| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departamento de Genética y Microbiología, Facultad de Medicina, Universidad de Murcia, Campus Universitario de Espinardo, 30100 Murcia, Spain
Correspondence
Pedro L. Valero-Guillén
plvalero{at}um.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Funke et al., 1997). Recently, several new taxa have been recognized with pathogenic potential in immunocompromised patients (Funke & Bernard, 1999; Funke et al., 1997).
Corynebacterium amycolatum, Corynebacterium jeikeium and Corynebacterium urealyticum are inhabitants of human skin and the most frequently isolated corynebacteria in clinical samples (Funke & Bernard, 1999). They may cause a variety of opportunistic infections, mainly in patients with underlying conditions (Funke & Bernard, 1999; Funke et al., 1997). Several reports have implicated C. urealyticum in urinary tract infections (Soriano et al., 1990), and it is now clear that infections by C. amycolatum are frequently catheter-related (Esteban et al., 1999; Funke & Bernard, 1999). Strains of these species are often multiantibiotic resistant (Funke & Bernard, 1999; Funke et al., 1997), and the establishment of this phenotype is important for the treatment of infections produced in immunocompromised patients.
Biochemical and/or genomic variation has been documented in C. amycolatum and C. jeikeium (Aubel et al., 1997; Funke et al., 1996, 1997; Wauters et al., 1996), leading in the past to misidentification with other species within the genus. Nowadays, clinically relevant Corynebacterium species can be identified by biochemical tests, complemented by chemical analysis of cell constituents (Funke & Bernard, 1999).
In chemotaxonomic studies, fatty acids can be easily determined by GLC and/or GLC-MS, whereas phospholipids are currently evaluated by TLC. In recent years, FAB-MS has been recommended as an accurate and rapid technique to analyse phospholipids of a great variety of micro-organisms (Cole & Enke, 1991). In general, MS improves on the data obtained by TLC and offers major possibilities of applications in systematic and biochemical investigations of these compounds. A few studies employing FAB-MS have dealt with phospholipids of the genus Corynebacterium (Niepel et al., 1998; Yagüe et al., 1997). To gain more insight into the phospholipid composition of corynebacteria we have applied MS (FAB and ESI) to the study of these compounds in some multiantibiotic resistant and clinically relevant species of Corynebacterium.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Strains of C. amycolatum were differentiated from Corynebacterium kroppenstedtii and Turicella otitidis by fatty acid profile and several biochemical tests (Collins et al., 1998). The laboratory strain Mycobacterium tuberculosis 11912 (grown on Löwenstein–Jensen medium for 4 weeks at 37 °C) was employed as a phospholipid control, mainly in the analysis of triacylphosphatidylinositol dimannosides (Ac3PIM2).
Phospholipid analysis.
Corynebacteria were cultivated for 72 h at 37 °C in Mueller–Hinton agar, supplemented with 0·2 % (v/v) Tween 80 in the cases of C. urealyticum and C. jeikeium due to the fact that these species are lipophilic and do not grow, or grow poorly, without this supplement. After harvesting, the strains were successively extracted for 2 h at 4 °C with chloroform/methanol at 1 : 2 (v/v), 1 : 1 (v/v) and 2 : 1 (v/v). The extracts were combined, evaporated to dryness with nitrogen and finally redissolved in 0·2 ml chloroform/methanol (1 : 1, v/v). This extract was partitioned in chloroform/methanol/water (8 : 4 : 2, by vol.) and the lower (chloroformic) phase collected, evaporated to dryness with nitrogen, redissolved in chloroform and applied to a small silica gel 60 (70–230 mesh) (Merck) column that was successively irrigated with chloroform, acetone and methanol. The methanol fraction (containing phospholipids) was evaporated to dryness and examined by one-dimensional TLC (20x20 aluminium sheets, silica gel 60 F254 plates; Merck) employing chloroform/methanol/water (60 : 30 : 6, by vol.) as solvent to determine the general composition of the isolates. Specific reagents for lipids (7 %, w/v, molybdophosphoric acid in ethanol), sugars (0·1 %, w/v, orcinol in 40 %, v/v, sulphuric acid), phosphorus (Dittmer–Lester spray) and amino groups (0·2 %, w/v, ninhydrin in acetone) were used (Minnikin et al., 1977), together with authentic standards of diphosphatidylglycerol (DPG), phosphatidylglycerol (PG), phosphatidylethanolamine (PE) and phosphatidylinositol (PI) (Sigma).
All phospholipid fractions were analysed by negative FAB-MS, employing m-nitrobenzyl alcohol as matrix and a Fisons VG AutoSpec mass spectrometer provided with a caesium gun at 25 kV. The acceleration voltage was 8 kV, the m/z range between 200 and 2400, and the resolution 1/1000. The spectra were obtained during 20 min, with 2 s of time scan and a delay of 0·5 s. In general, 10–15 scans were signal averaged.
Tandem mass spectrometry (MS-MS) analyses of Ac3PIM2 present in the extracts of C. amycolatum 14806, C. jeikeium 17 and C. urealyticum ATCC 43042T were carried out in an MSD-IonTrap-VL (Agilent) instrument, operated in the negative mode and with direct infusion of the sample (0·6 ml h-1). The electrospray needle was set at 3·5 kV and compressed N2 was employed as nebulization gas (69 kPa, 325 °C). Full-scan mass spectra were recorded over the mass range m/z 100–1600 at 3–5 s.
Purified acylphosphatidylglycerol (APG) and phosphatidylglycerol (PG) from C. amycolatum NCFB 2678T, C. jeikeium ATCC 43754T and C. urealyticum ATCC 43042T were obtained using preparative TLC (20x20 cm TLC plastic sheets, silica gel 60 F254; Merck) and chloroform/methanol/water (60 : 30 : 6, by vol.) as solvent. The phospholipids were detected with iodine vapour, scraped from the gel and extracted with chloroform/methanol (1 : 1, v/v) (overnight, 4 °C). Silica gel was eliminated by centrifugation (1500 g for 15 min) and the organic extract evaporated to dryness and analysed by FAB-MS as above. In a similar way, mixtures of PI+Ac3PIM2 were obtained from M. tuberculosis 11912 and C. urealyticum ATCC 43042T, and analysed as above. Alternatively, these mixtures were studied by 1H-NMR in a 400 MHz Brucker spectrometer, with the sample in deuterochloroform/deuteromethanol (2 : 1, v/v) (room temperature). Data from the bibliography assisted in the assignation of the different proton resonances (Gilleron et al., 1999; Niepel et al., 1998).
Phospholipid fatty acid analysis.
All phospholipid fractions and the mixture PI+Ac3PIM2 from C. urealyticum ATCC 43042T were subjected to acid methanolysis (overnight, 75 °C) in 4 ml methanol/toluene/sulphuric acid (30 : 15 : 1, by vol.) (Minnikin et al., 1980). The released fatty acid methyl esters were extracted with hexane, and analysed by GLC and GLC-MS. GLC was carried out in a KNK-3000 HRGLC gas chromatograph (Konik group) fitted with a DB1 column (15 m, 1 µm film), programmed from 100 °C to 270 °C at 6 °C min-1. GLC-MS was performed in a Hewlett Packard GLC (HP 5890)-MS (HP 5790) system provided with an HP-5 capillary column (25 m, 0·25 µm film), programmed from 60 °C to 280 °C at 10 °C min-1. GLC-MS with selected ion monitoring was performed in the GLC-MS system in the conditions described above, with multiple detection at m/z 310 (methyl esters of 10-methyloctadecenoic, 10-methyleneoctadecanoic acid and nonadecenoic acid – see below) and m/z 312 (methyl ester of tuberculostearic acid, TBS) (Larsson et al., 1981).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Some of them were presumptively identified on the basis of chromatographic behaviour and detection with several reagents. Thus, APG (a phospholipid related to PG containing an acyl group on the glycerol polar head), DPG, PG, PI and Ac3PIM2 were present in all the corynebacteria studied. Accordingly, FAB-MS (Fig. 2) identified several molecular classes compatible with PG at (M–H)-/z 745–775, PI at (M–H)-/z 833–877, APG at (M–H)-/z 1009–1041, and DPG and Ac3PIM2 at (M–H)-/z 1395–1459. The molecular masses of these compounds were deduced from the carboxylate composition obtained for each species examined (Table 1), and the distribution of the major molecular classes detected in the species studied is given in Table 2.
|
|
|
|
), C. jeikeium (Fig. 2b) and C. urealyticum (Fig. 2c), respectively. Most molecular classes predicted contain an even number of carbon atoms; however, in the case of C. amycolatum and C. urealyticum, significant amounts of PG and APG with an odd number of carbon atoms could be formulated due to the presence of C17 and C19 fatty acyl chains, respectively, in these species (Table 1).
In agreement with previous data for C. amycolatum (Yagüe et al., 1997), the predominant APG of this species appeared at (M–H)-/z 1011 (Fig. 2a), whereas in C. jeikeium and C. urealyticum the major APGs were found at (M–H)-/z 1037 (Fig. 2b) and (M–H)-/z 1009 (Fig. 2c), respectively. A combination of two octadecenoyls and one hexadecanoyl was demonstrated for C. amycolatum (Yagüe et al., 1997), with one octadecenoyl on the glycerol polar head. APGs from C. jeikeium and C. urealyticum seemed to follow a similar pattern of substitution, since FAB mass spectra of these compounds, purified by TLC from the respective type strains, presented fragments at m/z 435 (HPO3–glycerol–C18 : 1), m/z 475 (CH2=CH–CH2–PO3–glycerol–C18 : 1) and m/z 491 (CH2(O)CH–CH2–PO3–glycerol–C18 : 1) (see inset of Fig. 2b), defining the polar head of these molecules (Yagüe et al., 1997). Hence, the major APG of C. jeikeium contained three C18 : 1 fatty acyl moieties, and that of C.urealyticum two C18 : 1 fatty acyl moieties and one C16 : 1 (Table 2).
Some of the FAB-MS data from purified PG and APG indicated that the sn-1 position should be preferentially occupied by a C18 : 1 fatty acyl moiety in the three species studied, as revealed by an ion at m/z 417 [(R1–COO–CH=CH–CH2O–PO3H)-, R1=C18 : 1)] in the mass spectra of these molecules (see inset of Fig. 2b). This ion was generated by a preferential loss of the fatty acyl moiety located at the sn-2 position (Murphy & Harrison, 1994) from the different glycerophosphatidic acid ions [R1–COO–CH2–CH(OOR2)–CH2O–PO3H]- derived from PG and APG. The loss of fatty acyl moieties at position sn-1 produced other minor fragments of lower intensities that did not emerge from the background of the mass spectra (see inset of Fig. 2b). Glycerophosphatidic acid ions were found mostly at m/z 673 (formulated with R1, R2=C16 : 0, C18 : 1 fatty acyl moieties) and m/z 701 (C18 : 0+C18 : 1) in C. amycolatum, at m/z 671 (C16 : 1+C18 : 1) and m/z 699 (C18 : 1+C18 : 1) in C. jeikeium, and at m/z 671 (C16 : 1+C18 : 1) in C. urealyticum (see Fig. 2a, b, c).
The different molecular species of PIs from C. amycolatum and C. jeikeium should present patterns of substitution similar to those described for their respective PGs and APGs (Table 2), with major pseudomolecular ions situated at (M–H)-/z 835 (C16 : 0+C18 : 1) and (M–H)-/z 861 (C18 : 1+C18 : 1), respectively (Fig. 2a,b). In C. urealyticum signals at (M–H)-/z 847, (M–H)-/z 849 and (M–H)-/z 875 (Fig. 2c) predominated, thus defining a family of PIs where a C19 : 1 fatty acyl moiety should combine, respectively, with C16 : 1, C16 : 0 and C18 : 1 fatty acyl moieties. This finding was in general agreement with the carboxylate content found for the PI (+Ac3PIM2, see below) purified from C. urealyticum ATCC 43042T, whose composition was as follows: m/z 253 (C16 : 1), 15 %; m/z 255 (C16 : 0), 12 %; m/z 279 (C18 : 2), 8 %; m/z 281 (C18 : 1), 27 %; m/z 283 (C18 : 0), 3 %; m/z 295 (C19 : 1), 28 %, and m/z 297 (C19 : 0), 7 %. The diversity of PI in this species seemed to be higher, because compounds at (M–H)-/z 833 (C16 : 1+C18 : 1), (M–H)-/z 835 (C16 : 0+18 : 1), (M–H)-/z 851 (C19 : 0+C16 : 0), (M–H)-/z 861 (C18 : 1+C18 : 1), (M–H)-/z 873 (C18 : 1+C18 : 2) and (M–H)-/z 877 (C19 : 1+C18 : 0) were also detected in variable amounts. However, significant glycerophosphatidic ions appeared only at m/z 671 (C16 : 1+C18 : 1), m/z 685 (C16 : 1+C19 : 1), m/z 687 (C16 : 0+C19 : 1) and m/z 713 (C18 : 1+C19 : 1) in the FAB mass spectrum of PI (+Ac3PIM2) from C. urealyticum ATCC 43042T (not shown)
Preferential loss of the fatty acyl moiety at the sn-2 position (Murphy & Harrison, 1994) from the major PIs of C. urealyticum ATCC 43042T [(M–H)-/z 847, (M–H)-/z 849 and (M–H)-/z 875] led to the generation of the fragments located at m/z 593 and m/z 611 (593+18 mass units) (not shown). This suggests that, when present, C19 : 1 should be at position sn-1. This finding was supported by the existence of a fragment at m/z 431 in the mass spectrum (not shown), formulated as (R1–COO–CH=CH–CH2O–PO3H)-, with R1=C19 : 1, and derived by loss of C16 : 1, C16 : 0 and C18 : 1 (all of them at the sn-2 position) from, respectively, m/z 685, m/z 687 and m/z 713.
As mentioned above, pseudomolecular ions compatible with DPG and Ac3PIM2 were seen around (M–H)-/z 1400 (Fig. 2). DPGs from C. amycolatum predominated at (M–H)-/z 1403 (2C16 : 0+2C18 : 1) and (M–H)-/z 1431 (2C16 : 0+C18 : 1+C18 : 0) (Fig. 2a). As a whole, C. jeikeium presented two major DPGs at (M–H)-/z 1429 (3C18 : 1+C16 : 0) and (M–H)-/z 1455 (4C18 : 1), although the display of these molecules was broader in this species (Table 2; Fig. 2b). DPGs from C. urealyticum were mostly detected at (M–H)-/z 1399 (2C16 : 1+2C18 : 1) and (M–H)-/z 1401 (C16 : 1+C16 : 0+2C18·1) (Fig. 2c, Table 2).
As already mentioned, spots on TLC stained with reagents for lipids, sugars and phosphorus were compatible with Ac3PIM2 and, in agreement with previous observations (Dobson et al., 1985), they migrated together with PI in the solvent system employed. In this way a mixture of these compounds was obtained from C. urealyticum ATCC 43042T and from M. tuberculosis 11912 by preparative TLC. The samples were analysed by FAB-MS, and the presence of Ac3PIM2 in the mixture subsequently assessed by NMR. Signals for anomeric (H-1) protons of the two mannoses were centred at 5·03 p.p.m. and 5·09 p.p.m., whereas the inositol protons appeared at 4·16 p.p.m. (attributed to H-1), 4·18 p.p.m. (H-2), 3·77 p.p.m. (H-3), 3·89 p.p.m. (H-4), 3·21 p.p.m. (H-5) and 3·91 p.p.m. (H-6); the glycerol protons were detected at 4·42 p.p.m. (H-1a), 4·13 p.p.m. (H-1b), 5·23 p.p.m. (H-2), and 3·97 p.p.m. (H-3a and H-3b) (Gilleron et al., 1999; Niepel et al., 1998). The ratio H-2 of glycerol to anomeric protons of mannoses was approximately 4 for M. tuberculosis 11912 and approximately 5 for C. urealyticum ATCC 43042T, indicating a higher abundance of PI in the extracts obtained. Other phospholipids present in M. tuberculosis 11912 were identified as PE [(M–H)-/z 716 (C18 : 1+C16 : 0), (M–H)-/z 732 (C19 : 0+C16 : 0) and (M–H)-/z 744 (C18 : 1+C18 : 0), PG [(M–H)-/z 745 (C18 : 1+C16 : 1)], PI [(M–H)-/z 851 (C16 : 0+C19 : 0)] and DPG [(M–H)-/z 1401 (C16 : 1+C16 : 0+2C18 : 1), (M–H)-/z 1403 (2C16 : 0+2C18 : 1), (M–H)-/z 1429 (C16 : 0+3C18 : 1) and (M–H)-/z 1431 (C16 : 0+2C18 : 1+C18 : 0)]. Ac3PIM2 from M. tuberculosis 11912 were attributed to an ion at (M–H)-/z 1413 (2C16 : 0+C19 : 0; C19 : 0=10-methyloctadecanoic acid, or tuberculostearic acid – TBS), which was consistent with the major Ac3PIM2 reported for this species (Gilleron et al., 2001; Khoo et al., 1995; Schaeffer et al., 1999).
Molecular classes of Ac3PIM2 in C. amycolatum were attributed to (M–H)-/z 1425 (C16 : 0+C18 : 1+C18 : 0) and (M–H)-/z 1453 (C18 : 1+2C18 : 0). MS-MS experiments with both pseudomolecular ions revealed a prominent fragment at m/z 831 (Fig. 3b, fragment C), which was interpreted as belonging to a compound formulated as R3O–mannose–inositol(PO3-)–mannose (R3=C18 : 0 in Fig. 3a), according to the general fragmentation pattern of Ac3PIM2 (Gilleron et al., 2001; Schaeffer et al., 1999). In the MS-MS spectrum of (M–H)-/z 1425 (Fig. 3a), the presence of a significant fragment at m/z 1169 (Fig. 3b, fragment A; loss of R2OH; R2=C16 : 0 in Fig. 3a), together with m/z 417 (see above), was indicative that the sn-2 position of the glycerol was occupied by C16 : 0. Thus, in such a compound, a mannose residue is esterified by a C18 : 0, while C18 : 1 and C16 : 0 are, respectively, at positions sn-1 and sn-2 of the glycerol moiety. In a similar way, an additional significant fragment at m/z 1169 (loss of C18 : 0) in the MS-MS spectrum of (M–H)-/z 1453 (not shown) revealed a C18 : 0 at position sn-2. Therefore, in the last compound the remaining C18 : 1 and C18 : 0 are, respectively, at position sn-1 and in a mannose residue.
|
; Fig. 2a, b). The MS-MS spectrum of (M–H)-/z 1423 contained prominent fragments at m/z 803 (Fig. 3c, fragment C) (indicative of a C16 : 0 on a mannose residue: see Gilleron et al., 2001; Schaeffer et al., 1999), m/z 829 (Fig. 3c, fragment C) (a C18 : 1 on a mannose residue), m/z 1141 (Fig. 3c, fragments A, D; loss of C18 : 1) and m/z 1167 (Fig. 3c, fragments A, D; loss of C16 : 0). As a whole, these fragments were indicative of the existence of two molecular species for (M–H)-/z 1423, one of them with C16 : 0 on a mannose residue (R3=C16 : 0 in Fig. 3a), C18 : 1 at position sn-1 (R1=C18 : 1 in Fig. 3a) and C18 : 1 at position sn-2 (R2=C18 : 1 in Fig. 3a). The other molecular species appears to show a C18 : 1 on a mannose residue (R3=C18 : 1 in Fig. 3a) and C18 : 1 and C16 : 0 at, respectively, the sn-1 (R1=C18 : 1 in Fig. 3a) and sn-2 (R2=C16 : 0 in Fig. 3a) positions. Similarly, (M–H)-/z 1451 (not shown) contained two molecular species formulated as above, with C16 : 0 being replaced by C18 : 0.
Data on the Ac3PIM2 content of C. urealyticum were partially informative when the total phospholipid extract was examined (Fig. 2c), and better results were obtained by FAB-MS of the mixture (PI+Ac3PIM2) obtained by preparative TLC. Thus, pseudomolecular ions at (M–H)-/z 1395 (C16 : 1+C16 : 0+C18 : 1), (M–H)-/z 1411 (2C16 : 0+C19 : 1), (M–H)-/z 1423 (C16 : 0+2C18 : 1) and (M–H)-/z 1439 (C16 : 0+C18 : 0+C19 : 1) were detected in C. urealyticum ATCC 43042T (not shown), in agreement with the values predicted from the PI content of this species (Table 2). Also it was noteworthy that a major Ac3PIM2 from C. urealyticum 154 was detected at (M–H)-/z 1441 (C16 : 0+C18 : 0+C19 : 0) and that this strain contained high levels of C19 : 0 (TBS) (the ratio C19 : 0/C19 : 1 in the carboxylate content of the total phospholipid extract was approximately 5). Accordingly, the major PI of this strain were detected at (M–H)-/z 877 (C18 : 0+C19 : 0), also reaching high amounts (M–H)-/z 849 (C16 : 0+C19 : 1) and (M–H)-/z 851 (C16 : 0+C19 : 0); a significant fragment at m/z 433 detected in the mass spectrum of this strain should indicate that C19 : 0 is located at the sn-1 position of the glycerol moiety.
In (M–H)-/z 1439 from C. urealyticum ATCC 43042T the existence of a C16 : 0 fatty acyl moiety on a mannosyl residue of the polar head (Fig. 4a) was verified by the presence of two related significant fragments at m/z 803 and m/z 859 (Gilleron et al., 2001; Schaeffer et al., 1999) in the MS-MS spectrum (Fig. 4b). Moreover, fragments at m/z 1155 (loss of C18 : 0), m/z 1143 (loss of C19 : 1) (with lower relative abundance than m/z 1155) and m/z 431 (Fig. 4b) [which further fragments into m/z 153 (CH2=COH–CH2–PO3-) and m/z 295 (C19 : 1), not shown] indicated that C19 : 1 and C18 : 0 occupied, respectively, positions sn-1 and sn-2. Additional minor fragments at m/z 1181, m/z 1038 and m/z 1068 were in accordance with the formulation given in Fig. 4(a).
|
Phospholipid fatty acids
Phospholipid fatty acids of the strains studied ranged from 14 to 19 carbon atoms, the major ones being octadecenoic and hexadecanoic (Table 3). Saturated fatty acids were tetradecanoic (C14 : 0), pentadecanoic (C15 : 0), hexadecanoic (C16 : 0), heptadecanoic (C17 : 0) and octadecanoic (C18 : 0), and the mass spectra of all of them presented a base peak at m/z 74 with less intense molecular ions at, respectively, m/z 242, m/z 256, m/z 270, m/z 284 and m/z 298. Unsaturated fatty acids were identified as hexadecenoic (C16 : 1, m/z 268), heptadecenoic (C17 : 1, m/z 282), octadecadienoic (C18 : 2, m/z 294), octadecenoic (C18 : 1, m/z 296) and nonadecenoic (C19 : 1, m/z 310). The base peak of unsaturated fatty acids was m/z 55 (m/z 67 in the case of C18 : 2), being accompanied by fragments at m/z - 32 (loss of methanol) or m/z - 31 (loss of –OCH3).
|
). One of them was characterized as 10-methyloctadecanoic acid (tuberculostearic acid, TBS) (m/z 312, base peak at m/z 74, and diagnostic fragments at m/z 199 and m/z 167). A second compound was identified as 10-methyleneoctadecanoic acid because it showed the same chromatographic features as characterized previously in C. urealyticum (Couderc et al., 1991). This compound eluted just after TBS (not shown) and its mass spectrum (Fig. 5) was dominated by a base peak at m/z 55 (not shown), characteristic of unsaturated fatty acids. The molecular ion of this compound was situated at m/z 310 and accompanied by one at m/z 279 (M/z - 31). Important fragments for this acid were detected at m/z 197, m/z 165, m/z 153, m/z 157 and m/z 125, probably related to the presence of a methylene group at position C-10 (Fig. 5). An identical mass spectrum (not shown) was revealed for a third compound eluting just after C18 : 0, which had characteristics of 10-methyloctadecenoate. However, no definitive identification could be given for this compound and further analyses are required to establish its precise structure, and particularly the location of the double bond. Thus, the C19 : 1 carboxylate ion (m/z 295) (Table 1) corresponded to 10-methyloctadecenoic, 10-methyleneoctadecanoic and nonadecenoic acid in the case of C. amycolatum, whereas in C. urealyticum it could be identified as 10-methyloctadecenoic and 10-methyleneoctadecanoic acids, the latter being, in general, predominant in inositol-containing phospholipids – PI+Ac3PIM2 – of C. urealyticum, as revealed by analysis of strain ATCC 43042T.
|
) by GLC or GLC-MS led us to determine more precisely if other MAS isolates did not really synthesize the three compounds detected in MAR strains. GLC-MS with selected ion monitoring revealed in all cases the presence in MAS isolates of fatty acids with retention times and m/z values identical to 10-methyloctadecenoic, TBS and 10-methyleneoctadecanoic acids. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
), this technique provided useful information on the different molecular classes of a variety of phospholipids and revealed a great complexity in the species studied, a general rule already noticed in bacteria (Dowhan, 1997). The whole picture obtained is in agreement with data previously obtained by TLC in members of the genus Corynebacterium (Collins & Cummins, 1986; Yagüe et al., 1997). Nevertheless, the species studied differed from each other, at least at the optimal conditions for in vitro growth, since their major molecular classes of APG, DPG, PG, PI and Ac3PIM2 were not identical with respect to the fatty acyls located in the glycerol moiety. Thus, a combination of C16 : 0+C18 : 1 predominated among APG, PG and DPG of C. amycolatum, whereas in C. jeikeium and C. urealyticum, the predominant combinations were based on C18 : 1+C18 : 1 and C16 : 1+C18 : 1, respectively. This rule was extended to PI and Ac3PIM2 of C. amycolatum and C. jeikeium, but in C. urealyticum the major molecular classes of both phospholipids contained a C19 : 1 fatty acyl moiety, eventually characterized as (mainly) 10-methyleneoctadecanoic acid. The identification of this compound relied on its retention time and mass spectrum, which were identical (as methyl ester) to those of a fatty acid previously reported in C. urealyticum and whose structure was clearly established (Couderc et al., 1991; Couderc, 1995). Indeed, it is an unusual fatty acid, but considered as a precursor of the synthesis of TBS in several species of the genus Mycobacterium (Akamatsu & Law, 1970; Couderc, 1995). The genetic and/or enzymic reasons for its accumulation in some strains of C. urealyticum are, at present, unknown. 10-Methyleneoctadecanoyl was also found in low levels in MAR C. amycolatum strains, but its distribution among the phospholipids of this species could not be established. On the other hand, TBS and another fatty acyl moiety, tentatively identified as 10-methyloctadecenoyl, were also detected in C. urealyticum and in MAR strains of C. amycolatum. It should be mentioned that the minor differences between MAR and MAS strains of C. amycolatum were reproducible but difficult to explain, although they could be the result of more profound changes in the biochemistry of MAR C. amycolatum, deserving further investigation.
Branched fatty acids are not very common in Corynebacterium, and with the exception of Corynebacterium bovis (Collins & Cummins, 1986) and C. urealyticum (Couderc et al., 1991; Herrera-Alcaraz et al., 1990) even TBS constitutes a minor component in several species of the genus (Bernard et al., 1991; Herrera-Alcaraz et al., 1990). A recent report, however, has demonstrated high levels of 9-methyl-10-octadecenoic acid among the phospholipids and glycolipids of C. variabilis (Niepel et al., 1998). Its molecular mass is also 310 (as methyl ester), although it coelutes with octadecanoic acid (Niepel et al., 1998), suggesting that it differs from 10-methyloctadecenoic acid detected by us in C. amycolatum and C. urealyticum. 9-Methyl-10-octadecanoic acid is found in the major molecular classes of APG, PG and DPG in C. variabilis (Niepel et al., 1998) and this feature distinguishes this species from the corynebacteria included in the present work. Moreover, the diversity of PI observed in C. amycolatum, C. jeikeium and C. urealyticum was broader than that reported for C. variabilis, where only PI at (M–H)-/z 851 was detected (Niepel et al., 1998). This molecular species, composed of C16 : 0+TBS, is identical to the only one detected in several members of the genus Mycobacterium (see, for example, Gilleron et al., 2001), and also appears in C. urealyticum, together with those based on 10-methyleneoctadecanoyl as we have established in the present study.
Although the species studied showed a rather similar general phospholipid composition, a noteworthy feature is the particular substitution pattern on the glycerol moiety suggested by MS data, which includes an unsaturated fatty acyl moiety (octadecenoyl) at the sn-1 position of PG and APG (and probably of DPG). This finding corrects a previous work by Yagüe et al. (1997) on APG, where octadecenoyl was thought to be at the sn-2 position. Thus, the general trend of substitution in PG and APG found in C. amycolatum, C. jeikeium and C. urealyticum is similar to that noticed in C. variabilis (Niepel et al., 1998), although this species contains 9-methyl-10-octadecenoyl at the sn-1 position, among the major molecular classes of phospholipids, instead of a C18 : 1. FAB-MS data obtained in this work, together with those revealed in MS-MS experiments of Ac3PIM2, also indicate that 10-methyleneoctadecanoyl, detected among major molecular classes of PI and Ac3PIM2 of C. urealyticum, should be also at the sn-1 position. This agrees with the observation made in mycobacteria indicating that synthesis of TBS, and its precursor 10-methyleneoctadecanoic acid, requires the previous esterification of oleic acid to a phospholipid (Akamatsu & Law, 1970).
A variety of functions, other than structural roles, have been attributed to phospholipids and, probably, the great molecular diversity reflects differential tasks in many cellular processes (Dowhan, 1997). In this context, the APG content of corynebacteria is striking (Niepel et al., 1998; Yagüe et al., 1997). Also intriguing is the high amount of 10-methyleneoctadecanoyl observed in some strains of C. urealyticum (Couderc et al., 1991) and it remains to be established if this compound confers some kind of advantage to this bacterium, as suggested for 9-methyl-10-octadecenoyl in the case of Corynebacterium variabilis (Niepel et al., 1998).
In this work, FAB-MS and MS-MS experiments provided convincing evidence for the presence of Ac3PIM2 in C. amycolatum, C. jeikeium and C. urealyticum. In general, phosphatidylinositol mannosides (PIMs) are widespread in mycobacteria and related micro-organisms, as revealed by TLC (Dobson et al., 1985; Minnikin et al., 1977). Structural analyses carried out in the past (see Minnikin, 1982) indicated that these compounds appear mainly as triacyl and tetracyl forms of both phosphatidylinositol dimannosides (PIM2) and phosphatidylinositol hexamannosides (PIM6). In recent years, detailed structural (Gilleron et al., 1999, 2001; Khoo et al., 1995; Nigou et al., 1999) and biosynthetic (Korduláková et al., 2002; Schaeffer et al., 1999) studies have been performed on PIMs and also on their possible role during the infectious processes of pathogenic species of Mycobacterium (Gilleron et al., 2001). Moreover, PI, the precursor in the synthesis of these compounds (Korduláková et al., 2002), is considered an essential phospholipid for M. tuberculosis (Jackson et al., 2000). The fine structure of Ac3PIM2 remains to be elucidated in Corynebacterium, but the molecular masses found for the species investigated are close to those detected in M. tuberculosis (Khoo et al., 1995) and in Mycobacterium bovis (Gilleron et al., 1999, 2001; Nigou et al., 1999). The partial analysis carried out in the present work appears to indicate that, like mycobacteria, the species investigated esterify a mannosyl residue with a saturated fatty acyl moiety (C16 : 0 or C18 : 0), a pattern also observed in Rhodococcus equi (Garton et al., 2002). However, MS-MS experiments also indicated that in C. jeikeium and C. urealyticum unsaturated fatty acyl moieties are similarly present. PIMs are structurally (Gilleron et al., 1999, 2001; Khoo et al., 1995; Nigou et al., 1999) and biosynthetically (Besra et al., 1997) related to lipoarabinomannans and lipomannans of the cell envelope of Mycobacterium, and according to recent evidence are probably present in Corynebacterium (Puech et al., 2001). Thus, it should be interesting to investigate in the future if some biological functions attributed to these molecules (Gilleron et al., 2001; Nigou et al., 2002) can be also shared by Ac3PIM2 and more complex lipopolysaccharides present in opportunistic pathogens of the genus Corynebacterium.
In conclusion, the pathogenic corynebacteria analysed presented the same type of phospholipids, but in spite of the overall similarity observed, FAB-MS allowed them to be distinguished on the basis of the fatty acyl substituents. More notable differences were related to inositol-containing phospholipids, and particularly striking was the high content of 10-methyleneoctadecanoic acid found among the major molecular classes of PI and Ac3PIM2 in C. urealyticum.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Aubel, D., Renaud, F. N. R. & Freney, J. (1997). Genomic diversity of several Corynebacterium species identified by amplification of the 16S-23S rRNA gene spacer regions. Int J Syst Bacteriol 47, 767–772.
Bernard, K. A., Bellefeuille, M. & Ewan, E. P. (1991). Cellular fatty acid composition as an adjunct to the identification of asporogenous, aerobic gram-positive rods. J Clin Microbiol 29, 83–89.
Besra, G. S., Morehouse, C. B., Rittner, C. M., Waechter, C. J. & Brennan, P. J. (1997). Biosynthesis of mycobacterial lipoarabinomannan. J Biol Chem 272, 18460–18466.
Cole, M. J. & Enke, C. G. (1991). Direct determination of phospholipid structures in microorganisms by fast atom bombardment triple quadrupole mass spectrometry. Anal Chem 63, 1032–1038.[Medline]
Collins, M. D. & Cummins, C. S. (1986). Genus Corynebacterium. In Bergey's Manual of Systematic Bacteriology, vol. 2, pp. 1266–1276. Edited by P. H. A. Sneath, N. A. Mair, M. E. Sharpe & J. G. Holt. Baltimore, MD: Williams & Wilkins.
Collins, M. D., Falsen, E., Akervall, E., Sjöden, B. & Álvarez, A. (1998). Corynebacterium kroppenstedtii sp. nov., a novel corynebacterium that does not contain mycolic acids. Int J Syst Bacteriol 48, 1449–1454.
Couderc, F. (1995). Gas chromatography/tandem mass spectrometry as an analytical tool for the identification of fatty acids. Lipids 30, 691–699.[Medline]
Couderc, F., de Briel, D., Demont, N., Gilard, V. & Promé, J. C. (1991). Mass spectrometry as a tool for identifying group D2 corynebacteria by their fatty acid profiles. J Gen Microbiol 137, 1903–1909.[Medline]
Dobson, G., Minnikin, D. E., Minnikin, S. M., Parlett, J. H., Goodfellow, M., Ridell, M. & Magnusson, M. (1985). Systematic analysis of complex mycobacterial lipids. In Chemical Methods in Bacterial Systematics, pp. 237–265. Edited by M. Goodfellow & D. E. Minnikin. London: Academic Press.
Dowhan, W. (1997). Molecular basis of membrane phospholipid diversity: why are there so many lipids? Annu Rev Biochem 66, 199–232.[CrossRef][Medline]
Esteban, J., Nieto, E., Calvo, R., Fernández-Roblas, R., Valero-Guillén, P. L. & Soriano, F. (1999). Microbiological characterization and clinical significance of Corynebacterium amycolatum strains. Eur J Clin Microbiol Infect Dis 18, 518–521.[CrossRef][Medline]
Funke, G. & Bernard, K. A. (1999). Coryneform gram-positive rods. In Manual of Clinical Microbiology, 7th edn, pp. 319–345. Edited by P. R. Murray, E. J. Baron, M. A. Faller, F. C. Tenover & R. H. Yolken. Washington, DC: American Society for Microbiology.
Funke, G., Lawson, P. A., Bernard, K. A. & Collins, M. D. (1996). Most Corynebacterium xerosis strains identified in the routine clinical laboratory correspond to Corynebacterium amycolatum. J Clin Microbiol 34, 1124–1128.[Abstract]
Funke, G., von Graevenitz, A., Clarridge, J. E., III & Bernard, K. A. (1997). Clinical microbiology of coryneform bacteria. Clin Microbiol Rev 10, 125–159.[Abstract]
Garton, N. J., Gilleron, M., Brando, T., Dan, H.-H., Giguère, S., Puzo, G., Prescott, J. F. & Sutcliffe, I. C. (2002). A novel lipoarabinomannan from the equine pathogen Rhodococcus equi. J Biol Chem 277, 31722–31733.
Gilleron, M., Nigou, J., Cahuzac, B. & Puzo, G. (1999). Structural study of the lipomannans from Mycobacterium bovis BCG: characterisation of multiacylated forms of the phosphatidyl-myo-inositol anchor. J Mol Biol 285, 2147–2160.[CrossRef][Medline]
Gilleron, M., Ronet, C., Mempel, M., Monsarrat, B., Gachelin, G. & Puzo, G. (2001). Acylation state of the phosphatidylinositol mannosides from Mycobacterium bovis bacillus Calmette Guérin and ability to induce granuloma and recruit natural killer T cells. J Biol Chem 276, 34896–34904.
Herrera-Alcaraz, E., Valero-Guillén, P. L., Martín-Luengo, F. & Soriano, F. (1990). Taxonomic implications of the chemical analysis of the D2 group of corynebacteria. FEMS Microbiol Lett 72, 341–344.
Jackson, M., Crick, D. C. & Brennan, P. J. (2000). Phosphatidylinositol is an essential phospholipid of mycobacteria. J Biol Chem 275, 30092–30099.
Khoo, K.-H., Dell, A., Morris, H. R., Brennan, P. J. & Chatterjee, D. (1995). Structural definition of acylated phosphatidylinositol mannosides from Mycobacterium tuberculosis: definition of a common anchor for lipomannan and lipoarabinomannan. Glycobiology 5, 117–127.
Korduláková, J., Gilleron, M., Mikusová, K., Puzo, G., Brennan, P. J., Gicquel, B. & Jackson, M. (2002). Definition of the first mannosylation step in phosphatidylinositol mannoside synthesis. pimA is essential for growth of mycobacteria. J Biol Chem 277, 31335–31344.
Larsson, L., Mardh, P. A., Odham, G. & Westerdahl, G. (1981). Use of selected ion monitoring for detection of tuberculostearic and C32 mycocerosic acid in mycobacteria and in five-day-old cultures of sputum specimens from patients with pulmonary tuberculosis. Acta Pathol Microbiol Scand B 89, 245–251.[Medline]
Minnikin, D. E. (1982). Lipids: complex lipids, their chemistry, biosynthesis and roles. In The Biology of the Mycobacteria, vol. 1, pp. 95–184. Edited by C. Ratledge & J. L. Stanford. London: Academic Press.
Minnikin, D. E., Patel, P. V., Alshamaony, L. & Goodfellow, M. (1977). Polar lipid composition of Nocardia and related bacteria. Int J Syst Bacteriol 27, 104–117.
Minnikin, D. E., Hutchinson, I. G., Caldicott, A. B. & Goodfellow, M. (1980). Thin layer chromatography of methanolysates of mycolic acid-containing bacteria. J Chromatogr 188, 221–233.[CrossRef]
Murphy, R. C. & Harrison, K. A. (1994). Fast atom bombardment mass spectrometry of phospholipids. Mass Spectrom Rev 13, 57–75.[CrossRef]
Niepel, T., Meyer, H., Wray, V. & Abraham, W.-R. (1998). Intraspecific variation of unusual phospholipids from Corynebacterium spp. containing a novel fatty acid. J Bacteriol 180, 4650–4657.
Nigou, J., Gilleron, M. & Puzo, G. (1999). Lipoarabinomannans: characterization of the multiacylated forms of the phosphatidyl-myo-inositol anchor by NMR spectroscopy. Biochem J 337, 453–460.
Nigou, J., Gilleron, M., Rojas, M., García, L. F., Thurnher, M. & Puzo, G. (2002). Mycobacterial lipoarabinomannans: modulators of dendritic cell function and the apoptotic response. Microbes Infect 4, 945–953.[CrossRef][Medline]
Puech, V., Chami, M., Lemassu, A., Lanéelle, M. A., Schiffler, B., Gounon, P., Bayan, N., Benz, R. & Daffé, M. (2001). Structure of the cell envelope of corynebacteria: importance of the noncovalently bound lipids in the formation of the cell wall permeability barrier and fracture plane. Microbiology 147, 1365–1382.
Schaeffer, M. L., Khoo, K.-H., Besra, G. S., Chatterjee, D., Brennan, P. J., Belisle, J. T. & Inamine, J. (1999). The pimB gene of Mycobacterium tuberculosis encodes a mannosyltransferase involved in lipoarabinomannan biosynthesis. J Biol Chem 274, 31625–31631.
Soriano, F., Aguado, J. M., Ponte, C., Fernández-Roblas, R. & Rodríguez-Tudela, J. L. (1990). Urinary tract infection caused by Corynebacterium group D2. Report of 82 cases and review. Rev Infect Dis 6, 1019–1034.
Wauters, G., Driessen, A., Ageron, E., Janssen, M. & Grimont, P. A. D. (1996). Propionic acid-producing strains previously designated as Corynebacterium xerosis, C. minutissimum, C. striatum, and CDC group I2 and group F2 coryneforms belong to the species Corynebacterium amycolatum. Int J Syst Bacteriol 46, 653–657.
Yagüe, G., Segovia, M. & Valero-Guillén, P. L. (1997). Acyl phosphatidylglycerol: a major phospholipid of Corynebacterium amycolatum. FEMS Microbiol Lett 151, 125–130.[Medline]
Received 24 December 2002;
revised 2 April 2003;
accepted 10 April 2003.
This article has been cited by other articles:
|
|
 |
|
  A. Tauch, O. Kaiser, T. Hain, A. Goesmann, B. Weisshaar, A. Albersmeier, T. Bekel, N. Bischoff, I. Brune, T. Chakraborty, et al. Complete Genome Sequence and Analysis of the Multiresistant Nosocomial Pathogen Corynebacterium jeikeium K411, a Lipid-Requiring Bacterium of the Human Skin Flora J. Bacteriol., July 1, 2005; 187(13): 4671 - 4682. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||
