropane], and two oxygenated methoxy (MeO)-MA [X = –CH(Me)–CH(OMe)–] and keto-MA [X = –CH(Me)–CO–

SC

]. With respect to the vicinal group [Y], each is further grouped into three main types, with the major type-1 having [Y] as a cis- or α-methyl trans-

NU

cyclopropane group, and two minor types consisting of type-2 containing [Y] as a cis double bond and type-3 containing [X] as a trans double bond [1, 2].

MA

Here, α-MA and type-1 keto-MA (hereafter referred to as keto-MA) were

D

examined.

PT E

Figure 1: General structure of mycobacterial MA. Previous monolayers of MAs on water revealed that all the three major

CE

mycobacterial MAs, i.e., α-, MeO- and keto-MAs, had compact conformations where all four methylene chain segments of lengths, l, m, n and p were parallel at low outside pressure. These findings were deduced primarily from

AC

the fact that the Langmuir monolayers with average molecular areas near ˚2 molecule−1 , which corresponded to four times the cross-sectional area 80 A of a long alkyl chain, had high area elasticity values ranging from 100 to 1000 mN m−1 [3, 4]. This indicated dense packing of the methylene chains in the monolayer. Zuber et al. showed that a mycolic acid layer in such 1

ACCEPTED MANUSCRIPT

compact structures formed in the outer membrane inner leaflet, according to cryo-electron microscopy of vitreous sections of Mycobacterium bovis BCG, Mycobacterium smegmatis, and Corynebacterium glutamicum in their native states [5].

RI PT

A Langmuir monolayer is a single layer of insoluble molecules adsorbed on the surface of water. Intermolecular interaction between the film forming substances and water, and those between the film forming molecules deter-

SC

mine the molecular conformation. For a large molecule like MA with 60–90 carbon atoms, the intramolecular interactions become as significant as the

NU

intermolecular interactions. The fine structure of the compact conformation cannot be determined from thermodynamic analyses of the Langmuir mono-

MA

layer alone. Molecular structures of folded MAs were hypothesized from molecular dynamics (MD) simulations. Because keto-MA and MeO-MA respectively have hydrophilic carbonyls and methoxys as intrachain functional

D

groups in the meromycolate chains, W-form conformations were expected

PT E

with oxygen-containing intrachain groups in contact with a water surface. MD results that reflect the intramolecular interaction of a single molecule in vacuum and the thermodynamic and the optical properties of their monolay-

CE

ers confirmed the W-form conformations for keto-MA and MeO-MA [4]. The average molecular area occupied by α-MA in its compact form and

AC

the area elasticity of the Langmuir monolayer were somewhat smaller than those for keto-MA and MeO-MA, suggesting a different folded structure of α-MA molecule in the compact conformation. One reason may be that α-MA did not have a hydrophilic intrachain functional group in its meromycolate chain. MD studies showed that intramolecular interactions also greatly con-

2

ACCEPTED MANUSCRIPT

tribute to the less-tightly folded structure. Additionally matched lengths of the long l- and p- chains (1) has been implied [6]. Previous monolayer studies revealed that the stability of the W-form conformation on water varied for different MAs that had various α-methyl trans-

RI PT

cyclopropane content in the [Y] group [7]. Analogous results were obtained by macromodel energy calculation that yielded W-form conformations for all the MAs examined. MD showed that different MAs with different intra-chain

SC

groups had various tendencies for conformational changes [7].

Groenewald et al. applied atomistic MD simulations to stereochemically

NU

precise MAs from M. tuberculosis over a significantly extended timeframe relative to previous works. The α-MA was capable of a wide variety of conforma-

MA

tions, keto-MA underwent more rapid folding than that observed for α-MA and MeO-MA, and MeO-MA exibited folding properties that were intermediate between those of α-MAs and keto-MAs. The results suggested a pos-

D

sibility that the W-form conformation of monolayer keto-MA and MeO-MA

PT E

did not necessarily depend on the interactions between the oxygen-containing functional groups and the subphase water [8]. Nevertheless, the correlation between the MA chemical and folding struc-

CE

tures revealed through MD simulations does not exclude the importance of intermolecular interactions in W-form formation in the Langmuir monolayer.

AC

The large decrease in transition surface pressure with increasing temperature of MeO-MA from a W-form state to an extended state implied participation of hydrogen bonding between the hydrophilic groups and the water. That is, increased thermal motion of the molecules ruptured hydrogen bonds. To clarify how each factor contributes to the macroscopic properties and struc-

3

ACCEPTED MANUSCRIPT

tures of the monolayer, it would be necessary to conduct systematic thermodynamic studies using pure, synthetic MAs with well-defined chemical structures. Given that the α-methyl trans-cyclopropane group plays an active role in

RI PT

retaining the solid W-form conformation of the MA monolayers and probably in the cell wall, the physical nature of mixed MA monolayers was analyzed to reveal their behavior in the structure and function in bacilli. The

SC

macrostructure of the MA monolayers is attributed to the net molecular interactions. A Langmuir monolayer, a viscoelastic monolayer formed between

NU

two bulk phases having very different relative permittivities (ϵr (air) ≃ 1 and ϵr (water) ≃ 80), is a simplified model of a biological membrane that is

MA

suitable for a study focused on physical chemistry. Previous studies have shown the effects of temperature and surface pressure (or lateral pressure) on monolayers formed from different MA classes. Moreover, a monolayer

PT E

computer simulations.

D

of mixed MAs with a density of 1018 molecules per m2 , is too complex for

Two mixed Langmuir monolayers were examined via surface pressure measurements. One was prepared with α-MA and keto-MA, both from My-

CE

cobacterium tuberculosis (M. tb) AoyamaB, and the other was prepared with α-MA from AoyamaB and keto-MA from Mycobacterium bovis BCG (BCG).

AC

The systems were comprised of three main components: cis-cyclopropane containing α-MA, cis-cyclopropane containing keto-MA, and α methyl transcyclopropane containing keto-MA. Therefore, experimental and analytical results yielded a comprehensive understanding of how trans-cyclopropane content affected the physical properties of mixed MA monolayers.

4

ACCEPTED MANUSCRIPT

From this detailed analysis of mixed MA monolayers having various amounts cis/trans-cyclopropane, its effect was more decisively confirmed and provided

2

Materials and Methods

RI PT

a better understanding of the MA role in the mycobacterial cell wall.

Materials used for monolayer studies: α-MA came from Mycobacterium

SC

tuberculosis (M. tb) AoyamaB, and keto-MA came from M. tb AoyamaB and Mycobacterium bovis BCG (BCG) Tokyo 172, as described previously [1, 2]

NU

(Fig. 1). MA structures, compositions, and molecular weights are summarized in Table 1.

p = 23 for all the MAs.

a

MA

Table 1: Molecular structures, compositions and molecular weights of MAs. Values of major components.

b

Implied by 1 H-NMR spectra.

Samples details can be found in the literature [1, 2].

M. tb

α-MA

Aoyama B

CE

keto

M. bovis BCG

Distal group [X]

D

Classes

Proximal cyclopropane [Y]

PT E

Strains

keto

cis-cyclopropane

cis/trans

Av. MW

cis/trans, n–m–la

ratiob

cis 19–14–11

all cis

1150

1/2.7

1267

0.33/1

1246

cis 19–14–13

CH(CH3 )-CO

cis 15–18–17 trans 16–18–17

CH(CH3 )-CO

cis 15–18–17

Tokyo 172

cis 17–18–17

AC

trans 16–18–17

Other reagents: Distilled reagent-grade chloroform (Wako chemicals)

was used as the spreading medium. Water was distilled once and deionized by Milli-Q Plus (resistance 18.2 MΩ cm). 5

ACCEPTED MANUSCRIPT

Surface pressure vs. mean molecular area (π vs. A) isotherms measurement: Surface pressure (π) vs. mean molecular area (A) isotherms of two mixed Langmuir monolayers of MA were measured for various mole fractions of keto-MA Xketo , defined by cketo , cα + cketo

RI PT

Xketo =

(1)

where cα and cketo were molarities of α- and keto-MAs, respectively in the

SC

chloroform.

The mole fraction of the α-methyl-trans-cyclopropane keto-MA was defined as follows. Because the ratio of the methyl trans-cyclopropane con-

NU

taining keto-MA and the cis-cyclopropane containing keto-MA for M. tb was 2.7 : 1, and that for BCG was 0.33 : 1, Xtrans for the α-MA (M. tb) – keto-MA

MA

(M. tb) system and α-MA (M. tb) – keto-MA (BCG) system were defined by Xtrans = Xketo ×2.7/(2.7+1) and Xtrans = Xketo ×0.33/(0.33+1), respectively.

D

The Langmuir monolayers were prepared by spreading a chloroform so-

PT E

lution of MA mixtures (1 ml, ca. 3 ∼ 6 × 10−5 M) on the water surface. Measurement were performed on a Lauda Filmwaage (FW1) (LAUDA, Lauda-K¨onigshofen, Germany) or a FACE HBM700LB (Kyowa Interface Science Co., Ltd., Saitama, Japan) tough. The monolayer compression rate was

CE

˚2 molecule−1 min−1 . π vs. A isotherms were measured at 25 ◦C and 14 A 37 ◦C. The subphase temperature was controlled to within ±0.2 ◦ C. The

AC

room temperature was 23 ± 1 ◦ C. Each measurement was repeated 2–10 times.

Collapse pressures were determined as follows. Hysteresis curves of π vs.

A isotherms were repeatedly measured with different turnback surface pressures on the same monolayer. Surface pressures just below and above kinks 6

ACCEPTED MANUSCRIPT

on π vs. A isotherms were the turnback surface pressures, and the hysteresis measurements were repeated with increased turnback surface pressures, starting with the lowest. If a hysteresis curve with a turnback pressure above a kink did not reproduce the one conducted with a turnback pressure below

RI PT

the kink, but instead shifted to smaller molecular areas, the kink was determined to be a π cp . As an example, π vs. A hysteresis curves of a α-MA (M. tb) – keto-MA (BCG) mixed monolayer (Xketo = 0.900) measured at 37 ◦C were

SC

plotted in Fig. 2. The π vs. A curves for the first and second compressions exactly overlapped; however, the π vs. A curve for the third compression did

NU

not reproduce them. To obtain the same π value, the monolayer had to be compressed to smaller molecular area in the third compression relative to the

MA

first and second compressions. Thus the π cp was 27 mN m−1 . The error bars in the graphs obtained by analyzing the data shown in Figs. √ 3 and 4 represented the uncertainties calculated by 1.960σ/ N , a product

Results

CE

3

PT E

D

of the confidence level (95%) and the standard deviation σ, divided by the √ square root of the data number N .

The π vs. A isotherms for different Xketo obtained at 37 ◦C and 25 ◦C are

AC

shown in Figs. 3 and 4, respectively. The phase transition was defined as described elsewhere [6]. Phase diagrams were constructed by plotting the π values where the phase transitions occurred (π tr ), and where the films collapsed (π cp ) vs. Xketo . The resulting phase diagrams at 37 ◦C and 25 ◦C are shown in Figs. 5

7

ACCEPTED MANUSCRIPT

60

40

20 0

10 0 20

(

40

60 (

&

RI PT

.

/

30

SC

π!"!#$!#

%&

50

80

100

%&

120

NU

'!"!' !#)*+,-*+

MA

Figure 2: The collapse pressure π cp was determined by hysteresis measurements of a π vs. A isotherm (e. g. Xketo = 0.900 at 37 ◦C of the α-MA (M. tb) – keto-MA (BCG) monolayer): (1) first compression up to π = 25 mN

D

m−1 ; (2) fist expansion after holding for 10 min; (3) second compression up

PT E

to π = 25 mN m−1 ; (4) second expansion; (5) third compression. and 6, respectively. The π cp was greatly reduced by mixing the two MAs at all the temperatures. With respect to the α-MA (M. tb) – keto-MA (BCG)

CE

mixed monolayer, at 37 ◦C, there was a phase transition at a low π for every composition, whereas at 25 ◦C, there was no phase transition up to π cp . In

AC

contrast, the α-MA (M. tb) – keto-MA (M. tb) mixed monolayers at both temperatures partly collapsed at low π, as indicated by filled square symbols in Fig. 5 B and by filled triangles in Fig. 6 B near Xketo = 0. The mean molecular area (A) was plotted vs. Xketo at π = 5, 10 and 15 mN m−1 in Figs. 7 and 8 at 37 ◦C and 25 ◦C, respectively. Broken lines were 8

ACCEPTED MANUSCRIPT

60

60

"

40 8 3

20 1

10

2

6 4

0 20

40

7

10 9

7

30 20 1

8

6

3

10

5

60 80 o A / A2 molecule-1

40

2

4

9

5

0 100

120

20

RI PT

30

!

50 π / mN m-1

π / mN m-1

50

40

60 80 o A / A2 molecule-1

100

120

Figure 3: π vs. A isotherms of the mixed Langmuir monolayer of α- and

SC

keto-MAs at 37 ◦C: (A) α-MA (M. tb) – keto-MA (BCG); (1) Xketo = 0 (α-MA); (2) 0.150; (3) 0.250; (4) 0.299; (5) 0.377; (6) 0.500; (7) 0.618; (8)

NU

0.750; (9) 0.900; (10) 1 (keto-MA); (B) α-MA (M. tb) – keto-MA (M. tb); (1) Xketo = 0 (α-MA); (2) 0.050; (3) 0.100; (4) 0.200; (5) 0.348; (6) 0.500; (7)

MA

0.667; (8) 0.900; (9) 1 (keto-MA). Filled arrowheads show collapse pressures and open arrowheads phase transition pressures.

D

drawn to connect the molecular areas for α-MA-only and for keto-MA-only. For both temperatures and surface pressures, the overall trend was that the

PT E

mean molecular areas of the mixed monolayer deviated positively from the straight lines. Only in the α-MA (M. tb) – keto-MA (M. tb) system at 37 ◦C, for all the surface pressures, did the A values in a limited composition region

CE

(0 ≤ Xketo < 0.2) appear linear. Those for Xketo ≥ 0.2 deviated upward from the straight lines.

AC

The data were replotted vs. the mole fraction of α methyl trans-cyclopropane

containing keto-MA, Xtrans , to more clearly illustrate the effect of transcyclopropane content on monolayer properties. In the α-MA (M. tb) – keto-MA (BCG) mixed monolayer at 37 ◦C and 25 ◦C, A increased steeply with increasing Xtrans , and reached values very 9

ACCEPTED MANUSCRIPT

60

60

"

40

8

30

7

5

20

4

9 6

10

1

10 3

!

50 π / mN m-1

40 5

30 2

3

20

6

7

4

8 9

1

10

2

0

0

20

40

60 80 o A / A2 molecule-1

100

120

RI PT

π / mN m-1

50

20

40

60 80 o A / A2 molecule-1

100

120

Figure 4: π vs. A isotherms of the mixed Langmuir monolayer of α- and

SC

keto-MAs at 25 ◦C: (A) α-MA (M. tb) – keto-MA (BCG); (1) Xketo = 0 (α-MA); (2) 0.150; (3) 0.250; (4) 0.299; (5) 0.377; (6) 0.500; (7) 0.618; (8)

NU

0.750; (9) 0.900; (10) 1 (keto-MA); (B) α-MA (M. tb) – keto-MA (M. tb); (1) Xketo = 0 (α-MA); (2) 0.050; (3) 0.100; (4) 0.200; (5) 0.348; (6) 0.500; (7)

MA

0.667; (8) 0.900; (9) 1 (keto-MA). Filled arrowheads show collapse pressures and open arrowheads phase transition pressures.

30 20 10

0

0.2 0.4 0.6 0.8 X

CE

0

π / mN m-1

!

PT E

π / mN m-1

40

D

50

50 40 30 20 10 0

1

keto

"

0

0.2 0.4 0.6 0.8 X

1

keto

AC

Figure 5: Phase diagrams of the mixed Langmuir monolayer of α- and ketoMAs at 37 ◦ C: (A) α-MA (M. tb) – keto-MA (BCG); (B) α-MA (M. tb) –



keto-MA (M. tb). Filled symbols ( , !): π cp vs. Xketo ; open circle: π tr vs. Xketo .

10

ACCEPTED MANUSCRIPT

50 40

!

π / mN m-1

π / mN m-1

40 30 20 10 0

0

0.2 0.4 0.6 0.8 X

30 20 10 0

1

"

0

keto

RI PT

50

0.2 0.4 0.6 0.8 X

1

keto

SC

Figure 6: Phase diagrams of the mixed Langmuir monolayer of α- and ketoMAs at 25 ◦ C: (A) α-MA (M. tb) – keto-MA (BCG); (B) α-MA (M. tb) –



NU

keto-MA (M. tb). Filled symbols ( , !, ", #): π cp vs. Xketo ; open circle: π tr vs. Xketo .

90

D

60 50 40

o2

70

A / A molecule

-1

MA

!

80

PT E

o2

A / A molecule

-1

90

0

0.2 0.4 0.6 0.8 X

1

80

"

70 60 50 40

0

keto

0.2 0.4 0.6 0.8 X

1

keto

CE

Figure 7: Mean molecular area vs. Xketo plots of the mixed Langmuir monolayer of α- and keto-MAs at 37 ◦ C: (A) α-MA (M. tb) – keto-MA (BCG);



AC

(B) α-MA (M. tb) – keto-MA (M. tb): π = 5 ( ); 10 (!); 15 mN m−1 ("). Broken lines connecting the molecular areas of α-MA and keto-MA show the deviation of the areas from the area additivity.

11

ACCEPTED MANUSCRIPT

90

70

o2

60 50 40

0

0.2 0.4 0.6 0.8 X

70 60 50 40

1

"

80

RI PT

-1

!

80

A / A molecule

o2

A / A molecule

-1

90

0

0.2 0.4 0.6 0.8 X

keto

1

keto

SC

Figure 8: Mean molecular area vs. Xketo plots of the mixed Langmuir monolayer of α- and keto-MAs at 25 ◦ C: (A) α-MA (M. tb) – keto-MA (BCG);



NU

(B) α-MA (M. tb) – keto-MA (M. tb): π = 5 ( ); 10 (!); 15 mN m−1 ("). Broken lines connecting the molecular areas of α-MA and keto-MA show the

MA

deviation of the areas from the area additivity.

60 50 40

0

0.1

0.2 X

0.3

o2

D

70

A / A molecule

-1

!

80

PT E

o2

A / A molecule

-1

90

0.4

"

80 70 60 50 40

0

0.2 0.4 0.6 0.8 X

1

trans

CE

trans

90

Figure 9:

Mean molecular area vs. mole fraction of α methyl trans-

AC

cyclopropane containing keto-MA, Xtrans plots of the mixed Langmuir monolayer of α- and keto-MAs at 37 ◦C: (A) α-MA (M. tb) – keto-MA (BCG);



(B) α-MA (M. tb) – keto-MA (M. tb): π = 5 ( ); 10 (!); 15 mN m−1 (").

12

ACCEPTED MANUSCRIPT

90

70

o2

60 50 40

0

0.1

0.2 X

0.3

0.4

"

80 70 60 50 40

0

trans

RI PT

-1

!

80

A / A molecule

o2

A / A molecule

-1

90

0.2 0.4 0.6 0.8 X

1

trans

SC

Figure 10: Mean molecular area vs. mole fraction of α methyl transcyclopropane containing keto-MA, Xtrans plots of the mixed Langmuir mono-

NU

layer of α- and keto-MAs at 25 ◦C: (A) α-MA (M. tb) – keto-MA (BCG);



(B) α-MA (M. tb) – keto-MA (M. tb): π = 5 ( ); 10 (!); 15 mN m−1 (").

MA

˚2 molecule−1 at Xtrans ≃ 0.2 (Panel A in Figs. 9 and 10). Also close to 80 A in the α-MA (M. tb) – keto-MA (M. tb) system, A increased steeply with

D

increasing Xtrans (Panel B in Figs. 9 and 10).

PT E

The mechanical properties of the membranes maintain the characteristic shapes of living cells. The area elasticity of a monolayer (E) is defined by

CE

the following equation in analogy to the modulus of volume elasticity: ! " ∂π E = −A ∂A p,T

(2)

E values of the mixed MA monolayers were evaluated at π = 5, 10 and 15

AC

mN m−1 , at 37 ◦C and 25 ◦C, respectively, in Figs. 11 and 12. In the α-MA (M. tb) – keto-MA (BCG) system at 37 ◦C and Xketo near 0, the E values were close to those of the α-MA monolayer at all temperatures and surface pressures (Panel A of Figs. 11 and 12). At 37 ◦C, for 0.2 < Xketo < 0.7, the E values were almost the same as those of the monolayer of α-MA. Above 13

ACCEPTED MANUSCRIPT

Xketo = 0.7, the E values increased abruptly and approached those of ketoMA. These values, 100–270 mN m−1 , were equivalent to that of a liquid condensed monolayer of fatty acids [9]. At 25 ◦C, the 80–350 E values for 0.2 < Xketo < 0.7 were larger than those for a α-MA monolayer. The E

RI PT

values for 0.7 < Xketo increased abruptly to 200–550 mN m−1 , especially for π = 10 and 15 mN m−1 . 300

300 250

150 50

100

50

0

0.2 0.4 0.6 0.8 X keto

1

MA

0

150

NU

100

)

200

SC

!"#"$%"$

200

&'

(

!"#"$%"$

&'

250

0

0

0.2 0.4 0.6 0.8 X

1

keto

Figure 11: Area elasticity vs. Xketo for mixed Langmuir monolayers of α- and

D

keto-MAs at 37 ◦ C: (A) α-MA (M. tb) – keto-MA (BCG); (B) α-MA (M. tb)



PT E

– keto-MA (M. tb): π = 5 ( ); 10 (!); 15 mN m−1 ("). In the α-MA (M. tb) – keto-MA (M. tb) system, at 37 ◦C, the E values

CE

of the mixed monolayers were larger than those of the α-MA monolayer. Above Xketo = 0.7, the E values increased even more and approached those of the keto-MAs, 100–270 mN m−1 . At 25 ◦C, the 80–350 E values for 0.2 <

AC

Xketo < 0.7 were larger than those of the α-MA monolayer, and the E values for 0.7 < Xketo increased to 200–550 mN m−1 , especially for π = 10 and 15 mN m−1 . In the α-MA (M. tb) – keto-MA (M. tb) system, the sudden increase in the E values took place at Xketo = 0.2. The E values were 200– 550 mN m−1 , which were similar to those in the α-MA (M. tb) – keto-MA 14

ACCEPTED MANUSCRIPT

600

600 500

400

&'

(

!"#"$%"$

300 200 100 0

)

400 300 200 100

0

0.2 0.4 0.6 0.8 X

0

1

0

keto

RI PT

!"#"$%"$

&'

500

0.2 0.4 0.6 0.8 X

1

keto

SC

Figure 12: Area elasticity vs. Xketo for mixed Langmuir monolayers of α- and keto-MAs at 25 ◦ C: (A) α-MA (M. tb) – keto-MA (BCG); (B) α-MA (M. tb)



NU

– keto-MA (M. tb): π = 5 ( ); 10 (!); 15 mN m−1 ("). (BCG) system. Additonally, E = 550 mN m−1 is quite a large value for a

MA

liquid condensed film, it is similar to values for a solid condensed film [9]. 300 (

&'

!"#"$%"$

D

200

!"#"$%"$

&'

250

PT E

150 100

50

0

CE

0

0.1

0.2 X

300 )

250 200 150 100 50

0.3

0.4

trans

0

0

0.2 0.4 0.6 0.8 X

1

trans

AC

Figure 13: Area elasticity vs. mole fraction of α methyl trans-cyclopropane containing keto-MA, Xtrans for mixed Langmuir monolayers of α- and ketoMAs at 37 ◦C: (A) α-MA (M. tb) – keto-MA (BCG); (B) α-MA (M. tb) –



keto-MA (M. tb): π = 5 ( ); 10 (!); 15 mN m−1 ("). The E values were also replotted vs. the mole fraction of α methyl trans15

ACCEPTED MANUSCRIPT

600

600 ( &'

400 300 200

400 300 200 100

0

0.1

0.2 X

0.3

0

0.4

0

trans

RI PT

100 0

)

500

!"#"$%"$

!"#"$%"$

&'

500

0.2 0.4 0.6 0.8 X

1

trans

SC

Figure 14: Area elasticity vs. mole fraction of α methyl trans-cyclopropane containing keto-MA, Xtrans for the mixed Langmuir monolayers of α- and



NU

keto-MAs at 25 ◦C: (A) α-MA (M. tb) – keto-MA (BCG); (B) α-MA (M. tb) – keto-MA (M. tb): π = 5 ( ); 10 (!); 15 mN m−1 (").

MA

cyclopropane containing keto-MA, Xtrans . For the α-MA (M. tb) – keto-MA (BCG) system at 37 ◦C, the E values of the mixed monolayer increased at

D

Xtrans = 0.2 to values larger than 100 mN m−1 (Panel A in Fig. 13). At 25 ◦C, except at π = 5 mN m−1 , the E values exceeded 200 mN m−1 at Xtrans = 0.2

PT E

(Panel A in Fig. 14).

For the α-MA (M. tb) – keto-MA (M. tb) system, at 37 ◦C, E depended positively on Xtrans , and for π = 10 and 15 mN m−1 , E was greater than

CE

100 mN m−1 at Xtrans = 0.2. At 25 ◦C, E increased at Xtrans = 0.2 and had large E values, including 550 mN m−1 , which was close to those for a solid

AC

condensed monolayer (Panel B in Fig. 14).

16

ACCEPTED MANUSCRIPT

4

Discussion

4.1

Miscibility of α- and keto-MAs in the monolayer

The π vs. A curves of the mixed monolayers did not exhibit simple additive

RI PT

properties. As shown in Figs. 3A and 4A, the π vs. A curves of α-MA (M. tb)–keto MA (BCG) system were divided into five groups: curve (1: Xketo = 0); (2: Xketo = 0.150) and (3: Xketo = 0.250); (4: Xketo = 0.299)–(6:

SC

Xketo = 0.500); (7: Xketo = 0.618); and (8: Xketo = 0.667)–(10: Xketo = 1). Curve (7) followed the π vs. A curve of the last group at low surface pressure,

NU

and then followed the π vs. A curve of the fourth group at high surface pressure.

MA

Although thermodynamic studies did not reveal microscopic information, the A and E vs. Xketo diagrams had clues concerning the MA conformations. It was important to assess whether the MAs formed homogeneous mono-

D

layers over the entire Xketo range. If the monolayer components behaved

PT E

independently, individual film collapses would be observed at the original collapse pressures at all Xketo values. Simultaneously, the A vs. Xketo plots would be linear like the ones in Figs. 7 and 8. The actual π cp changed with

CE

Xketo ; therefore, α-MA and keto-MA (BCG) were miscible in the monolayer. Furthermore, the A vs. Xketo plots deviated positively from straight lines,

AC

indicating that the different classes MAs mixed with each other in the monolayer.

However, the reduced π cp suggested that the different classes of MAs were

not very miscible. In partcular, for the α-MA rich region (near Xketo = 0), the decrease of π cp with increasing Xketo was steep, indicating that keto17

ACCEPTED MANUSCRIPT

MA was expelled from the α-MA-rich condensed monolayer. Stable mixed monolayers were formed only when the monolayer was not densely packed. Therefore, from Xketo = 0 to Xketo = 0.1 where a minimum π cp existed, the monolayer was extremely rich in the α-MA. When the monolayer was

RI PT

compressed to a certain surface pressure, keto-MA was segregated into a bulk phase. In Panel A in Fig. 3, as the surface pressure increased, the π vs. A curve of Xketo = 0.150 (2) approached the π vs. A curve of Xketo = 0 (1),

SC

and overlapped it at high surface pressure. In the keto-MA rich region, the variation of π cp with Xketo was less steep, which implied that keto-MA and

NU

α-MA were more miscible in the keto-MA-rich monolayer than in the α-MArich monolayer. The π vs. A curves of Xketo = 0.618 at 25 and 37 ◦C (curve 7

MA

in Panel A of Figs. 3 and 4) revealed that, in the keto-MA rich region, α-MA and keto-MA formed a mixed monolayer at relatively low surface pressure. However, when the monolayer was compressed again, keto-MA was excluded

D

and the monolayer became enriched with α-MA.

PT E

In the α-MA (M. tb)–keto-MA (M. tb) system, the miscibility of α-MA and keto-MA was worse than that of the α-MA (M. tb)–keto-MA (BCG) system. The phase diagrams of α-MA (M. tb)–keto-MA (M. tb) (Figs. 5B and

CE

6B) indicated that, in α-MA rich monolayers, the first film collapse occurred at low surface pressure (ca. 2–3 mN m−1 at 37 ◦C and ca. 9 mN m−1 at 25 ◦C)

AC

and the second film collapse followed at a higher surface pressure. The monolayer of α-MA at 25 ◦C had two collapse pressures (π cp = 31

and 42 mN m−1 ). This was possible because the two main components of α-MA (Table 1) were not miscible at high surface pressures, and one of the components was segragate into a bulk crystalline phase. At a higher

18

ACCEPTED MANUSCRIPT

surface pressure, the remaining component started to crystalize. At Xketo = 0.100 and 25 ◦C, there were four collapse pressures. The lowest one probably corresponded to the pressure at which keto-MA was removed. Because αMA and keto-MA consisted of multiple components, the number of collapse

4.2

RI PT

pressures could have reflected the number of the monolayer components.

Implied conformations of MA molecules

SC

The A values of α-MA at all the surface pressures at both temperatures were 48–57 ˚ A2 molecule−1 (Figs. 7 and 8). If the cross-sectional area of a methylene

NU

chain in a condensed monolayer was 20 ˚ A2 molecule−1 [10], these values ˚2 molecule−1 corresponded to twice or three times that value. For A = 57 A

MA

at π = 5 mN m−1 , the corresponding E value was 40 mN m−1 for 37 ◦C and 80 mN m−1 for 25 ◦C. At these values, the α-MA monolayer would be in a liquid expanded state. Thus, it was possible that the MA molecules were unfolded,

D

with the alpha and meromycolate chains subjected to little constraint. At

PT E

π = 10 and 15 mN m−1 , the E values were 100–150 mN m−1 , with the monolayers in a liquid condensed state. Furthermore, for A values of 48– 54 ˚ A2 molecule−1 , the α-MA molecules were packed in a denser environment

CE

with less space than at π = 5 mN m−1 for the alpha and meromycolate chains, which were possibly fully extended. The latter speculation was reasonable

AC

because the meromycolate chains were not straight alkyl chains. Instead, they contained two intrachain functional groups, and could occupy larger cross-sectional areas than that of an alkyl chain. For the surface pressures used in here, α-MA had an extended conformation, where the keto-MA had a compact W-conformation, irrespective of 19

ACCEPTED MANUSCRIPT

the surface tension, that occupied twice as large an area as α-MA [3, 4]. Generally, the A vs. Xketo plots were not linear. In the α-MA (M. tb)–ketoMA (BCG) system at 37 ◦C, A was constant within experimental error for Xketo = 0.150 and 0.250 at π = 5 and 10 mN m−1 (Plots for π = 15 mN m−1

RI PT

were missing because the monolayers collapsed at π ≃ 10 mN m−1 ). A was

singularly large at Xketo = 0.618 at π = 5 mN m−1 (Fig. 7A). At 25 ◦C, A was constant at Xketo = 0.150 and 0.250 and between 0.299 ≤ Xketo ≤ 0.500

SC

(Fig. 8A).

In the mixed monolayer system of α-MA (M. tb)–keto-MA (M. tb) at

NU

37 ◦C, the A vs. Xketo data at Xketo = 0–0.150 and 1 was linear connecting the A values of α- and keto-MAs. The rest of the points positively deviated

MA

from linearity at all the π values.

From Figs. 7–12, the MAs in the mixed monolayers at Xketo values near ˚2 molecule−1 , and the 1 took W-forms, because A had values of ca. 80 A

4.3

PT E

D

monolayers were rigid (E > 100 mN m−1 ).

Possible effect of α-methyl trans-cyclopropane con-

CE

taining keto-MA on the mixed monolayer The dependence of monolayer properties on Xtrans changed at Xtrans = 0.2. In the α-MA (M. tb) – keto-MA (BCG) mixed monolayer at 37 ◦C and 25 ◦C,

AC

A reached 80 ˚ A2 molecule−1 at Xtrans ≃ 0.2 (Panel A in Figs. 9 and 10), and E of the mixed monolayer increased to values corresponding to condensed monolayers at Xtrans = 0.2 (Panel A in Figs. 13 and 14). In the α-MA (M. tb) – keto-MA (M. tb) mixed monolayers at 25 ◦C, E increased at Xtrans = 0.2 and was nearly equal to the values for the keto-MA (M. tb) monolayer (Fig. 20

ACCEPTED MANUSCRIPT

14). Previous studies of surface monolayers with different trans-cyclopropane content [4, 7], such as keto-MA from M. tb H37Ra (trans-cyclopropane content: 0%), did not observe W-form conformations. This may suggest a

4.4

RI PT

physcal importance of trans-cyclopropane content in the mixed MA layer.

Relevance of the role of α-methyl trans-cyclopropane

SC

containing keto-MA to bacilli

It is not clear whether it is the inner leaflets (mycolyl-arabinogalactan-peptidoglycan

NU

complex), the outer leaflets (esters of trehalose or glycerol) of the cell wall, or unbound MAs, that are most relevant to the mixed monolayer results.

MA

Despite many studies on biosynthesis pathways of mycolic acids in Mycobecteria [11–16], they are not yet fully understood. There are at least

D

five arabinofuranosyltransferase (Aft) enzymes involved in building mycolylarabinogalactan-peptidoglycan complexes of the cell wall [17–22]. A mutant

PT E

strain of Corynebacterium glutamicum lacking AftB had a partially disrupted cell surface, and bound mycolates in the cell wall decreased by 40%, accom-

CE

panied by a 1.7-fold increase in free mycolates [22]. Raad et al. suggested that mycolic acids were synthesized prior to binding to the arabinogalactan. Mixed MA molecules of different classes could form the optimum arrange-

AC

ment that fortifies the MA monolayer before esterification with arabinoses in the cell wall.

Although most MAs from mycobacterial origins are covalently linked to

arabinogalactan, some free MAs are exported and found in mycobacterial biofilms. They may be required for biofilm maturation in the drug-tolerant, 21

ACCEPTED MANUSCRIPT

persistence stage of the bacilli [23]. Keto-MA was required in pellicle biofilm growth of M. tuberculosis, which was essential in conferring drug tolerance to the bacteria [24]. The mixed monolayer results could explain the importance of keto-MA in this biofilm growth.

RI PT

Here and previously [4, 7], it was shown that small amounts of the trans isomer of keto-MA were required to form rigid monolayers. Oxygenated MAs were more antigenic than the non-oxygenated α-MA, and the trans

SC

isomer was shown to be more antigenic than the cis isomer. [25,26]. Another possibility was the relevance to the interaction with the host upon entering

NU

macrophages and the long-term survival of the mycobacteria [27, 28]. The rigid monolayer rendered by keto-MA, having low surface energy (ca. 40 mJ

MA

m−2 at 37 ◦C) could have high affinity for cholesterol inside the macrophages. Data showing the effect of the trans-cyclopropane content on an actual mycobacterial cell wall layer also emphasized its physical and biological im-

D

portance. The ratio between the major cell-wall MAs, α-, methoxy and keto

PT E

MA is various. For example, it is 1 : 0.04 : 1.5 in M. bovis BCG Danish 1331 strain and 1 : 0.97 : 2.7 in the Tokyo 174 strain. The ratio is hard to be said the same, but these two strains are both used as prequalified UNICEF BCG

CE

vaccines, and are expected to have the same immunological effect or behavior in the human body. However, when the trans-cyclopropane contents in the

AC

mycobacterial cell wall MA of various human pathogenic mycobacteria were compared, the ratio was in the same range, as shown in the Table 2. With this ratio, the cell wall may retain more resistance to outside pressure, yet be strong enough to retain the W-form conformation (same layer thickness).

22

ACCEPTED MANUSCRIPT

Table 2: α-methyl trans-cyclopropane content in cell wall MAs derived from the data published in [29, 30]. *UNICEF’s prequalified BCG vaccine [31– 33]; **Used only regionally in Vietnam and Indonesia; ***not used as BCG vaccine today.

RI PT

α-Me trans-cyclopropane content (%) Strains

in cell wall MAs

M. tb Aoyama B

18.38

SC

M. tb MNC 1397 M. tb 9829/87

NU

M. tb 4610/91 M. tb 2668/92

MA

M. tb H37Rv M. bovis MNC 8 M. bovis MNC 27

15.84 18.09 17.78 17.18 17.18 19.58 20.81 17.43

M. bovis BCG Danish MNC 5***

17.43

M. bovis BCG Glaxo 10-F***

11.52

M. bovis BCG Pasteur 1173 P2**

11.84

M. bovis BCG Tokyo 172*

14.71

M. bovis BCG Tokyo 172-V

15.92

M. microti OV 248

9.53

M. kansasii 20-01

32.47

M. avium complex KK41-24

43.1

AC

CE

PT E

D

M. bovis BCG Danish 1331*

23

ACCEPTED MANUSCRIPT

5

Conclusions

The possible effects of MAs having trans-cyclopropane rings on biomembrane functions or pathogenicity have been considered important [11, 12]. Here,

RI PT

Langmuir monolayers were examined with respect to trans-cyclopropane ring content. It was found that at Xtrans > 0.2 (20 %), all the MAs were in the W-shape conformation (Figs. 9 and 10) and that mixed monolayers had an equivalent rigidness to that of liquid or solid condensed films (Figs. 13 and

SC

14). These forms were the most likely for keto-MAs [3, 4]. These observations (Figs. 9, 10, 13, and 14) and the data shown in Table 2 suggest that

NU

trans-cyclopropane is important in cell wall leaflets. However, only about 20% content of trans-cyclopropane was enough to retain the W-shape con-

MA

formation of the MAs and to fortify the cell wall. Using natural MAs, the role of the α-methyl trans-cyclopropane group containing keto-MA in mixed α-MA and keto-MA monolayers was deter-

D

mined. More detailed monolayer properties and information on functions

PT E

of MAs could be obtained via thermodynamic studies on mixed monolayers prepared from synthetic MAs with known absolute stereochemistry, defined

Acknowledgements

AC

6

CE

chain lengths, and functional group content [28, 34–38].

Special thanks go to Dr. Motoko Watanabe for providing her precious mycolic acid samples. We thank Alan Burns, PhD, from the Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript. This research was partly supported by Grant-in-Aid for Scientific Research (C) 24

ACCEPTED MANUSCRIPT

17K05836.

References

RI PT

[1] M. Watanabe, Y. Aoyagi, M. Ridell, and D. E. Minnikin. Separation and characterization of individual mycolic acids in representative mycobacteria. Microbiology, 147:1825–1837, 2001.

SC

[2] M. Watanabe, Y. Aoyagi, H. Mitome, T. Fujita, H. Naoki, M. Ridell, and D. E. Minnikin. Location of functional groups in mycobacterial

NU

meromycolate chains; the recognition of new structural principles in mycolic acids. Microbiology, 148:1881–1902, 2002.

MA

[3] Masumi Villeneuve, Mizuo Kawai, Hideki Kanashima, Motoko Watanabe, David E Minnikin, and Hiroo Nakahara. Temperature dependence

D

of the Langmuir monolayer packing of mycolic acids from MycobacBiochimica et Biophysica Acta-Biomembranes,

PT E

terium tuberculosis.

1715(2):71–80, September 2005. [4] Masumi Villeneuve, Mizuo Kawai, Motoko Watanabe, Yutaka Aoy-

CE

agi, Yukio Hitotsuyanagi, Koichi Takeya, Hiroaki Gouda, Shuichi Hirono, David E Minnikin, and Hiroo Nakahara. Conformational behav-

AC

ior of oxygenated mycobacterial mycolic acids from Mycobacterium bovis BCG. Biochimica et Biophysica Acta-Biomembranes, 1768(7):1717– 1726, July 2007.

[5] B Zuber, M Chami, C Houssin, J Dubochet, G Griffiths, and M Daff´e. Direct Visualization of the Outer Membrane of Mycobacte25

ACCEPTED MANUSCRIPT

ria and Corynebacteria in Their Native State. Journal of Bacteriology, 190(16):5672–5680, July 2008. [6] Masumi Villeneuve, Mizuo Kawai, Motoko Watanabe, Yutaka Aoyagi,

RI PT

Yukio Hitotsuyanagi, Koichi Takeya, Hiroaki Gouda, Shuichi Hirono, David E Minnikin, and Hiroo Nakahara. Differential conformational behaviors of -mycolic acids in Langmuir monolayers and computer sim-

SC

ulations. Chemistry and Physics of Lipids, 163(6):569–579, June 2010. [7] Masumi Villeneuve, Mizuo Kawai, Kieko Horiuchi, Motoko Watanabe,

NU

Yutaka Aoyagi, Yukio Hitotsuyanagi, Koichi Takeya, Hiroaki Gouda, Shuichi Hirono, and David E Minnikin.

Conformational folding of

MA

mycobacterial methoxy- and ketomycolic acids facilitated by α-methyl trans-cyclopropane groups rather than cis-cyclopropane units. Microbiology, 159(11):2405–2415, November 2013.

D

[8] Wilma Groenewald, Mark S Baird, Jan A Verschoor, David E Minnikin,

PT E

and Anna K Croft. Differential spontaneous folding of mycolic acids from Mycobacterium tuberculosis. Chemistry and Physics of Lipids, 180:15–

CE

22, May 2014.

[9] E. K. Davies and E. K. Rideal. Interfacial Phenomena, page 265. Aca-

AC

demic Press, New York, 1961. [10] Takanori Takiue and Dieter Vollhardt. Miscibility of alkanol and fluoroalkanol in Langmuir film at the air/water interface. Colloids and Surfaces A, 198-200:797–804, February 2002.

26

ACCEPTED MANUSCRIPT

[11] Cox Jeffry S. Glickman, Michael S. and William S. Jacobs, Jr. A novel mycolic acid cyclopropane synthetase is required for cording, persistence, and virulence of Mycobacterium tuberculosis. Molecular cell, 5(4):717–

RI PT

727, April 2000. [12] Cahill Sean M. Glickman, Michael S. and William S. Jacobs, Jr. The Mycobacterium tuberculosis cmaA2 gene encodes a mycolic acid transcyclopropane synthetase.

J. biol. Chem, 276(3):2228–2233, January

SC

2001.

NU

[13] C´ecile Asselineau, Jean Asselineau, Gilbert Lan´eelle, and MarieAntoinette Lan´eelle. The biosynthesis of mycolic acids by Mycobac-

MA

teria: current and alternative hypotheses. Progress in lipid research, 41(6):501–523, November 2002.

[14] Lynn G Dover, Ana M Cerde˜ no-T´arraga, Mark J Pallen, Julian Parkhill,

D

and Gurdyal S Besra. Comparative cell wall core biosynthesis in the my-

PT E

colated pathogens, Mycobacterium tuberculosis and Corynebacterium diphtheriae. FEMS microbiology reviews, 28(2):225–250, May 2004.

CE

[15] Kuni Takayama, Cindy Wang, and Gurdyal S Besra. Pathway to synthesis and processing of mycolic acids in Mycobacterium tuberculosis.

AC

Clinical microbiology reviews, 18(1):81–101, January 2005. [16] Hedia Marrakchi, Marie-Antoinette Lan´eelle, and Mamadou Daff´e. Mycolic acids: structures, biosynthesis, and beyond. Chemistry & biology, 21(1):67–85, January 2014.

27

ACCEPTED MANUSCRIPT

[17] Luke J Alderwick, Eva Radmacher, Mathias Seidel, Roland Gande, Paul G Hitchen, Howard R Morris, Anne Dell, Hermann Sahm, Lothar Eggeling, and Gurdyal S Besra. Deletion of Cg-emb in corynebacterianeae leads to a novel truncated cell wall arabinogalactan, whereas

RI PT

inactivation of Cg-ubiA results in an arabinan-deficient mutant with a cell wall galactan core. J. biol. Chem, 280(37):32362–32371, September 2005.

SC

[18] Luke J Alderwick, Mathias Seidel, Hermann Sahm, Gurdyal S Besra, and Lothar Eggeling. Identification of a novel arabinofuranosyltrans-

NU

ferase (AftA) involved in cell wall arabinan biosynthesis in Mycobacterium tuberculosis. J. biol. Chem, 281(23):15653–15661, June 2006.

MA

[19] Mathias Seidel, Luke J Alderwick, Helen L Birch, Hermann Sahm, Lothar Eggeling, and Gurdyal S Besra.

Identification of a novel

D

arabinofuranosyltransferase AftB involved in a terminal step of cell

PT E

wall arabinan biosynthesis in Corynebacterianeae, such as Corynebacterium glutamicum and Mycobacterium tuberculosis. J. biol. Chem, 282(20):14729–14740, May 2007.

CE

[20] Helen L Birch, Luke J Alderwick, Apoorva Bhatt, Doris Rittmann, Karin Krumbach, Albel Singh, Yu Bai, Todd L Lowary, Lothar Eggeling,

AC

and Gurdyal S Besra. Biosynthesis of mycobacterial arabinogalactan: identification of a novel alpha(1–¿3) arabinofuranosyltransferase. Molecular Microbiology, 69(5):1191–1206, September 2008.

ˇ [21] Henrieta Skovierov´ a, G´erald Larrouy-Maumus, Jian Zhang, Devinder Kaur, Nathalie Barilone, Jana Kordul´akov´a, Martine Gilleron, 28

ACCEPTED MANUSCRIPT

St´ephanie Guadagnini, Martina Belanov´a, Marie-Christine Prevost, Brigitte Gicquel, Germain Puzo, Delphi Chatterjee, Patrick J Brennan, J´erˆome Nigou, and Mary Jackson. AftD, a novel essential arabinofuranosyltransferase from mycobacteria. Glycobiology, 19(11):1235–1247,

RI PT

November 2009.

[22] R B Raad, X M´eniche, C De Sousa-D’Auria, Mohamed Chami, Christophe Salmeron, Marielle Tropis, Cecile Labarre, Mamadou Daff´e,

SC

Christine Houssin, and Nicolas Bayan. A deficiency in arabinogalactan biosynthesis affects Corynebacterium glutamicum mycolate outer mem-

NU

brane stability. Journal of Bacteriology, 192(11):2691–2700, June 2010.

MA

[23] Anil K Ojha, Anthony D Baughn, Dhinakaran Sambandan, Tsungda Hsu, Xavier Trivelli, Yann Guerardel, Anuradha Alahari, Laurent Kremer, William R Jacobs, and Graham F Hatfull. Growth of Mycobac-

D

terium tuberculosis biofilms containing free mycolic acids and harbour-

2008.

PT E

ing drug-tolerant bacteria. Molecular Microbiology, 69(1):164–174, July

[24] Dhinakaran Sambandan, Dee N Dao, Brian C Weinrick, Catherine

CE

Vilch`eze, Sudagar S Gurcha, Anil Ojha, Laurent Kremer, Gurdyal S Besra, Graham F Hatfull, and William R Jacobs. Keto-mycolic acid-

AC

dependent pellicle formation confers tolerance to drug-sensitive Mycobacterium tuberculosis. mBio, 4(3):e00222–13, May 2013.

[25] Jiongwei Pan, Nagatoshi Fujiwara, Shiro Oka, Ryoji Maekura, Takeshi Ogura, and Ikuya Yano. Anti-Cord Factor (Trehalose 6, 6’-Dimycolate)

29

ACCEPTED MANUSCRIPT

IgG Antibody in Tuberculosis Patients Recognizes Mycolic Acid Subclasses. MICROBIOLOGY and IMMUNOLOGY, 43(9):863–869, 1999. [26] Mervyn Beukes, Yolandy Lemmer, Madrey Deysel, Jumaa R AL DU-

RI PT

LAYYMI, R Al Dulayymi Juma’a, Mark S Baird, S Baird Mark, Gani Koza, Maximiliano M Iglesias, M Iglesias Maximiliano, Richard R Rowles, R Rowles Richard, Cornelia Theunissen, Johan Grooten, Gianna Toschi, Vanessa V Roberts, V Roberts Vanessa, Lynne Pilcher,

SC

Sandra Van Wyngaardt, Nsovo Mathebula, Mohammed Balogun, Anton C Stoltz, C Stoltz Anton, and Jan A Verschoor. Structure-function

NU

relationships of the antigenicity of mycolic acids in tuberculosis patients.

MA

Chemistry and Physics of Lipids, 163(8):800–808, November 2010. [27] Y Yuan, Y Zhu, D D Crane, and C E Barry. The effect of oxygenated mycolic acid composition on cell wall function and macrophage growth in

PT E

September 1998.

D

Mycobacterium tuberculosis. Molecular Microbiology, 29(6):1449–1458,

[28] Jan A Verschoor, Mark S Baird, and Johan Grooten. Towards understanding the functional diversity of cell wall mycolic acids of Mycobac-

CE

terium tuberculosis. Progress in lipid research, 51(4):325–339, October

AC

2012.

[29] C. Asselineau and J. Asselineau. St´er´eochimique de l’acide corynomycolique. Bulletin de la Soci´et´e Chimique de France, pages 1992–1999, 1966.

30

ACCEPTED MANUSCRIPT

´ [30] J. F. Tocanne and C. Asselineau. Etude st´er´eochimique des acides aliphatiques α-ramifi´es β-hydroxyl´es. configuration de l’acide corynomycolique. Bulletin de la Soci´et´e Chimique de France, pages 4519–4525,

RI PT

1968. [31] Masaaki Seki, Ikuro Honda, Isao Fujita, Ikuya Yano, Saburo Yamamoto, and Akira Koyama. Whole genome sequence analysis of Mycobacterium bovis bacillus Calmette-Gu´erin (BCG) Tokyo 172: a comparative study

SC

of BCG vaccine substrains. Vaccine, 27(11):1710–1716, March 2009.

NU

[32] SAGE Working Group on BCG and WHO Secretariat. Report on BCG vaccine use for protection against mycobacterial infections including tu-

fections, October 2017.

MA

berculosis, leprosy, and other nontuberculous mycobacteria (NTM) in-

[33] Tania Cernuschi, Stefano Malvolti, Emily Nickels, and Martin Friede.

D

Bacillus Calmette-Gu´erin (BCG) vaccine: A global assessment of de-

PT E

mand and supply balance. Vaccine, 36(4):498–506, January 2018. [34] Juma’a R Al Dulayymi, Mark S Baird, and Evan ROBERTS. The syn-

CE

thesis of a single enantiomer of a major α-mycolic acid of Mycobacterium tuberculosis. Chemical Communications, 0(2):228–229, December 2002.

AC

[35] Juma’a R Al Dulayymi, Mark S Baird, and Evan Roberts. The synthesis of a single enantiomer of a major α-mycolic acid of M. tuberculosis. Tetrahedron, 61(50):11939–11951, 2005.

[36] Juma’a R Al Dulayymi, Mark S Baird, Hayder Mohammed, Evan Roberts, and William Clegg.

The synthesis of one enantiomer of 31

ACCEPTED MANUSCRIPT

the α-methyl-trans-cyclopropane unit of mycolic acids. Tetrahedron, 62(20):4851–4862, May 2006. [37] Juma’a R Al Dulayymi, Mark S Baird, Roberts, and David E Minnikin.

RI PT

The synthesis of single enantiomers of meromycolic acids from mycobacterial wax esters. Tetrahedron, 62(51):11867–11880, October 2006. [38] Gani Koza and Mark S Baird. The first synthesis of single enantiomers

AC

CE

PT E

D

MA

NU

SC

of ketomycolic acids. Tetrahedron Letters, 48:2165–2169, 2007.

32

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI PT

Graphical abstract

ACCEPTED MANUSCRIPT Highlights Mixed monolayers of well-defined trans-cyclopropane content were prepared.



Surface tension measurements allowed detailed study of mixed MA monolayers



-methyl trans-cyclopropane promotes W-shape conformation of MAs.



-methyl trans-cyclopropane promotes rigid MA monolayer formation.

AC

CE

PT E

D

MA

NU

SC

RI PT