• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbcAbout JBCASBMBSubmissionsSubscriptionsContactJBCThis Article
J Biol Chem. Jul 1, 2011; 286(26): 22769–22776.
Published online May 3, 2011. doi:  10.1074/jbc.M111.231316
PMCID: PMC3123044

The Acylation State of Surface Lipoproteins of Mollicute Acholeplasma laidlawii*An external file that holds a picture, illustration, etc.
Object name is sbox.jpg

Abstract

Acylation of the N-terminal Cys residue is an essential, ubiquitous, and uniquely bacterial posttranslational modification that allows anchoring of proteins to the lipid membrane. In Gram-negative bacteria, acylation proceeds through three sequential steps requiring lipoprotein diacylglyceryltransferase, lipoprotein signal peptidase, and finally lipoprotein N-acyltransferase. The apparent lack of genes coding for recognizable homologs of lipoprotein N-acyltransferase in Gram-positive bacteria and Mollicutes suggests that the final step of the protein acylation process may be absent in these organisms. In this work, we monitored the acylation state of eight major lipoproteins of the mollicute Acholeplasma laidlawii using a combination of standard two-dimensional gel electrophoresis protein separation, blotting to nitrocellulose membranes, and MALDI-MS identification of modified N-terminal tryptic peptides. We show that for each A. laidlawii lipoprotein studied a third fatty acid in an amide linkage on the N-terminal Cys residue is present, whereas diacylated species were not detected. The result thus proves that A. laidlawii encodes a lipoprotein N-acyltransferase activity. We hypothesize that N-acyltransferases encoded by genes non-homologous to N-acyltransferases of Gram-negative bacteria are also present in other mollicutes and Gram-positive bacteria.

Keywords: Bacteria, Fatty Acid, Lipoprotein, Mass Spectrometry (MS), Membrane Proteins, 2D PAGE

Introduction

Essential cellular activities such as transport, sensing, signal transduction, growth, adhesion, and pathogenesis require proteins localized on the cell surface. One of the strategies to anchor a protein at the lipid membrane is covalent modification of proteins by fatty acids or other lipids. All bacteria rely on acylation of N-terminal Cys residues of lipoproteins to allow tight anchorage on the membrane surface (1). This uniquely bacterial modification was first observed in 1969 in a major outer membrane protein of Escherichia coli called Braun's lipoprotein and later detected in every bacterial organism studied (2). Adhesion proteins and substrate-binding subunits of ATP-binding cassette (ABC)2 transporters are known to be attached to the membrane in this manner. Thousands of proteins predicted to be lipoproteins “by similarity” according to the presence of consensus sequences of Sec Type II or Tat Type II signal peptides are also thought to be acylated (3).

Bacteria of the Mollicutes class are widespread in nature and are characterized by the lack of a rigid cell wall and drastically reduced genome sizes. In fact, representatives of this class are the simplest self-replicating organisms able to independently subsist, generate energy, and adapt to changing environments (4 and references therein). Most mollicutes depend on cholesterol and are partially or totally incapable of fatty acid synthesis and therefore depend on the host or the culture medium for a constant supply of these substances (5). Acholeplasma laidlawii is a convenient model mollicute because of its relatively simple nutritional requirements and fast growth rate. A. laidlawii can synthesize limited amounts of saturated fatty acids when grown in a thoroughly lipid-depleted medium (6). Because of the absence of the periplasmic space, Mollicutes need an unusually large number of surface-exposed lipoproteins. The genome of A. laidlawii was recently determined in our laboratory (NCBI Project ID 19259). Among 325 predicted A. laidlawii membrane proteins, 62 have predicted Sec Type II signal peptides and can therefore be N-terminally acylated with long chain fatty acids to anchor on the cell surface.

A conserved motif in Type II signal peptides directs future lipoprotein export through the Sec or Tat pathway to the membrane and then to the lipoprotein biogenesis machinery (3, 7). In Gram-negative bacteria, N-terminal acylation proceeds in three steps (schematically represented in Fig. 1). (i) Prolipoprotein diacylglyceryltransferase (Lgt) uses membrane phospholipid substrates and catalyzes the transfer of a diacylglyceryl moiety onto a strictly conserved cysteine via a thioether linkage. (ii) The signal peptide is then cleaved by lipoprotein signal Type II peptidase (Lsp in E. coli), leaving the lipid-modified Cys at the newly formed N terminus of the mature lipoprotein. (iii) Lipoprotein N-acyltransferase (Lnt) adds a third fatty acid residue to the backbone amino group of lipidated cysteine. The last step is important for lipoprotein transport to the outer membrane of Gram-negative bacteria via the lipoprotein localization 1 pathway (8). Depletion of Lnt results in the accumulation of lipoproteins in the plasma membrane, which may be lethal to the cell.

FIGURE 1.
Pathway of protein acylation in Gram-negative bacteria. The three steps leading to the triacylated lipoproteins in Gram-negative bacteria and the enzymes responsible are identified. See text for details.

The first two steps of lipoprotein biogenesis are universally conserved in eubacteria as judged by the presence of genes whose products are homologous to diacylglyceryltransferase and signal Type II peptidase (3, 7). During A. laidlawii genome annotation, lgt and lsp were readily detected by sequence homology. However, A. laidlawii as well as other mollicutes and Gram-positive bacteria lack recognizable lnt gene homologs (9). A question therefore arises whether the third step of protein acylation occurs in these organisms.

There are conflicting data concerning N-acylation in Mollicutes. In a number of studies, major lipoproteins such as variable lipoproteins VlpA and VlpC from Mycoplasma hyorhinis, LP44 of Mycoplasma salivarium, and macrophage-activating lipoproteins from Mycoplasma fermentans and Mycoplasma gallisepticum were shown to lack this modification, consistent with the lack of lnt homologs in their genomes (10, 11). On the other hand, the surface lipoprotein MG149 of Mycoplasma genitalium (12) and spiralin of Spiroplasma melliferum were found to be N-acylated (13).

A standard way to study the acylation state of total lipoproteins present in the cell is to extract lipoproteins with a detergent, selectively hydrolyze ester or amide bonds connecting the lipid and protein moieties, and determine the ratio of [O-ester-bound chains + N-linked chains]/[O-ester bound chains] for cell lipoproteins. For Mycoplasma mycoides and A. laidlawii, this parameter is 1.2 (14) and 1.1 (15), respectively, suggesting the predominant occurrence of O-ester-linked fatty acids in acylated proteins in these mollicutes. On the other hand, similar measurements performed for Mycoplasma agalactiae and M. gallisepticum lipoproteins revealed a ratio of 1.4 (16, 17), suggesting that a considerable fraction of lipoproteins is N-acylated and implying the existence of the third step of lipoprotein biogenesis. This notion, however, appears to contradict the finding with M. gallisepticum macrophage-activating lipoproteins (no N-acylation) referred to above (11). Clearly, development of robust methods to quantitatively measure the acylation state is necessary for resolving the existing controversies for Mollicutes as well as Gram-positive bacterial lipoproteins. Mass spectrometric methods offer a unique advantage in this regard as they allow analyzing the protein acylation state directly without the requirement for prior hydrolysis of the lipid-protein juncture. In this study, we report the development of a new method of determination of the acylation state of lipoproteins that relies on a combination of two-dimensional gel electrophoresis protein separation, transfer to nitrocellulose membrane, and MALDI-MS visualization of modified N-terminal tryptic peptides. We used this method to examine the acylation of eight major lipoproteins from A. laidlawii. We show definitively that all A. laidlawii lipoproteins examined are fully triacylated, implying the existence of robust N-acyltransferase activity in this organism.

EXPERIMENTAL PROCEDURES

Cell Growth

A. laidlawii PG8-A was cultivated in modified Edward medium (18) containing (per liter) 20 g of tryptose, 5 g of yeast extract, 5 g of NaCl, 5 g of CH3COONa, 1.3 g of KCl, and 3.6 g of Tris-HCl, pH 8. Before inoculation, 6% (w/w) horse serum, 0.5% (w/w) glucose, and 500 units/ml penicillin were added to the medium. A. laidlawii cultures were grown for 18 h at 30, 37, or 42 °C. Cells were collected by centrifugation at 7000 × g for 30 min at 4 °C.

In Vivo Incorporation of Exogenous 14C-labeled Fatty Acids into A. laidlawii Lipoproteins

74 kBq of either palmitic acid (16:0), oleic acid (18:1c), stearic acid (18:0), or linoleic acid (18:2c) (Amersham Biosciences) were added per 1 ml of growth medium. Exponentially growing cells were collected by centrifugation; washed twice with a buffer containing 150 mm NaCl, 50 mm Tris-HCl, and 2 mm MgCl2, pH 7.4; and treated with a nuclease mixture (Amersham Biosciences). Proteins were resolved by two-dimensional PAGE and silver-stained as described elsewhere (19). Gels were dried and exposed to a storage phosphor screen for 2 weeks. Images were acquired on a Typhoon scanner (Amersham Biosciences). Proteins were identified by in-gel trypsin digestion and MS analysis (20).

Lipoprotein Extraction with Triton X-114 Detergent

Cell pellets (10 μl) were resuspended in 500 μl of 0.1% Triton X-114, 150 mm NaCl, and 50 mm Tris-HCl, pH 8.0 and incubated for 1 h at 4 °C. After 15-min centrifugation at 16,000 × g (4 °C), the supernatants were transferred into a new tube and incubated at 37 °C for 15 min at which point the clear solution became cloudy due to micellae formation. Phase separation was performed by centrifugation at 16,000 × g at room temperature. The upper aqueous phase was discarded, and proteins from the lower organic phase were precipitated by the addition of 4 volumes of acetonitrile.

Electroblotting, in Situ Tryptic Digest, and Peptide Extraction from Nitrocellulose Membranes

After two-dimensional or one-dimensional PAGE separation, lipoproteins were transferred onto Hybond-C Extra membrane (Amersham Biosciences) via a standard electroblotting procedure and stained with Ponceau S (Fluka). The top layer of the membrane-containing stained protein spots was scrapped off, and crumbs of nitrocellulose (2–4 μl) were rinsed twice with 50 μl of 50 mm NH4HCO3, pH 7.5 in 10% acetonitrile. Proteins were digested overnight with trypsin (Promega; 5 μl of a 20 mg/ml solution in 50 mm NH4HCO3) at 37 °C. Peptides were extracted stepwise at room temperature with 20 μl of aqueous 0.5% trifluoroacetic acid for 30 min, 0.5% trifluoroacetic acid in 10% acetonitrile for 10 min, and 0.5% trifluoroacetic acid in 20% acetonitrile for 10 min. Finally, a 10-min extraction with 10 μl of chloroform, methanol, and 20% formic acid (1:3:1, v/v/v) or chloroform, acetonitrile, and 0.5% trifluoroacetic acid (1:3:1, v/v/v) was performed.

MALDI-MS Analysis

Sample aliquots (1–2 μl) were mixed on a steel target with 0.3 μl of 20 mg/ml 2,5-dihydroxybenzoic acid in 0.5% trifluoroacetic acid and 30% acetonitrile water solution (Aldrich). Mass spectra were recorded on an Ultraflex II MALDI-TOF-TOF mass spectrometer (Bruker Daltonics) equipped with a neodymium laser. The [MH]+ or [M − H] (for lipids) molecular ions were measured in reflector or linear mode; the accuracy of mass peak measurement was within 0.007%. Fragment ion spectra were obtained in Lift mode. The accuracy of fragment ion mass peak measurements was within 1 Da.

Protein Identification

Protein identification was carried out by peptide fingerprint search with the use of Mascot software (Matrix Science) through the NCBI A. laidlawii protein database. One missed cleavage, Met oxidation, and Cys-propionamide were permitted. Protein scores greater than 44 were considered to be significant (p < 0.05). Interpretation of MS-MS spectra and assignments of acylpeptides were performed with Biotools v.3 software (Bruker Daltonics), revealing a variety of attached long chain fatty acids (palmitate, stearate, etc.). To estimate the ratio of protein modification with a particular fatty acid, averaged values of normalized MS intensities Pi were used. For example, the extent of palmitoylation (PPal) was calculated using the following formula: PPal = P3P × 100% + (P2P1M + P3P+14 + P1P2S) × 66.6% + (P2P1M+14 + P2P1S+14 + P1P2S) × 33.3%. The extent of modification with other fatty acids was calculated in a similar way.

Fatty Acid Analysis

Fatty acid analysis of A. laidlawii lipid membranes was performed after methyl esterification by gas chromatography according to Russian state standard specification R 51483-99.

RESULTS

Identification of A. laidlawii Lipoproteins by Radioactive Labeling

To identify the most abundant acylated proteins of A. laidlawii, various 14C-labeled fatty acids (palmitic (16:0), oleic (18:1c), stearic (18:0), and linoleic (18:2c) acids) were added to A. laidlawii growth medium. Cells were grown and harvested, and proteins were resolved by two-dimensional PAGE. The profile of A. laidlawii proteins revealed after two-dimensional PAGE separation was highly reproducible (supplemental Fig. 1). Radioactively labeled proteins were revealed using a phosphorimaging system. The most efficient radioactive labeling was observed when cells were grown in the presence of [14C]palmitic acid, revealing a total of 17 major labeled spots (Fig. 2A). Much less efficient labeling was observed when proteins from cells grown on stearic acid-containing medium were subjected to two-dimensional PAGE analysis; however, the pattern of labeled proteins was the same (data not shown). No detectable labeling with oleic and linoleic acids was observed (not shown).

FIGURE 2.
Two-dimensional page analysis of A. laidlawii lipoproteins. A, cell extracts were prepared from A. laidlawii cultures grown at 37 °C in the presence of [14C]palmitic acid, proteins were resolved by two-dimensional PAGE, and radioactive spots were ...

We performed mass spectrometric identification of 209 A. laidlawii proteins from silver-stained gels subjected to autoradiography (shown in supplemental Fig. 1A) by tryptic peptide fingerprinting. The results are presented in supplemental Table 1. A total of 19 proteins from 17 radioactively labeled on-gel spots were identified (Fig. 2A and Table 1). 17 of the identified proteins possess Sec-(SPII) signal peptides and are predicted to be extracellular lipoproteins. Two proteins, translation elongation factor EF-G and dihydrolipoamide acetyltransferase, lack signal peptides and are generally known to be cytoplasmic. Their presence in the [14C]palmitate-labeled protein set is likely due to imperfect separation of proteins on two-dimensional gels. These proteins were therefore excluded from further analysis. None of the 192 proteins identified in non-radioactively labeled spots contained a predicted Sec-(SPII) signal.

TABLE 1
A. laidlawii proteins identified from [14C]palmitate-labeled two-dimensional PAGE spots

All 17 [14C]palmitate-labeled A. laidlawii proteins were identified unambiguously (see the supplemental representative HTML Mascot files) and were found to be related to substrate-binding subunits of ABC transporters of carbohydrates, amino acids, and inorganic ions or are annotated as hypothetical surface-anchored proteins of unknown function (Table 1). These same proteins were identified in the case of stearic acid labeling (data not shown).

We next performed protein extraction from A. laidlawii cells with the detergent Triton X-114, which selectively transfers some membrane proteins to the micelle detergent phase. Extracted lipoproteins were resolved by standard one-dimensional or two-dimensional PAGE (Fig. 2B). Major membrane proteins from one-dimensional and two-dimensional gels matched each other, indicating that most A. laidlawii lipoproteins are resolved by two-dimensional PAGE at our conditions. A total of 28 protein spots were observed on a Coomassie-stained two-dimensional gel. A. laidlawii proteins present in two-dimensional gel spots were identified by MALDI-MS and are presented in supplemental Table 2. Protein spots that matched radioactively labeled spots in Fig. 2A are indicated in the gel presented in Fig. 2B by corresponding numbers. Three radioactively labeled proteins (spots 1, 6, and 11) were absent from the Triton X-114 extracts probably due to their low abundance. Coomassie-stained proteins from the Triton X-114 extract that did not have a radioactively labeled counterpart did not contain a Sec-(SPII) signal (see supplemental Table 2). Most of them are major A. laidlawii proteins that should be regarded as contaminants.

MS Visualization of Acylpeptides from A. laidlawii Lipoproteins

To determine the acylation state of eight major A. laidlawii lipoproteins extracted with Triton X-114, peptide fingerprints after in-gel tryptic digests were examined. Although numerous internal tryptic peptides were readily observed on MALDI mass spectra for each protein studied, we were unable to detect the presence of N-terminal peptides for any of the proteins despite the numerous attempts. The possible reasons for our failure to detect the expected acylated N-terminal peptides may include (but are not limited to) (i) low intensity of corresponding MS signals due to interference from other peptides and (ii) poor extraction of peptides bearing linked long chain fatty acids from polyacrylamide gels. To observe and analyze acylated peptides directly, we developed a new method that included transblotting of proteins separated by two-dimensional PAGE to nitrocellulose membranes. After transfer to the membrane, proteins were stained, and the most abundant protein spots were treated with trypsin. We expected that peptides modified with lipids, if present in a peptide spot, would be better retained on the membrane due to their hydrophobicity. Accordingly, trypsin-treated membrane samples were gently washed as described under “Experimental Procedures” to remove unmodified peptides. MALDI-MS analysis of peptides present in the washes confirmed that only unmodified internal peptides of A. laidlawii lipoproteins were present in the sample (data not shown). The final extraction of the membrane was performed by a chloroform-methanol-formic acid-water solution, and extracted material was subjected to MALDI-MS analysis. As can be seen in Fig. 3, the material eluted at this step contained, for each lipoprotein spot tested, a series of peaks, which were absent from the wash samples. Each peak in the series could be interpreted as a defined triacylated N-terminal tryptic peptide derived from a corresponding A. laidlawii lipoprotein. No peaks corresponding to diacylated N-terminal tryptic peptides were observed. We therefore conclude that the new procedure indeed allows efficient direct visualization of the lipoprotein acylation state. We further conclude that despite the absence of a recognizable N-acyltransferase homolog in A. laidlawii the third step of protein acylation is carried out to completion.

FIGURE 3.
Mass spectrometric analysis of N-acylpeptides from A. laidlawii lipoproteins. N-Acylpeptides of A. laidlawii lipoproteins GI:162447521 (predicted ligand-binding component of an ABC-type sugar transport system; A), GI:162448245 (predicted substrate-binding ...

Detailed Analysis of Acylation of A. laidlawii Lipoproteins

For each lipopeptide series, a total of seven mass peaks were observed. Two major peaks were shifted by 788 and 816 mass units from calculated values of unmodified peptides. These peaks were interpreted, respectively, as triple palmitoyl- (3P; +788) and dual palmitoyl/monostearoyl (2P1S; +816)-modified versions of N-terminal lipoprotein peptides. Two minor peaks in each series were shifted by +760 and +844 mass units and correspond to double palmitoyl/monomyristoyl (2P1M; +760) and double stearoyl/monopalmitoyl (1P2S; +844) modifications. The four peaks described above can be arranged in an ascending series with each peak differing from another by 28 mass units corresponding to the presence of two CH2 groups. Curiously, in addition to these peaks, three peaks shifted by 14 mass units (a single CH2 group) were present. The appearance of these peaks cannot be explained in terms of the usual even-numbered saturated fatty acid modifications and may be due to infrequent modification with monomethyl-branched saturated fatty acids.

The lipopeptide mass spectral profiles for each of the eight major A. laidlawii lipoproteins analyzed appeared to be highly reproducible because the ratios of individual peaks in each series were found to be constant for samples prepared from independently grown cell cultures (Table 2). The precision of these measurements was within 5% for major 3P and 2P1S peak signals and 2% for minor peaks. We assume that for each peak in the series the peak intensity ratio approximately reflects the quantitative ratio of differently acylated peptides and therefore reports the overall acylation state of a lipoprotein. Although this assumption is difficult to prove experimentally, published data on influenza type A virus hemagglutinin seem to support this notion (21). Because the ratio of lipopeptide peaks for different A. laidlawii proteins was the same within the experimental error, we calculated an average frequency of modifications with different fatty acids for eight lipoproteins studied. By far, the most frequent modification is palmitic acid: it accounts for ~74% of the total fatty acid modifications detected. The frequencies of stearic and myristic acid modifications are 16 and 3%, correspondingly. The remaining 7% of lipid modifications are due to potential odd-numbered fatty acid modifications (6%) and modifications with unsaturated fatty acids (<1% of total modifications; revealed as low intensity signals neighboring major peaks in Fig. 3). For major fatty acid modifications, these estimates are in good agreement with observed intensities of the 14C-fatty acid protein labeling experiment (Fig. 2).

TABLE 2
Relative intensities of individual mass peaks in N-terminal peptide series of eight A. laidlawii lipoproteins

Comparison of Fatty Acid Composition of A. laidlawii Lipoproteins and Membranes

The pronounced bias in frequencies of modification with individual fatty acids is at odds with published data, which suggest that the frequencies of various long fatty chains in this organism (myristic, palmitic, and stearic) are more balanced (22). To investigate whether the assortment of protein-attached fatty acids reflects the composition of A. laidlawii lipid membrane, standard analysis of A. laidlawii membrane fatty acids was performed (Fig. 4). The results indicated that stearate was the most abundant A. laidlawii membrane fatty acid (~27%) followed by palmitate (25%) and oleic and linoleic acids (17 and 10%, respectively). Myristate was present in small (2–3%) amounts. Comparison of fatty acid frequencies in A. laidlawii lipoproteins and membranes thus reveals that the acylation machinery has a clear preference for palmitate and possibly myristate. Conversely, stearate and unsaturated fatty acids do not appear to be preferred acylation substrates.

FIGURE 4.
Comparison of fatty acid composition of A. laidlawii lipid membranes and lipoproteins. The average percentages of individual fatty acids present in cell membrane (gray) and lipoproteins (white) are shown. For each lipoprotein, individual modifications ...

Fatty Acid Composition of A. laidlawii Lipoproteins Is Not Temperature-dependent

To regulate the lipid composition of its membrane and maintain a proper balance between various lipids, A. laidlawii alters the polar head group structure of phospho- and glycolipids (e.g. the proportion of mono-/diglucosyldiacylglycerols and monoacyldi-/monoglucosyldiacyl glycerols). In this way, the lipid composition of the membrane is changed as a function of growth temperature (2225). We were interested in whether these changes lead to changes in protein acylation. Accordingly, we determined whether the composition of protein-associated fatty acids depends on growth temperature. A. laidlawii cultures were grown at 37, 30, or 42 °C for an identical amount of time. Because the A. laidlawii growth rate is highly temperature-dependent, considerably less biomass was collected from 30 and 42 °C cultures, which affected the quality of the resulting mass spectra (Fig. 5). Nevertheless, comparisons of Fig. 5 spectra with Fig. 2 data (obtained with material prepared from cultures grown at 37 °C) revealed that peak intensities of differently acylated peptides were the same at all three temperatures. This lack of differences in protein acylation was in stark contrast with the experimentally observed changes in lipid composition of membranes of A. laidlawii grown at different temperatures (Fig. 5B), which was in agreement with published data (2225). Thus, the composition of A. laidlawii protein-associated fatty acids does not depend on either growth temperature or membrane composition.

FIGURE 5.
Fatty acid composition of lipoproteins and membrane lipid composition from A. laidlawii cells grown at different temperatures. A, A. laidlawii cells were grown at three different temperatures, and N-acylpeptides of the indicated lipoproteins were analyzed ...

Specificity of Acylation at Various Stages of Lipoprotein Genesis

Because a strong preference for incorporation of palmitate in A. laidlawii lipoproteins exists, we were interested to know at which step of the acylation process the less common stearate is incorporated. To address this question, we compared MS-MS spectra of 3P and 2P1S forms for N-terminal peptides from two lipoproteins, GI:162448066 and GI:162448245 (Fig. 6). For each lipopeptide variant, series of C-terminal y- and z-type daughter ions, which confirmed the amino acid structure of the peptide moieties (CTPK for GI:162448066 and CIGGER for GI:162448245), were observed. Most interestingly in each MS-MS spectrum, signals corresponding to a 256-mass unit loss from the parent ion were detected (Fig. 6, marked by asterisks). The only possibility to get such a mass difference is the elimination of the entire palmitic acid (CH3(CH2)14COOH) from the S-diacylglycerol moiety. Elimination of the CH3(CH2)16COOH stearic acid should have produced a shift of 284 mass units. However, the corresponding ions were not observed, suggesting that in 2P1S-modified lipopeptides stearate is not found on S-diacylglycerol. In other words, palmitates are predominantly O-linked in 2P1S forms. It therefore follows that stearate residues are mainly amide-linked in these forms, reflecting stage specificity of the acylation process.

FIGURE 6.
MS-MS analysis of A. laidlawii lipoprotein N-acylpeptides. 3P- (upper panels) and 2P1S (lower panels)-modified N-acylpeptides of A. laidlawii lipoprotein GI:162448066 (A; parent ions of m/z 1236 and 1264) and GI:162448245 (B; parent ions of m/z 1422 and ...

DISCUSSION

In this work, we report the development of a robust MS-based method that allows direct determination of the acylation status of bacterial lipoproteins. Using this method, we show that every studied lipoprotein of the mollicute A. laidlawii is triacylated. By extension, we hypothesize that lipoproteins in other mollicutes are also triacylated. The existence of A. laidlawii lipoproteins in triacylated form is unexpected because a gene encoding recognizable homologs of N-acyltransferases responsible for incorporation of the third fatty acid in lipoproteins from Gram-negative bacteria is absent from the A. laidlawii genome. Identification of triacylation as the major (perhaps the only) state of A. laidlawii lipoproteins is in contrast with published data, which suggest the predominant involvement of O-ester-linked fatty acids but not amide-linked fatty acids in protein acylation in A. laidlawii (14, 26). In the latter work, the authors showed that globomycin, a cyclic lipopeptide specifically inhibiting signal peptidase II, had no dramatic effect on the extent of acylation of A. laidlawii lipoproteins, suggesting the preferential absence of amide-linked acyl chains. The reasons for the discrepancy with previous data are unknown but may be related to the fact that both Le Hénaff and co-workers (14) and Nyström et al. (26) analyzed fatty acids that were released from lipoproteins by various chemical treatments. The failure to detach even trace amounts of non-covalently associated lipids from their lipoprotein preparations or imperfection of releasing reactions could have heavily influenced the outcome of their measurements.

In our work, we visualized acyl chains attached to N-terminal peptides directly, thus avoiding the issue of contaminating lipids. The only possible source of biases in our method is related to acylpeptide solubility and extraction efficiency. However, the procedure that we developed for lipopeptide extraction from nitrocellulose membranes (mild washes followed by elution at harsh conditions) and thorough control of the wash and elution stages by mass spectrometry makes this possibility unlikely. In addition, the results of two-dimensional gel electrophoresis separation of radioactively labeled A. laidlawii lipoproteins reveal that most lipoproteins form predominantly single spots (Fig. 2A), indicating the existence of one major modification, which according to MS analysis is triacylation. The enzyme that is responsible for A. laidlawii lipoprotein acylation remains to be identified.

Our method enables us not just to detect lipoprotein N-peptide acylation but also to determine the amounts of different fatty acids linked to a peptide. The ratios of various long chain fatty acids were found to be similar for all A. laidlawii lipoproteins studied. At the same time, these ratios were dramatically different from the fatty acid composition of lipid membranes. Specifically, a strong preference for incorporation of palmitate in lipoproteins was revealed. Changes in membrane composition that resulted from growth temperature change had no effect on the ratios of fatty acid incorporation in lipoproteins. Hence, the fatty acids attached to lipoproteins of A. laidlawii are not correlated with lipid composition of its membrane.

MS-MS analysis of A. laidlawii lipoprotein N-terminal peptides revealed that palmitate is exclusively attached via S-glycerol. It therefore follows that selection of N-linked fatty acids is less stringent. Together, these observations point toward specificity of the protein acylation process. The preferential occurrence of palmitates in O-linkages indicates that A. laidlawii lipoprotein diacylglyceryltransferase (Lgt) uses mainly dipalmitoylglycerophospholipid substrate at the first step of lipoprotein biogenesis. This preference may be caused by true specificity of Lgt at the stage of lipid substrate selection or may be an indirect consequence of the local lipid content of the immediate membrane surroundings of Lgt. Development of in vitro protein acylation systems with membranes of defined content should help clarify this important issue.

While our work was under review, an important study by Asanuma et al. (27) appeared where the authors report that SitC and four other major lipoproteins from Staphylococcus aureus are triacylated despite the apparent absence of the lnt (N-acyltransferase) gene in this organism. It therefore appears that triacylation of lipoproteins is a global phenomenon in the bacterial world. It thus follows that a third stage of protein acylation (Fig. 1) exists in microorganisms that lack identifiable lnt homologs. At least in E. coli, the lnt gene is essential (8). By extension, genes coding for Lnt analogs may be essential in Mollicutes and Gram-positive bacteria. However, in these organisms, N-acylation cannot serve to direct proteins to the outer membrane as this membrane is absent. The physiological significance of N-acylation in Mollicutes and Gram-positive bacteria remains to be determined. Identification of novel enzymes responsible for triacylation in organisms lacking lnt homologs is another important avenue of future research.

Supplementary Material

Supplemental Data:

Acknowledgments

We thank Dr. V. G. Baykov (Nutrition Institute of the Russian Academy of Medical Science) for standard fatty acid analysis of A. laidlawii lipid membranes and Dr. K. Severinov (Waksman Institute, Piscataway, NJ) for valuable discussions and help with manuscript preparation.

*This work was supported by Ministry of Education and Science of the Russian Federation State Contracts 14.740.11.0763 and 16.740.11.0371 (to V. M. G.) and Molecular and Cellular Biology program grant from the Russian Academy of Sciences (to Konstantin Severinov).

An external file that holds a picture, illustration, etc.
Object name is sbox.jpgThe on-line version of this article (available at http://www.jbc.org) contains supplemental Fig. 1, Tables 1 and 2, and representative HTML Mascot files.

2The abbreviations used are:

ABC
ATP-binding cassette
3P
triple palmitoyl
2P1S
dual palmitoyl/monostearoyl
1P2S
double stearoyl/monopalmitoyl
2P1M
double palmitoyl/monomyristoyl
SPII
signal peptidase II.

REFERENCES

1. Schultz A. M., Henderson L. E., Oroszlan S. (1988) Annu. Rev. Cell Biol. 4, 611–647 [PubMed]
2. Braun V., Rehn K. (1969) Eur. J. Biochem. 10, 426–438 [PubMed]
3. Babu M. M., Priya M. L., Selvan A. T., Madera M., Gough J., Aravind L., Sankaran K. (2006) J. Bacteriol. 188, 2761–2773 [PMC free article] [PubMed]
4. Razin S., Yogev D., Naot Y. (1998) Microbiol. Mol. Biol. Rev. 62, 1094–1156 [PMC free article] [PubMed]
5. Andersson A. S., Rilfors L., Orädd G., Lindblom G. (1998) Biophys. J. 75, 2877–2887 [PMC free article] [PubMed]
6. Saito Y., Silvius J. R., McElhaney R. N. (1978) J. Bacteriol. 133, 66–74 [PMC free article] [PubMed]
7. Bos M. P., Tommassen J. (2004) Curr. Opin. Microbiol. 7, 610–616 [PubMed]
8. Terada M., Kuroda T., Matsuyama S. I., Tokuda H. (2001) J. Biol. Chem. 276, 47690–47694 [PubMed]
9. Hutchings M. I., Palmer T., Harrington D. J., Sutcliffe I. C. (2009) Trends Microbiol. 17, 13–21 [PubMed]
10. Mühlradt P. F., Kiess M., Meyer H., Süssmuth R., Jung G. (1998) Infect. Immun. 66, 4804–4810 [PMC free article] [PubMed]
11. Jan G., Le Hénaff M., Fontenelle C., Wróblewski H. (2001) Arch. Microbiol. 177, 81–90 [PubMed]
12. Shimizu T., Kida Y., Kuwano K. (2008) Infect. Immun. 76, 3672–3678 [PMC free article] [PubMed]
13. Le Hénaff M., Fontenelle C. (2000) Arch. Microbiol. 173, 339–345 [PubMed]
14. Jan G., Fontenelle C., Verrier F., Le Hénaff M., Wróblewski H. (1996) Curr. Microbiol. 32, 38–42 [PubMed]
15. Le Hénaff M., Chollet A., Fontenelle C. (2001) Curr. Microbiol. 43, 424–428 [PubMed]
16. Le Hénaff M., Guéguen M. M., Fontenelle C. (2000) Curr. Microbiol. 40, 23–28 [PubMed]
17. Jan G., Fontenelle C., Le Hénaff M., Wróblewski H. (1995) Res. Microbiol. 146, 739–750 [PubMed]
18. Liska B., Smith P. F. (1974) Folia Microbiol. 19, 107–117 [PubMed]
19. Golovina A. Y., Sergiev P. V., Golovin A. V., Serebryakova M. V., Demina I., Govorun V. M., Dontsova O. A. (2009) RNA 15, 1134–1141 [PMC free article] [PubMed]
20. Shevchenko A., Jensen O. N., Podtelejnikov A. V., Sagliocco F., Wilm M., Vorm O., Mortensen P., Shevchenko A., Boucherie H., Mann M. (1996) Proc. Natl. Acad. Sci. U.S.A. 93, 14440–14445 [PMC free article] [PubMed]
21. Serebryakova M. V., Kordyukova L. V., Baratova L. A., Markushin S. G. (2006) Eur. J. Mass Spectrom. 12, 51–56
22. Niemi A. R., Rilfors L., Lindblom G. (1995) Biochim. Biophys. Acta 1239, 186–194 [PubMed]
23. Nyström S., Wallbrandt P., Wieslander A. (1992) Eur. J. Biochem. 204, 231–240 [PubMed]
24. Wieslander A., Christiansson A., Rilfors L., Khan A., Johansson L. B., Lindblom G. (1982) Rev. Infect. Dis. 4, (suppl.) S43–S49 [PubMed]
25. Lindblom G., Orädd G., Rilfors L., Morein S. (2002) Biochemistry 41, 11512–11515 [PubMed]
26. Nyström S., Johansson K. E., Wieslander A. (1986) Eur. J. Biochem. 156, 85–94 [PubMed]
27. Asanuma M., Kurokawa K., Ichikawa R., Ryu K. H., Chae J. H., Dohmae N., Lee B. L., Nakayama H. (2011) FEBS J. 278, 716–728 [PubMed]

Articles from The Journal of Biological Chemistry are provided here courtesy of American Society for Biochemistry and Molecular Biology
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...