Envelopes of Chlamydia psittaci in Alkaline Buffer and

Transcription

1 JOURNAL OF BACrTROLOGY, Jan. 1976, p Copyright i 1976 American Society for Microbiology Vol. 125, No. 1 Printed in U.S.A. Protein-Carbohydrate-Lipid Complex Isolated from the Cell Envelopes of Chlamydia psittaci in Alkaline Buffer and Ethylenediaminetetraacetate TOSHIHIKO NARITA AND G. P. MANIRE* Department of Bacteriology and Immunology, School of Medicine, University of North Carolina, Chapel Hill, North Carolina Received for publication 29 September 1975 Exposure of isolated cell envelopes from purified infectious elementary bodies (EB) of Chlamydia psittaci to sodium carbonate-bicarbonate buffer at ph 1 plus ethylenediaminetetraacetate (EDTA) results in partial solubiliation of the envelopes. The released materials represent 2% of the dry weight, 16% of the total protein, 4% of the total carbohydrate, and 9% of the total lipid of the cell envelopes. Sucrose density gradient centrifugation, and Sephadex G-2, Sepharose 4B, or diethylaminoethyl-cellulose column chromatography, reveal a protein-carbohydrate-lipid complex of several hundred thousand molecular weight that contains 5% protein. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the isolated EB cell envelopes reveals two major protein bands, A and B, with estimated molecular masses of approximately 85, and 53,, respectively, both of which also stain for the presence of carbohydrate and lipid. Gel electrophoresis of the protein-carbohydrate-lipid complex reveals two proteins bands, C and D, with estimated molecular weights of approximately 17, and 13,, respectively, which contain lipid and a small amount of carbohydrate; bands A and B are not present in the complex. Gel electrophoresis of the cell envelope residues after extraction of the complex with alkali and EDTA shows a single main band, corresponding to the position of band B, which contains protein, carbohydrate, and lipid; band A is completely missing. B and A is believed to be a component of the complex, which is split into two subunits on alkali solubiliation. In the preceding paper (15), we reported that exposure of cell envelopes isolated from purified elementary bodies (EB) of Chlamydia psittaci to alkaline buffer and ethylenediaminetetraacetate (EDTA) caused the liberation of spherical and rodlike particles, which apparently are surface components of the cell. This paper describes further investigation of the substances released from EB cell envelopes by alkali and EDTA, including studies on the chemical characteriation and the polypeptide composition as analyed by sodium dodecyl sulfate (SDS)- polyacrylamide gel electrophoresis of the released cell envelope components. MATERIALS AND METHODS General. The Cal 1 meningopneumonitis strain of C. psittaci was employed for these studies. The methods for cultivation and purification of the organism, and for isolation of the cell envelopes, were reported in the preceding paper (15). Alkali solubiliation and isolation of the cell envelope "soluble fraction." The envelope materials liberated from isolated cell envelopes of purified EB by alkali treatment, designated as "soluble fraction," were prepared as follows. For analytical experiments, isolated cell envelopes were exposed to ph 1,.1 M sodium carbonate-bicarbonate (carbonate) buffer with and without 1 mm EDTA. After 6 h of incubation at 38 C, the reaction mixture was centrifuged at 12, x g for 1 h at 4 C, and then the supernatant fluid was centrifuged a second time at 12, x g for 1 h at 4 C. The resulting supernatant fraction was then centrifuged at 3, x g for 1 h at 4 C. This clear supernatant fluid containing the soluble fraction was subjected to chemical analysis. To obtain maximum yield, the cell envelopes were treated essentially as described above, except that the sediment obtained after the first centrifugation at 12, x g for 1 h was further extracted three times for 8 h each at ph 1. The 3, x g supernatant fluid was dialyed for 72 h at 4 C against several changes of distilled water and then lyophilied. The lyophilied preparation was stored in a desiccator at 4 C. The 12, x g sediment after repeated alkali extraction, designated as the cell envelope residue, was also washed with chilled distilled water, lyophilied, and stored in a desiccator at 4 C. 38 Downloaded from on August 13, 218 by guest

2 VOL. 125, 1976 Chemical analyses. Total protein was measured by the method of Lowry et al. (11), using crystalline bovine serum albumin (Sigma Chemical Co., St. Louis, Mo.) as a standard. Total carbohydrate determination was performed by the phenol-sulfuric acid method as described by Hodge and Hofreiter (6). Carbohydrate concentration (expressed as glucose equivalents) was determined by reference to a glucose standard run in each experiment. Reducing sugars were assayed by the method of Park and Johnson (16) after hydrolysis with 3 N HCl for 3 h at 1 C in a sealed ampoule. For hexosamine estimation, samples were hydrolyed with 6 N HCl for 6 h at 95 C and measured according to the method of Ghuysen et al. (5), with D-glucosamine hydrochloride (Sigma) as a standard. No correction for hexosamine loss due to hydrolytic condition was applied. Lipids were extracted by the procedures of Bligh and Dyer (3). Samples were hydrolyed with 6 N HCl for 3 h at 1 C before extraction and dried in vacuo. The hydrolyed samples were mixed with chloroformmethanol (1:2, vol/vol), and then the chloroform phase was removed after centrifugation at 2, rpm for 1 min. The water-methanol phase containing nonlipid residues was dried to a constant weight in a vacuum desiccator. The amount of lipid was estimated by subtraction of dry weight of nonlipid residues from total dry weight. In some cases, lipids were detected qualitatively by spotting samples on a strip of Whatman no. 1 chromatography paper. These were then stained overnight with Sudan black B and destained in 5% ethanol. The comparative intensity of staining reaction was determined arbitrarily as +, + +, and Total phosphorus was determined by the method of Lowry et al. (1), employing KH2PO4 as a standard. Dry weight measurement. For analyses of dry weights of cell envelopes before and after the alkali treatment, the suspension was centrifuged twice at 12, x g for 1 h at 4 C. The supernatant fluid was discarded, and the precipitate was washed twice with chilled distilled water. The washed precipitate was lyophilied and followed by vacuum desiccation over phosphorus pentoxide until the sample had reached constant weight. IsopyCnic sucrose gradient centrifugation. Linear sucrose density gradients were prepared from to 3% (wt/vol) sucrose solution containing ph 1,.1 M carbonate buffer with 1 mm EDTA. Two milliliters of sample (1.5 mg/ml) dissolved in the same buffer was layered on the 24-ml gradients and spun in the SW25 swinging bucket rotor in a Spinco model L centrifuge at 22, rpm for 3 h at 4 C. Fractions were dialyed for 72 h at 4 C against several changes of distilled water to remove sucrose and then assayed for protein by absorbance at 28 nm and for carbohydrate (6). For the detection of lipid, fractions were lyophilied, reconstituted in 1,ul of distilled water, and assayed qualitatively as described above. Sephadex G-2 gel filtration. The buffer used to equilibrate the.9- by 41-cm column of Sephadex G-2 (Pharmacia Fine Chemicals, Inc., Upsala, Sweden) and to dissolve and elute the sample was composed of.1 M carbonate buffer, ph 1, with 1 RELEASE OF PROTEIN-CARBOHYDRATE-LIPID COMPLEX 39 mm EDTA. Prior to elution, the column was washed thoroughly with the same buffer. Four milligrams of sample was applied to the column and chromatographed at 25 C at a flow rate of 3. ml/h, and.5-ml fractions were collected. Blue dextran 2, (Sigma) was used to determine the void volume (V) of the column. Sepharose 4B gel filtration. A Sepharose 4B column (.9 by 5 cm) was equilibrated with.5 M tris(hydroxymethyl)aminomethane (Tris)-hydrochloride buffer, ph 8.5, with 1 mm EDTA. The sample containing 7 mg of the soluble fraction in the same buffer was applied to the column and eluted at a flow rate of 1.8 ml/h, collecting.6-ml fractions. Blue dextran was run to determine VO. All procedures were carried at at 4 C. Ion exchange chromatography. Diethylaminoethyl-cellulose column chromotography was carried out on Whatman DE52 (.9 by 16 cm) equilibrated with.5 M Tris-chloride buffer, ph 8.5, containing 1 mm EDTA. Four milligrams of sample was applied to the column and followed by partial elution with 36 ml of the same buffer. A linear gradient of sodium chloride was generated by mixing 9 ml of Tris buffer containing 1 mm EDTA with 9 ml of the same buffer containing 1. M sodium chloride. The column was run at 4 C. Fractions of 3.6 ml were collected at a flow rate of 18 ml/h. Polyacrylamide gel electrophoresis. SDS-polyacrylamide gel electrophoresis was performed on 1% polyacrylamide gel in the presence of.1% SDS at ph 7.2 as described by Maiel (12), and Weber and Osborn (25). Acrylamide, bisacrylamide, 2-mercaptoethanol, and SDS were purchased from Eastman Kodak Co., Rochester, N.Y. Samples were suspended in a solution containing 1% SDS-1% 2-mercaptoethanol in.1 M sodium phosphate (phosphate) buffer, ph 7.2, and heated at 1 C for 1 min. Since the addition of urea into the solubiliing solution was found to have no significant effect on the pattern of bands, urea was not routinely used. Prior to application to the gel, the sample mixture was mixed with sucrose solution and bromophenol blue to make a final concentration of 1 and.1%, respectively. The sample mixture (1 to 2 ug of protein per gel) was applied to a 5- by 8-mm gel, and electrophoresis was carried out at 6 ma/gel for 8 h at 2 C. Protein bands were detected by staining gels with.25% Coomassie brilliant blue (Sigma), and gels were destained by repeated washing with 7% acetic acid. In some cases, gels stained with Coomassie blue were scanned at 62 nm with a Gilford model 24 spectrophotometer equipped with a model 241 linear transport and a recorder. For carbohydrate or glycoprotein, gels were stained with a periodic acid-schiff reagent (27). Lipids were detected by placing gels in Sudan black B for 12 h and destained with 5% ethanol. After destaining, the gels were placed in distilled water for 1 week for rehydration. Molecular weight determination by gel filtration or by SDS-polyacrylamide gel electrophoresis. Molecular weight estimations were made by gel filtration on a Sephadex G-2 column prepared as described above. According to the methods of Andrews (1) and Downloaded from on August 13, 218 by guest

3 31 NARITA AND MANIRE Whitaker (26), molecular weight was determined from a log of the ratio of the elution volume (V.) to V versus the log of the molecular weight of the marker protein or glycoprotein. Molecular weight standards employed were: egg albumin (45,), bovine serum albumin (67, and 134,), human gamma globulin (14,), catalase (24,), and thyroglobulin (66,). V was determined with blue dextran 2,. Molecular weights of bands obtained in SDS-polyacrylamide gel electrophoresis were determined from a standard curve of log molecular weight versus relative electrophoretic mobility to bromophenol blue by the method described by Shapiro et al. (21) and Weber and Osborn (25). Cytochrome c (13,), trypsin (24,), egg albumin (45,), and bovine serum albumin (67,) were used as marker proteins (Sigma). Each was dissolved in.1 M phosphate buffer containing 1% SDS-1% 2-mercaptoethanol at 38 C for 3 h and then subjected to gel electrophoresis as described above. RESULTS The major components of isolated cell envelopes of EB and the percentage of each component per dry weight of the cell envelope are listed in Table 1. Protein and lipid constituted the predominant components of the envelopes, and a relatively small amount of carbohydrate was detected. The combination of ph 1 carbonate buffer and 1 mm EDTA resulted in the liberation of approximately 16% of cell envelope protein, 4% of carbohydrate, and 9% of lipid from the envelopes after 6 h of incubation, and consequently about 2% dry weight of the envelope was released (Table 2). This treatment had a greater effect on the envelopes and on the individual components of cell envelopes than did treatment in buffer alone or ph 8 phosphate buffer plus EDTA. It appears that the presence of EDTA is required for extensive release of carbohydrate in the envelopes, in addition to a greater solubiliation of reducing sugars and hexosamine, as compared with the use of buffer alone. As no assay for individual carbohydrate released from cell envelopes was performed, it is not clear if EDTA causes liberation of an increased quantity of specific carbohydrates. TABLE 1. Major components found in the isolated cell envelopes of EB Constituents % Dry wt Protein Carbohydrate Reducing sugars Hexosamine....6 Lipid Phosphorus The released envelope materials were subjected to sucrose density gradient centrifugation and Sephadex, Sepharose, or diethylaminoethyl-cellulose column chromatography. Centrifugation in sucrose density gradients in ph 1 carbonate buffer containing 1 mm EDTA resulted in the appearance of one relatively broad peak trailing toward the top of the gradients (Fig. 1). A similar pattern was obtained for a sample centrifuged for 14 h in sucrose gradients in ph 8.5 Tris buffer plus 1 mm EDTA. It should be noted, however, that when values for protein alone were plotted a shoulder was observed, suggesting the presence of two peaks. The elution profile of the released material suspended in ph 1 carbonate buffer plus EDTA and chromatographed on Sephadex G-2 column equilibrated with the same buffer is represented in Fig. 2. The bulk of the protein, TABLE 2. Cell envelope components liberated by incubation in ph 1 carbonate buffer with and without EDTA Components % of each component released in Buffer alone Buffer + EDTA Protein Carbohydrate Reducing sugars Hexosamine Lipid Phosphorus NDa 2. Dry weight and, Not done..6 o N < 4 o.2 ca J. BACTERIOL. FRACTION NUMBER FIG Analysis by centrifugation, filtration, and chromatography of EB cell envelope components liberated by incubation in ph 1 carbonate buffer and EDTA. Procedures are described in Materials and Methods. Symbols: () protein; () carbohydrate. Lipid was qualitatively detected by staining with Sudan black B. FIG. 1. Sucrose density gradient centrifugation (O to 3%). The top of the gradient is on the right. E N.7 wlj I m Downloaded from on August 13, 218 by guest

4 RELEASE OF PROTEIN-CARBOHYDRATE-LIPID COMPLEX 311 o.6 6 CY.4 4 D E I w :; on,.2 2 C - a- X. 2...~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ VOL. 125, E c 4 a: +++ I % O I a- FRACTION NUMBER m cr FIG. 2. Sephadex G-2 filtration. The void volume (arrow) was determined with Blue dextran a: 1 t SlDO carbohydrate, and lipid appeared as a single FRACTION NUMBER peak immediately after the void volume. Gel FIG. 3. Sepharose 4B column chromatography. The filtration by Sepharose 4B with a higher exclusion limit was also carried out at ph 8.5, taking void volume is indicated by arrow. care that the ph of the eluant never exceeded 9. according to the manufacturer's instructions. In this case, the protein, carbohydrate,.4-25 E.1 c _ and lipid components emerged together slightly R -1 -> 2- after the void volume as a broad peak trailing I 1 w -8 Ft almost the entire length of the column (Fig. 3). This suggests that the homogeneity in sie of O c 6 < 12 5 o - a4 the released material may be dependent upon O -2 the ph of the eluant or surrounding medium. C~~~~~~ 5 cr The protein, carbohydrate, and lipid components in the released material were eluted in u parallel as a single peak from a diethylaminoethyl-cellulose column when the concentration FRACTION NUMBER of the eluting solvent reached approximately.4 FIG. 4. Diethylaminoethyl-cellulose column chromatography. Dotted line represents linear gradient of M (Fig. 4). About 7% of the protein applied NaCI. had been eluted when the salt gradient was terminated. In short, none of the fractionation procedures used in this study could separate the 2.5 I I I I I I I I I components of the released material, thus indicating their existence as a protein-carbohy- (E) drate-lipid complex. 2.. B The molecular weight of the complex was between 4, and 5, (Fig. 5) when \(H) measured by Sephadex G-2 gel filtration, 1.5 assuming that the complex had globular configuration. However, Andrews (1) and Ward and Arnott (24) have claimed that the density of sie (T)...1 of glycoprotein is not the same as that of protein 1..v and that their elution behaviors on a dextran gel column differ from that of normal globular MOLECULAR WEIGHT x 1 4 protein. Hence, the gel filtration method may FIG. 5. Molecular weight determination of the protein-carbohydrate-lipid complex from the EB cell have limited accuracy when employed for molecular weight determinations of such complexes. Materials and Methods. Molecular weight of the envelope. Molecular weight was estimated by gel filtration on a Sephadex G-2 column as described in The fractions eluted as a single peak after complex is indicated by the arrow. (E) Egg albumin; Sephadex G-2 gel filtration were pooled, B, bovine serum albumin; (H) human gamma globulin; C, catalase; (T) thyroglobulin. The markers in dialyed against several changes of distilled water at 4 C, and lyophilied. Chemical analy- parentheses represent a glycoprotein. Downloaded from on August 13, 218 by guest

5 312 NARITA AND MANIRE ses of the isolated complex revealed that the complex consists of approximately 5% protein, 15% carbohydrate, and less than 35% lipid and other constituents (Table 3). Substantially similar results were obtained with the proteincarbohydrate-lipid complex from sucrose gradient centrifugation and from Sepharose 4B gel filtration. The lyophilied complex appeared TABLE 3. Composition of the isolated protein-carbohydrate-lipid complex Components % of dry wt of the complexa Mean Protein Carbohydrate Lipidb < 28- < 39 <32 avalues given represent the range of values obtained for three different preparations. Estimated by subtraction of protein and carbohydrate content from total dry weight I I A B C D J. BACTERIOL. to be insoluble in distilled water, buffer solutions near neutral ph, ethanol, ether, and acetone at room temperature, although sonic treatment resulted in a translucent suspension. The complex was soluble in 5. M urea and detergents such as SDS and Nonidet P-4. When cell envelopes of EB were dissolved in ph 7.2 phosphate buffer with SDS and 2- mercaptoethanol and subjected to polyacrylamide gel electrophoresis in the presence of SDS, more than 2 protein bands were resolved (Fig. 6). Two major bands, designated bands A and B, 17 smaller bands, designated 1 to 15, C and D, and several minor bands were observed. Minor bands included three near band 1, one near band 3, one near band 11, and rarely one below band D. Although the electrophoretic patterns obtained were reproducible, bands C and D stained more clearly in cell envelope preparations that had been subjected to repeated freeing and thawing or storage at 4 C. i a I.' I. I b i I i Downloaded from on August 13, 218 by guest FIG. 6. SDS-polyacrylamide gel electrophoresis of the EB cell envelopes and the protein-carbohydrate-lipid complex obtained after the treatment of EB envelopes with alkali and EDTA. The gel was stained for protein with Coomassie blue. In a schematic illustration of electrophoretic pattern of the envelope, the width of the lines is proportional to the relative intensity of bands. Hatching represents diffuse band. (a) EB cell envelopes; (b) the complex.

6 VOL. 125, 1976 RELEASE OF PROTEIN-CARBOHYDRATE-LIPID COMPLEX 313 The electrophoretic pattern of the isolated protein-carbohydrate-lipid complex derived from alkali solubiliation of EB cell envelopes is shown in Fig. 6b. The complex consists of two major protein bands, corresponding to the positions of bands C and D in the envelopes; bands A and B were completely missing. Densitometric quantitation of the relative amounts of bands C and D revealed that the former represents 5 to 55% of the protein in the complex, with the latter comprising the remaining 5 to 45%. The gels were stained with periodic acid- Schiff reagent to reveal the presence of carbohydrate-containing material (Fig. 7 [ii, a and b ]). Bands A and B were found to be glycoprotein. Bands C and D of both the envelope and the complex gels were stained only very lightly. It should be noted that a few additional periodic acid-schiff stain-positive a b a b c (i) A B C D A a b c o b c 11~~~~~ Downloaded from (ii) B m C D a b c a b A on August 13, 218 by guest AiiB D- FIG. 7. Comparison of electrophoretic patterns of bands A, B, C, and D in gels stained (i) with Coomassie blue to reveal protein, (ii) with a periodic acid-schiff reagent to reveal carbohydrate, and (iii) with Sudan black B to reveal lipid. Dotted line indicates bands that stained lightly or occasionally disappeared. (a) Isolated EB cell envelopes; (b) the complex; (c) cell envelope residue after extraction of the complex with alkali and EDTA.

7 314 NARITA AND MANIRE bands were observed occasionally in gels of the cell envelopes and the complex. Their positions were different from bands A, B, C, and D, and the staining reaction of each band was very weak. When the gels were stained with Sudan black B for lipid, bands A and B in the envelope and bands C and D in the complex were shown to be lipoprotein (Fig. 7 [iii, a and b]). The intensity of the staining reaction of band B was greater than that of band A. These results indicate that the two major bands of the cell envelopes, A and B, and those of the complex, C and D, are composed of protein, carbohydrate, and lipid, although the carbohydrate content of bands C and D appears to be very low. The cell envelope residue obtained after repeated treatment with alkli and EDTA was also dissolved in the same manner with SDS-2-mercaptoethanol in phosphate buffer, and the pattern of bands obtained after SDS-polyacrylamide gel electrophoresis was compared with those of the cell envelopes and the complex (Fig. 7c). The most striking difference was that band A was completely absent in the preparation of the cell envelope residue. There were no significant changes in the electrophoretic pattern of the other bands. Band B was the single major protein band, and this band also stained for the the presence of carbohydrate and lipid. The range of molecular weights of proteins extracted from the cell envelopes was estimated to be between 1, and 11, when compared to the positions of proteins of known molecular weight (Fig. 8). The molecular weights obtained for the two major cell envelope proteins were approximately 85, for band A and 53, for band B. Estimates obtained for the to1 x8 *t l I w4 BAND A 6 B BAND B E -JT T W -J o 2B BANDC BAND D RELATIVE MOBILITY FIG. 8. Estimation of molecular weights of bands A, B, C, and D by SDS-polyacrylamide gel electrophoresis. B, Bovine serum albumin; E, egg albumin; T, trypsin; C, cytochrome c. J. BACTERIOL. two complex peptides were approximately 17, for band C and 13, for band D. DISCUSSION In previous studies reported by Manire and Tamura (13), purified cell walls of Chlamydiae, which had been extracted with SDS and showed no evidence of attached cytoplasmic membranes, contained about 75% protein and 5% phospholipid. Such preparations were found to contain nine major peptide bands in acrylamide gel electrophoresis, and one of these stained as a glycoprotein (23). The subunit structures which line the inner aspect of the cell wall were composed of three peptide bands, one of which stained as a glycoprotein. In the studies reported above, we chose to look at whole cell envelopes prepared as previously reported, but without SDS extraction. These preparations contain both the outer cell and cytoplasmic membranes, as shown in the preceding paper (15). These envelopes were found to contain about 6% protein, 3% carbohydrate, and 3% lipid, with the latter apparently reflecting the SDS solubility of the cytoplasmic membrane. As reported in the previous paper, the incubation of purified EB envelopes in carbonate buffer and EDTA at ph 1 resulted in the release of spherical structures about 6 to 7 nm in diameter, which also appear as rodlike aggregates of two to four spheres (15). We have now demonstrated that the alkaline-edta treatment released a protein-carbohydrate-lipid complex with a molecular weight of several hundred thousand. The released complex represents about 2% of the dry weight of intact envelopes, including 16% of the protein, 4% of the carbohydrate, and 9% of the lipid. A similar release of protein-lipopolysaccharide complexes from isolated cell envelopes of gram-negative bacteria under similar conditions has been reported by Eagon et al. (4, 17). A protein-lipopolysaccharide complex of about 1" molecular weight, which was released from the cell envelopes of Pseudomonas aeruginosa by combined treatment with Tris buffer and EDTA, was reported to be representative of the in situ form of native endotoxin. Electron microscopy examination of that complex revealed rod-shaped structures composed of three spherical units. Leive (9) has reported that EDTA liberates 3 to 5% of lipopolysaccharide from envelopes of Escherichia coli. It has been suggested that lipopolysaccharide may be composed by subunits cross-linked via divalent cations (2). It Downloaded from on August 13, 218 by guest

8 VOL. 125, 1976 seems likely that the effect of EDTA on chlamydial envelopes is due to deprivation of cations, and our results indicate that EDTA is required for extensive release of carbohydrate from the envelopes. Hence, EDTA may preferentially attack the sites at which carbohydrate moieties are linked to other components through multivalent cations. The complex released by alkaline treatment is apparently a true complex, which could not be separated by Sephadex G-2 gel filtration, Sepharose 4B column chromatography, or diethylaminoethyl-cellulose column chromatography. In the case of Sepharose 4B gel filtration, where the eluant ph was 8.5, the protein, carbohydrate, and lipid components emerged together, but there was a broad peak trailing almost the length of the column which suggests that the homogeneity in sie of the released material may be dependent on ph of the eluant. None of the procedures, however, indicated that the carbohydrate, protein, and lipid could be eluted separately. SDS-polyacrylamide gel electrophoresis of EB cell envelopes resulted in a complex pattern of peptide bands similar to those found in typical gram-negative organisms (7, 19, 2, 22) and other biological membranes (8, 18). In contrast to the findings of Tamura et al. (23), who used SDS-extracted cell walls, several of the envelope bands were periodic acid-schiff stain positive. Of the 19 bands from whole envelopes, those designated A, B, C, and D were found to be of most interest. A and B were heavily stained with periodic acid-schiff stain, and C and D were very lightly stained. When the protein-carbohydrate-lipid complex was similarly solubilied and subjected to acrylamide gel electrophoresis, two major bands were found that corresponded to bands C and D. Densitometric quantitation revealed that band C represents 5 to 55% of the complex and band D represents 45 to 5%. They stained very lightly with periodic acid-schiff. It should be noted that occasional periodic acid-schiff stainpositive bands were observed, but their position was different from A, B, C, and D and they stained very lightly. When stained with Sudan black B for lipid, bands A and B from envelopes and C and D from the complex were found to be lipoprotein. Consequently, we can conclude that bands A and B of envelopes and C and D of the complex are composed of protein, lipid, and carbohydrate, although the carbohydrate content of C and D is low. Band A was completely absent in envelope residues after alkali and EDTA treatment. This RELEASE OF PROTEIN-CARBOHYDRATE-LIPID COMPLEX 315 suggests that band A is a component of the protein-carbohydrate-lipid fraction, and it consists of at least two subunit peptides, or, to account for the disparity in molecular weights, band A may consist of several subunits of bands C and D. The cell envelope residues did contain band B. In the preceding paper, morphological studies revealed that EB cell envelopes retained structural rigidity even after exposure to alkali and EDTA, so it is possible that band B may represent a structural protein responsible for maintenance of rigidity. Matsumoto and Manire (14) isolated hexagonal subunits from EB cell walls with hot formamide extraction and centrifugation in a potassium tartrate gradient, and it was suggested that these structures, which are absent in reticulate bodies, are involved in the rigidity of EB. The relationship of band B in cell envelope residues to the peptides of the cell wall subunit, at least one of which Tamura et al. found to contain carbohydrate (23), is not known, but they may well be related. It is not yet possible to assign specific roles to these bands from cell envelopes and the isolated complex, but experiments are in progress to relate immunological activities to them. ACKNOWLEDGMENTS The excellent technical assistance of Eliabeth Brownridge and Samuel Garrison is acknowledged. We are grateful to Sadako Iwata of Nayoya City University for valuable suggestions and discussions, and to Priscilla Wyrick for help in preparing this manuscript. This work was supported in part by Public Health Service grant AI-868 from the National Institute of Allergy and Infectious Diseases. LITERATURE CITED 1. Andrews, P The gel-filtration behaviour of proteins related to their molecular weights over a wide range. Biochem. J. 96: Asbell, M. A., and R. G. Eagon Role of multivalent cations in the organiation, structure, and assembly of the cell wall of Pseudomonas aeruginosa. J. Bacteriol. 92: Bligh, E. G., and W. J. Dyer A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37: Cox, S. T., Jr., and R. G. Eagon Action of ethylenediaminetetraacetic acid, tris(hydroxymethyl)- aminomethane, and lysoyme on cell walls of Pseudomonas aeruginosa. Can. J. Microbiol. 14: Ghuysen, J.-M., D. J. Tipper, and J. L. Strominger Enymes that degrade bacterial cell walls, p In E. F. Neufield and V. Ginsberg (ed.), Methods in enymology, vol. VIII, Complex carbohydrates. Academic Press Inc., New York. 6. Hodge, J. W., and B. T. Hofreiter Determination of reducing sugars and carbohydrates, p In R. L. Whistler and M. L. Wolforms (ed.), Methods in carbo- Downloaded from on August 13, 218 by guest

9 316 NARITA AND MANIRE hydrate chemistry, vol. I. Academic Press Inc., New York. 7. Inouye, M., and M.-L. Yee Homogeneity of envelope proteins of Escherichia coli separated by gel electrophoresis in sodium dodecyl sulfate. J. Bacteriol. 113: Kobylka, D., A. Khettry, B. C. Shin, and K. L. Carraway Proteins and glycoproteins of the erythrocyte membrane. Arch. Biochem. Biophys. 148: Leive, L Release of lipopolysaccharide by EDTA treatment of E. coli. Biochem. Biophys. Res. Commun. 21: Lowry,. H., N. R. Roberts, K. Y. Leiner, M.-L. Wu, and A. L. Farr The quantitative histochemistry of brain. I. Chemical methods. J. Biol. Chem. 27: Lowry,. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: Maiel, J. V Acrylamide gel electrophoresis of proteins and nucleic acid, p In K. Habel and N. P. Salman (ed.), Fundamental techniques in virology. Academic Press Inc., New York. 13. Manire, G. P., and A. Tamura Preparation and chemical composition of the cell walls of mature infectious dense forms of meningopneumonitis organisms. J. Bacteriol. 94: Matsumoto, A., and G. P. Manire Electron microscopic observations on the fine structure of cell walls of Chiamydia psittaci. J. Bacteriol. 14: Narita, T., P. B. Wyrick, and G. P. Manire Effect of alkali on the structure of cell envelopes of Chlamydia psittaci elementary bodies. J. Bacteriol. 125: Park, J. T., and M. J. Johnson A submicrodetermination of glucose. J. Biol. Chem. 181: Rogers, S. W., H. E. Gilleland, Jr., and R. G. Eagon Characteriation of a protein-lipopolysaccharide complex released from cell walls of Pseudomonas aeruginosa by ethylenediaminetetraacetic acid. Can. J. J. BACTERIOL. Microbiol. 15: Rottom, S., and S. Rain Electrophoretic pattems of membrane proteins of Mycoplasma. J. Bacteriol. 94: Schnaitman, C. A Examination of the protein composition of the cell envelope of Escherichia coli by polyacrylamide gel electrophoresis. J. Bacteriol. 14: Schnaitman, C. A Protein composition of the cell wall and cytoplasmic membrane of Escherichia coli. J. Bacteriol. 14: Shapiro, A. L., E. Viniuela, and J. V. Maiel Molecular weight estimation of polypeptide chains by electrophoresis in SDS-polyacrylamide gels. Biochem. Biophys. Res. Commun. 28: Stinnett, J. D., H. E. Gilleland, and R. G. Eagon Proteins released from cell envelopes of Pseudomonas aeruginosa on exposure to ethylenediaminetetraacetate: comparison with dimethylformamide-extractable proteins. J. Bacteriol. 114: Tamura, A., A. Tanaka, and G. P. Manire Separation of the polypeptides of Chlamydia and its cell walls by polyacrylamide gel electrophoresis. J. Bacteriol. 118: Ward, D. N., and M. S. Arnott Gel filtration of proteins, with particular reference to the glycoprotein, luteiniing hormone. Anal. Biochem. 12: Weber, K., and M. Osborn The reliability of molecular weight determinations of dodecyl sulfatepolyacrylamide gel electrophoresis. J. Biol. Chem. 244: Whitaker, J. R Determination of molecular weights of proteins by gel filtration on Sephadex. Anal. Chem. 35: Zacharius, R. M., and T. E. Zell Glycoprotein staining following electrophoresis on acrylamide gels. Anal. Biochem. 3: Downloaded from on August 13, 218 by guest