INFECTION AND IMMUNITY, Feb. 1986, p. 501-5060019-9567/86/020501-06$02.00/0Copyright C 1986, American Society for Microbiology

Isolation, Purification, and Partial Characterization of a

Lipopolysaccharide from Haemophilus pleuropneumoniaeJAMES R. MAUDSLEY,' SOLOMON KADIS,1* AND WILLIAM R. MAYBERRY2

Department of Medical Microbiology, College of Veterinary Medicine, University of Georgia, Athens, Georgia 30602,1and Department of Microbiology, Quillen-Dishner College of Medicine, East Tennessee State University, Johnson City,

Tennessee 376142

Received 17 June 1985/Accepted 10 November 1985

Lipopolysaccharide (LPS) from Haemophilus pleuropneumoniae 1536, $erotype 2, was isolated and purifiedby a procedure designed to be equally satisfactory for both smooth- and rough-type LPS. The LPS yield was

53%. Analysis of the preparations revealed that protein, nucleic acid, and cellular phospholipid contaminationwas negligible (less than 0.1%). Analysis of the sugar content of the LPS by gas-liquid chromatography andcolorimetric analysis revealed the presence of rhamnose, mannose, galactose, glucose, heptose, glucosamine,galactosamine, and 3-deoxy-D-manno-2-octulosonic acid. The heptose and glucose contents appeared to beunusually high. The fatty acids of the LPS consisted of a mixture of C14:0 and C16:0 in a ratio of about 4.5:1(50% of the total) and 3-hydroxy C14:0. When used as a preparatory dose for the dermal Shwartzman reaction,as little as 10 ,ug of the LPS injected intradermally in rabbits produced reddening and swelling. Afterintravenous injection of a 100-,ug LPS provoking dose, necrosis was observed at all intradermal injection sites.Limulus amebocyte lysate gelation was observed with an LPS concentration as low as 0.5 ng/ml. A typicalbiphasic fever response was noted in rabbits injected with as little as 0.25 ng of LPS per kg of body weight.

Haemophilus pleuropneumoniae is the causative agent ofa severe, contagious, often fatal respiratory disease knownas porcine Haemophilus pleuropneumonia (28, 29). Al-though the pathogenesis of this disease has not been eluci-dated completely, experimental studies have revealed a rolefor the lipopolysaccharide (LPS) endotoxin synthesized byH. pleuropneumoniae (28). No definitive studies on H.pleuropneumoniae LPS have been reported thus far. Studieson the LPS of related Haemophilus species, principally H.influenzae, indicate that the LPS may be of a rough-typelacking in smooth-type 0-antigenic side chains (13). It ap-peared conceivable that H. pleuropneumoniae could alsosynthesize a rough LPS. It has also been shown in the pastfew years that the LPS of a number of different bacterialorganisms is heterogeneous in that the LPS from someapparently smooth-type strains lacks a full complement ofthe 0-antigenic side chains (11, 30). To isolate and purify theH. pleuropneumoniae LPS, a method was sought that was

satisfactory in obtaining both smooth- and rough-type LPS.The recently reported method of Darveau and Hancock (6),which appears to satisfy this need, was used to isolate andpurify the H. pleuropneumoniae LPS.The prupose of this study was to obtain a pure LPS

preparation that has been sufficiently characterized withrespect to its chemical composition and biological propertiesto be used in future studies involving the pathogenesis of H.pleuropneumoniae.

MATERIALS AND METHODSBacteria and growth conditions. H. pleuropneumoniae

1536, serotype 2, was obtained from the American TypeCulture Collection (no. 27089) and stored as a lyophilizedpowder. The lyophilized material was reconstituted indeionized water, streaked on tryptic soy agar (DifcoLaboratories, Detroit, Mich.) containing 0.6% yeast extract

* Corresponding author.

and 0.001% NAD, incubated at 37°C for 24 h, and transferredto liquid medium of the same composition and allowed togrow at 37°C until a Klett reading of 160 was attained(approximately 10 h). Ten ml of this culture was inoculatedinto 1,600 ml of fresh cultivation medium. The cells wereincubated at 37°C without shaking for 48 h, harvested bycentrifugation in a Sorvall model RC-2B centrifuge with a

model GS-3 rotor at 13,000 x g for 15 min at 4°C, and washedtwice with deionized water. The fihal pellet was suspended indeionized water and lyophilized.

Isolation of LPS and analysis of purity. The LPS isolationprocedure was initiated with lyophilized bacterial cells bythe method of Darveau and Hancock (6). The final pellet wassuspended in 1.0 ml of deionized water and lyophilized.The purity of the LPS preparations was ascertained by

four different methods. (i) Overall nucleic acid and proteincontent was estimated by absorbance scans from 300 to 200nm with absorbance measurements at 260 and 280 nm. (ii)Protein contamination was also examined by sodium dode-cyl sulfate (SDS)-polyacrylamide gel electrophoresis(PAGE) followed by silver staining (22), and (iii) by a

modified Lowry procedure (20). (iv) Cellular phospholipidcontamination was evaluated by comparing the amount ofC16:1 (a fatty acid found in whole cells but not in LPS)present in whole cells with the amount of this fatty acidpresent in the final LPS product.The SDS-PAGE procedure included a buffer system de-

scribed by Laemmli (18). The separating gel contained 15%(wt/vol) acrylamide-bisacrylamide and 4 M urea. The stack-ing gel contained 4% (wt/vol) acrylamide-bisacrylamide and4 M urea. The loading buffer for protein standards (SigmaChemical Co., St. Louis, Mo.) is described in Sigma techni-cal bulletin MWS-877. Gels were stained by the Morrisseysilver staining procedure (22).The purity of the H. pleuropneumoniae LPS was com-

pared with that of a commercial Escherichia coli O111:B4LPS preparation (Sigma Chemical Co.).

501

Vol. 51, No. 2

on February 25, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

502 MAUDSLEY ET AL.

TABLE 1. Fatty acid content of H. pleuropneumoniae LPS'

Mol%

Fatty Total HCI Alkali-labile Alkali-stableacid' hydrolysate

TMS TFA TMS TFA TMS TFA

C14:0 39 38 50 52C16:0 9 8 31 30C14h 50 49 19 18 100 88

Total fatty acid content was 400 nmol/mg of LPS. Alkali-labile andalkali-stable fatty acid contents were 211 and 134 nmol/mg of LPS, respec-tively.

b All of these fatty acids are normal (straight chain) fatty acids. The suffix hindicates a monohydroxy fatty acid.

Sugar analysis. The LPS was hydrolyzed with 0.5 M HCIat 100°C for 4 h. The released sugars were analyzed as alditolacetates (26, 27) by gas-liquid chromatography (GLC) byusing a Varian model 3700 instrument with a column (0.3 by200 cm) packed with 0.75% HI-EFF-1BP (diethylene gly-col succinate)-0.1% phenyldiethanolamine-0.25% EG-SS-X (ethylenesuccinate methylsilicone copolymer) onGas-Chrom Q 60/80. Xylitol was used as the internal stan-dard.

Fatty acid analysis. All fatty acid analyses of H. pleuro-pneumoniae LPS were performed on 1.0-mg lyophilized LPSsamples by procedures similar to those used by Mayberry(21). Fatty acids were liberated from some samples of LPSby total acid hydrolysis with 2 M HCl at 100°C for 6 h. Othersamples were subjected to direct alkaline methanolysis in10.0 ml of chloroform-methanol (3:2 [vol/vol]) containing0.27 M KOH overnight at room temperature. The reactionmixture was neutralized with 0.3 M HCI and centrifuged in aBeckman TJ-6 centrifuge with a TA-24 rotor at 2,200 x g for30 min at 4°C. The supernatant fluid containing alkali-labilefatty acids (ester linked) was decanted, and the partiallydeacylated LPS pellet was washed by centrifugation asdescribed above with 10.0 ml of chloroform-methanol (3:2),evaporated to dryness, and subjected to acid hydrolysis withHCI as described above. The combined supernatant fluidwas dried.

All of the samples described above were subjected toesterification procedures. Both of the HCl hydrolysates wereextracted in 7.5 ml of chloroform. The aforementionedcombined supernatant fluid sample was reconstituted in 7.5ml of chloroform. Each of these samples contained 300 nmolof 9,10-dihydroxy stearic acid as the internal standard. All ofthe samples were treated with 2.5 ml of methanol and 0.2 mlof HCl (final concentration, 0.25 M) and heated at 55°C for1.0 h to completely esterify the fatty acids. The sampleswere partitioned, washed with water, dried, and treated witha trimethylsilylation mixture to convert hydroxy esters to0-trimethylsilyl (TMS) esters and analyzed by GLC on a25-m capillary column coated with OV-1 (Hewlett-PackardCo., Palo Alto, Calif.) under temperature-programmed con-ditions by using a Hewlett-Packard model 5840A instrumentequipped with flame ionization detectors. After GLC analy-sis as TMS derivatives, samples were converted to 0-trifluoracetyl (TFA) methyl esters by heatitng the samples inmethylene chloride-TFA anhydride for 30 min andreanalyzed. The fatty acid content of E. coli O111:B4 LPSwas determined with O-TFA methyl esters.

Fatty acid analyses of whole cell lyophilized preparationswere conducted as described by Mayberry (21). Relativemolar response factors for fatty acids of interest were

determined by using weighed standards (21), allowing thedetermination of individual fatty acids on a nanomole basiswith subsequent relative contribution to the total profile asmoles percent.

Colorimetric analyses of the LPS. The LPS was analyzedfor 3-deoxy-D-manno-2-octulosonic acid (KDO) by the reac-tion of KDO with thiobarbituric acid by two methods (4, 16).Heptose content was determined by the cysteine-sulfuricacid assay of Wright and Rebers (31) with D-glycero-L-mannoheptose (supplied by Paul Rebers, Ames, Iowa) as thestandard. Total carbohydrate was measured by the phenol-sulfuric acid method (3). Total phosphate was determined bythe Ames method (2). Amino sugars were assayed by amodified Elson-Morgan procedure (7) with glucosamine asthe standard. All analyses of LPS chemical compositionwere also performed with the commercial E. coli O111:B4LPS preparation that was subjected to DNase, RNase, andpronase treatment as specified in the procedure of Darveauand Hancock (6).

Biological properties of the LPS. The dermal Shwartzmanreaction was performed by injecting 2.0- to 2.2-kg male NewZealand albino rabbits intradermally along the latero-medianline of the shaved abdomen with 100-, 50-, 20-, 10-, and 0(control)-p,g doses of LPS in 0.2 ml of sterile, pyrogen-freewater followed 24 h later by a provoking intravenous injec-tion of 100 ,ug of LPS in 0.1 ml of sterile, pyrogen-free water.Injection sites were examined at 24 and 48 h after theinjection of the provoking dose. Lesions were measured incentimeters. Three rabbits received H. pleuropneumoniaeLPS, and three were administered E. coli O111:B4 LPS as apositive control.The Limulus amebocyte lysate (LAL) gel test was per-

formed by using commercially prepared LAL (WhittakerM.A. Bioproducts, Walkersville, Md.) and E. coli O111:B4LPS (Sigma Chemical Co.) as a standard. A test wasconsidered positive when the gelled LAL remained unbro-ken after inversion of the test sample of LPS and LAL.

Pyrogenicity was measured in 1.8- to 2.0-kg male NewZealand albino rabbits after intravenous injection of either0.10, 0.25, or 1.0 jig of LPS per kg of body weight. Injectionswere administered in 0.2 ml of sterile, pyrogen-free water.Each LPS dose was injected into three rabbits. After injec-tion, the temperature of each animal was recorded at 30-minintervals for 6 h with a thermocouple inserted approximately8 cm into the rectum. Sterile, pyrogen-free water and E. c oliO111:B4 LPS were used as controls.

TABLE 2. Fatty acid profiles of H. pleuropneumoniae wholecells

MoI%

TotalFatty Total extract- Total Alkali- Alkali-acid" acid able bound labile stable

hydro- lipid fatty fatty fattylysate fatty acids acids acids

acids

C120 2 1C14:0 36 20 3 31 7Ct6:1 26 43 36C16:0 22 32 17 27 23C181 1 1 1C180 2 2 6 1 5C14h 12 1 73 4 65

" See footnote b, Table 1.

INFECT. IMMUN.

on February 25, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

H. PLEUROPNEUMONIAE LPS 503

*9.a-I ,r

2 3

FIG. 1. Silver-stained SDS-polyacrylamide gel of the LPS of H.pleuropneumoniae 1536, serotype 2, and E. coli 0111:B4. Lane 1,

Protein standards (Sigma Chemical Co.). Top, middle, and bottomarrows indicate molecular weights of 66,000, 45,000, and 29,000,respectively. Each band represents 50 ng of protein. Lane 2, H.pleuropneumoniae LPS. Lane 3, E. coli 0111:B4 LPS. In lanes 2and 3, 20 [tg of LPS was added.

RESULTS

Yield and purity of LPS. To ascertain the effectiveness ofthe procedure used to isolate H. pleuropneumoniae LPS,which was originally designed to isolate and purify bothsmooth and rough LPS (6), the amount of an LPS-specificfatty acid (3-OH C14:0) present in our final LPS preparation(Table 1) was compared with the amount of this fatty acidpresent in whole cells (Table 2). The calculated yield basedon the recovery of this fatty acid was 53%.When UV absorbance scans from 300 to 200 nm were

performed on the final LPS preparations, no absorptionpeaks at 260 or 280 nm were detected, suggesting that atmost only traces of nucleic acid and protein were present ascontaminants in the LPS. In contrast, when a commercial E.coli O111:B4 LPS preparation was subjected to the sameanalysis, a small peak at 260 nm was observed. Calculationsof the nucleic acid content revealed that this LPS may becontaminated with 2.3% nucleic acid. These results taken asa whole indicated that the nucleic acid and protein contentsof the H. pleuropneumoniae LPS were each considerablyless than 0.1%. Moreover, no protein was detected in theLPS when it was subjected to SDS-PAGE followed by silverstaining (Fig. 1). Because this staining procedure reveals

TABLE 4. Gas chromatographic analysis of the sugar content ofH. pleuropneumoniae LPS

Sugar LPS (% [wt/wtj)

Rhamnose ........................................ 5.8Mannose ........................................ 0.4Galactose ........................................ 0.4Glucose ........................................ 36.8Mannoheptose....................................... 13.9Glucoheptose........................................ 10.8Glucosamine ........................................ 1.3Galactosamine....................................... 2.5

polypeptide bands of 1.0 ng or less (22), it was concludedthat any protein contamination in our LPS preparationscould not have been greater than 0.1%. When the H.pleuropneumoniae LPS was examined for the presence ofprotein by a modified Lowry assay (20), it was found that itcontained about 1.0% protein. In contrast, the E. coli0111:B4 LPS preparation contained 3.4% protein by thiscolorimetric analysis. However, because other compounds,including hexosamines, can interfere with this type of pro-tein assay (8, 12), the modified Lowry procedure may not beas reliable in detecting protein contamination in the LPS asare absorbance scanning and silver staining of polyacryl-amide gels.A determination was also made of the extent of cellular

phospholipid contamination in the final LPS preparations.The fatty acid C16:1 was present in whole cells (Table 2) butnot in the LPS (Table 1). Because not even a trace of C16:1was detected in the LPS preparations, it is apparent thatthere was extremely little if any cellular phospholipid in theH. pleuropneumoniae LPS.No attempt was made to detect the specific amount of SDS

contamination in the LPS preparations. Nevertheless, thelipid extraction of Folch et al. (10) was performed to removeany SDS that was present.Chemical analyses of the LPS. Analyses of the chemical

content of the H. pleuropneumoniae LPS compared withthat of the E. coli 0111:B4 LPS are shown in Table 3. Thegas chromatographic analysis of the sugar content of H.pleuropneumoniae LPS is shown in Table 4. Perhaps themost striking features of the sugar profile are the relativelyhigh glucose and heptose values. The total heptose content,consisting of mannoheptose and glucoheptose, was 24.7%.This agreed favorably with the value obtained by colorimet-ric analysis of LPS (25.0%). Colorimetric analysis of theKDO after hydrolysis of the LPS with 0.2 N H2SO4 and 4.0N HCI yielded a KDO content of 0.4 and 1.6%, respectively.The total sugar content of the H. pleuropneumoniae LPS ascalculated from the data in Table 4 was approximately

TABLE 3. Chemical content of the LPS of H. pleuropneumoniae and E. coli 0111:B4

Composition of LPS (% [wt/wtJ)SourceofLPSAmino Total Total TotalSource ofLPSKDHets

sugars KDO Heptose carbohydrate fatty acid phosphate

H. pleuropneumoniae 1536 8.4 0.4" 27.0 36.6 9.2 1.61.6b

E. coli 0111:B4 4.4 1.1" 7.7 27.5 2.0 3.910.b

"LPS hydrolyzed with 0.2 N H2SO.b LPS hydrolyzed with 4.0 N HCI.

1

VOL. 51, 1986

on February 25, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

504 MAUDSLEY ET AL.

TABLE 5. The dermal Shwartzman reaction elicited in rabbits byH. pleuropneumoniae and E. coli 0111:B4 LPS

LPS Preparative Avg size ofdose (,ug) necrotic lesion (cm)

H. pleuropneumoniae 100 1.6 by 1.350 1.4 by 1.120 1.0 by 0.810 0.9 by 0.70 0

E. coli 0111:B4 100 2.4 by 1.350 1.4 by 1.020 0.9 by 0.710 0.5 by 0.50 0

72.7%. This is approximately twice the value obtained fortotal carbohydrate by colorimetric analysis (36.6%). A lowphosphate content (1.6%) was also detected colori-metrically.A detailed analysis of the fatty acid content of H.

pleuropneumoniae LPS, which was determined with 1.0-mgsamples of lyophilized LPS, is given in Table 1. It is apparentthat the LPS contains about 400 nmol of fatty acid per mg ofLPS. Fifty percent of the fatty acid content consists of amixture of C14:0 and C16:0 in a ratio of approximately 4.5:1.The remaining 50% consists of 3-OH C14:0. Assuming anaverage fatty acid molecular weight of 229, the fatty acidsmake up 91.6 jig/mg of LPS or about 9.2% of the total LPS.It should be noted that essentially all of the nonhydroxy fattyacid is alkali labile (ester linked), whereas the hydroxy fattyacid is alkali stable (amide linked). It was also determinedthat the E. coli O111:B4 LPS contains about 80 nmol of fattyacid per mg of LPS. Consequently, by using calculationssimilar to those described above, the total fatty acid contentof the E. coli LPS was 2%.To compare the rather simple fatty acid composition of the

H. pleuropneumoniae LPS that is revealed from the data inTable 1 with the fatty acids that occur in intact H.pleuropneumoniae cells, a detailed analysis of the fatty acidcontent of whole H. pleuropneumoniae cells was conductedwith 25-mg samples of lyophilized cells. It appears that thetotal acid hydrolysate pattern is not uncommon comparedwith that of other Haemophilus species (Table 2; 15). Therewas a very small amount of C120, C18:1, and C18:0. Thenonhydroxy fatty acids, C14:0, C16:1, and C16:0, were presentat the highest levels, 36, 26, and 22%, respectively. Thehydroxy fatty acid, C14h, occurred at a level of about 12%.The extractable lipid and alkali-labile fractions were some-what enriched in C16:1 and C16:0, whereas the hydroxy aqcid inthe extractable lipid fraction was negligible. It is also appar-ent that at least 93% of the hydroxy acid was bound andpresumably in a cell wall component (Table 2). Also, at least71% of the hydroxy acid was in an alkali-stable (amide-linked) linkage.

Biological properties of the LPS. The results of the dermalShwartzman reaction with H. pleuropneumoniae and E. coliO111:B4 LPS are shown in Table 5. All H. pleuro-pneumoniae and E. coli LPS doses tested elicited a positiveresponse.H. pleuropneumoniae LPS produced a firm gel (positive

test) in the LAL test at a concentration of 0.5 ng/ml orhigher. In contrast, E. coli O111:B4 LPS caused firm gela-tion at a concentration of 0.25 ng/ml or higher.Fever response curves for H. pleuropneumoniae LPS in

rabbits are shown in Fig. 2. The lowest dose of H.pleuropneumoniae LPS that produced a biphasic fever re-sponse was 0.25 ,ug/kg of body weight. Positive controls withE. coli O111:B4 LPS gave essentially identical restilts.Control rabbits that were administered only sterile, pyrogen-free water exhibited no increase in temperature.

DISCUSSION

Absorbance scans, SDS-PAGE followed by silver stain-ing, and fatty acid analyses all indicated that the procedureof Darveau and Hancock (6) yields a highly pure H.pleuropneumoniae LPS preparation lacking detectableamounts of nucleic acid, protein, and phospholipid contam-ination. Furthermore, this procedure (6) contains specificsteps to exclude polysaccharide contamination. Studies con-ducted in Hancock's laboratory with antisera and monoclo-nal antibody preparations against capsular polysaccharideisolated and purified from the same strain of the heavilyencapsulated Pseudomonas aeruginosa as that used for theLPS preparations demonstrated conclusively that these LPSpreparations were completely free of contaminating capsularpolysaccharide (R. E. W. Hancock, personal communica-tion). The total LPS yield of 53% was comparable to theyields obtained by Darveau and Hancock for other bacterialspecies.

Analysis of the chemical composition of the H. pleuro-pneumoniae LPS revealed that in some respects it is similarto that of Haemophilus influenzae. As was found with H.influenzae, the KDO in the H. pleuropneumoniae LPS,when hydrolyzed with 0.2 N H2SO4 by the procedure ofKharkhanis et al. (16), is present in much smaller amounts(0.4%) than the KDO detected in the LPS of other bacteriasuch as Salmonella sp. (5 to 8%), P. aeruginosa (4.3%), andBordetella pertussis (4.6%) (9, 17, 19), as well as E. coli

2

0i4 I

2

<I-. I

2

O0

0.lImg/kg

I 1 2 3 4 5 6HOURS

0.25Mjg/kg

I 2 3 4HOURS

5 6

1.0 AMgl kg

O 1 2 3 4 5 6HOURS

FIG. 2. Fever response in rabbits to H. pleuropneumoniae LPS.

INFECT. IMMUN.

on February 25, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

H. PLEUROPNEUMONIAE LPS 505

O111:B4 (1.1%) (this paper). Parr and Bryan found that afterhydrolysis of the H. influenzae LPS with 4 N HCI, theamount of KDO detected was substantially greater than thatobserved after hydrolysis of the LPS with 0.2 N H2S04 (23).In this study, hydrolysis of the H. pleuropneimroniae LPSwith 4 N HCl resulted in a fourfold increase in the KDOcontent. Because hydrolysis of the LPS with 4 N HCI by themethod of Brade et al. (4) frees C-4- and C-5-substitutedKDO, it appears that the KDO in the H. pleuropneiuinoniaeLPS, as well as that in the H. influenzae LPS, is substitutedand unavailable to react in the assay without rigoroushydrolysis. It should also be noted that Raichvarg et al. (24)reported that no KDO was detected in their H. influienzaestrains. Moreover, within a given bacterial species, the KDOcontent of the LPS can vary considerably, as shown by therecent study in which the KDO content of the LPS ofBrucella ovis smooth- and rough-type strains ranged from0.3 to 9.5% (1). It should be noted that in a recent study ofthe composition and antigenicity of the lipid-free oligosac-charide moiety of the H. influenzae type b LPS (14), verylittle KDO was detected by the thiobarbituric acid assay.However, the KDO content as determined by the semi-carbizide assay was 13% for one strain and 34% for anotherstrain.

In contrast to those of H. influenzae LPS, the total fattyacid and phosphate contents of the H. pleuiopneuinoniaeLPS are comparatively low (9, 23). The total carbohydratecontent of the LPS of these two Haemophilus species asdetermined colorimetrically appears to be similar (9). How-ever, our GLC analysis showed that the carbohydrate con-tent of the H. pleuropneumoniae LPS must be at least twicethat obtained by the colorimetric analysis. The value ob-tained by gas-chromatographic analysis was probably a moreaccurate representation of the actual carbohydrate contentof the LPS. The heptose and glucose contents of the H.pleuropneumoniae LPS appeared to be high compared withthe values obtained for these sugars isolated from the LPS ofa variety of bacterial organisms including H. inflienzae (9,23), P. aeruginosa (17), Azospirillum lipoferulm (5), and E.coli O111:B4 (this paper). Comparable heptose values wereobtained for the H. pleuropneumoniae LPS in this study bytwo independent methods of analysis. The relatively highglucose values for the H. pleuropneumoniae LPS comparedwith those obtained for the H. influenzae LPS can perhapsbe accounted for by the fact that the H. influenzae LPS hasbeen reported to be a rough-type LPS (13), whereas thesilver-stained SDS-PAGE preparations of H. pleuropneu-moniae indicated that this LPS is a typical smooth-type LPS.Thus, the increased glucose content could have been areflection of the presence of glucose in the 0-antigenic sidechains of the H. pleuropneumoniae LPS. In addition, itshould be noted that characterization of the H. pleluropneut-moniae LPS has been confined to a representative strain ofonly one serotype. Other known H. pleuropnelimoniaeserotypes (25) may be found to contain different concentra-tions of glucose and possibly other sugars.With respect to the fatty acid composition, it should be

noted that as in other Haemophilus species, the number ofcell wall fatty acids found in H. pleuropneumoniae wholecells is comparatively low (15). The general pattern ischaracterized by relatively large amounts of C14:0, C160,C16:1, and 3-OH C14. The fatty acids C12:0, C18:1, and C18:0 arepresent but in low concentrations. Consequently, it was notsurprising that the H. pleuropneumoniae LPS containedpredominantly three fatty acids (C14:0, C16:0, and 3-OH C14).

Investigations on the LAL, pyrogenicity, and dermal

Shwartzman reactions revealed that H. pleiuropneiumnoniiaeLPS was similar in its biological activity to that of the H.influtzn-e LPS (9) and also to that of E. c oli O111:B4 LPS(this paper).

ACKNOWLEDGMENTSThis study was supported by U.S. Department of Agriculture

Special Animal Health Research grant 84-CRSR-2-2433 and grant85-024 from the Veterinary Medical Research Experiment Station,University of Georgia.The expertise of Joseph Mendicino, Department of Biochemistry,

University of Georgia, in performing GLC analysis of the sugarcontent of the LPS is very gratefully acknowledged.

LITERATURE CITED

1. Afzal, M., R. P Tengerdy, P. G. Squire, and R. P. Ellis. 1984.Characterization of Brlcella oivis lipopolysaccharide and its usefor diagnosis of ram epididymitis by enzyme-linked immunosor-bent assay. J. Clin. Microbiol. 20:1159-1164.

2. Ames, B. N. 1966. Assay of inorganic phosphate, total phos-phate and phosphatases. Methods Enzymol. 8:115-118.

3. Ashwell, G. 1966. New colorimetric methods of sugar analysis.Methods Enzymol. 8:85-95.

4. Brade, H., C. Galanos, and 0. Luderitz. 1983. Differentialdetermination of the 3-deoxy-D-mannooctulosonic acid residuesin lipopolysaccharides of Salninonella mininesotti rough mutants.Eur. J. Biochem. 131:195-220.

5. Choma, A., R. Russa, and Z. Lorkiewicz. 1984. Chemicalcomposition of lipopolysaccharide from Azospir illiiu lipoferiin.FEMS Microbiol. .,ett. 22:245-248.

6. Darveau, R. P., and R. E. W. Hancock. 1983. Procedure forisolation of bacterial lipopolysaccharides from both smooth andrough Pseiudoaionans ieru-i,ginosa and Sal/noanela typhuinuriimnstrains. J. Bacteriol. 155:831-838.

7. Dittmer, J. C., and M. A. Wells. 1969. Quantitative and quali-tative analysis of lipids and lipid components. MethodsEnzymol. 14:482-530.

8. Eichberg, J., and L. C. Mokrasch. 1969. Interference by oxi-dized lipids in the determination of protein by the Lowryprocedure. Anal. Biochem. 30:386-390.

9. Flesher, A. R., and R. A. Insel. 1978. Characterization oflipopolysaccharide of Haemnophlilius influtenzae. J. Infect. Dis.138:719-730.

10. Folch, J., M. Lees, and G. H. S. Stanley. 1957. A simple methodfor the isolation and purification of total lipids from animaltissues. J. Biol. Chem. 226:497-509.

11. Goldman, R. C., and L. Leive. 1980. Heterogeneity of antigenic-side chain length in lipopolysaccharide from Escherichlia coli0111 and Salmoniella typhimnurimn LT2. Eur. J. Biochem.107:145-153.

12. Herd, J. K. 1971. Interference of hexosamines in the Lowryreaction. Anal. Biochem. 44:404-410.

13. Inzana, T. J. 1983. Electrophoretic heterogeneity and interstrainvariation of the lipopolysaccharide of Haemopliiuis influenzae.J. Infect. Dis. 148:492-499.

14. Inzana, T. J., W. E. Seifert, Jr., and R. P. Williams. 1985.Composition and antigenic activity of the oligosaccharide moi-ety of Haeinophliluis infliuenzae type b lipooligosaccharide. In-fect. Immun. 48:324-330.

15. Jantzen, E., B. P. Berdal, and T. Omland. 1981. Cellular fattyacid taxonomy of Haienophlilius, Pasteiurella, and Actinobacil-lias, p. 197-203. In M. Kilian, W. Frederiksen, and E. L.Biberstein (ed.), Haeinophlilus, Pasteiurella, and Actinobacillus.Academic Press, Inc., New York.

16. Kharkkanis, Y. D., J. Y. Zeltner, J. J. Jackson, and D. J. Carlo.1978. A new and improved microassay to determine 2-keto-3-deoxyoctonate in lipopolysaccharide of gram-negative bacteria.Anal. Biochem. 85:595-601.

17. Kropinski, A. M., J. Kuzio, B. L. Angus, and R. E. W. Hancock.1982. Chemical and chromatographic analysis of lipopolysac-charide from an antibiotic-supersusceptible mutant of Psei-

VOL. 51, 1986

on February 25, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

INFECT. IMMUN.

domonas aeruginosa. Antimicrob. Agents Chemother.21:310-319.

18. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London)227:680-685.

19. Le Dur, A., R. Chaby, and L. Szab6. 1980. Isolation of twoprotein-free and chemically different lipopolysaccharides fromBordetella pertussis phenol-extracted endotoxin. J. Bacteriol.143:78-88.

20. Markwell, M. A. K., S. M. Haas, L. L. Bieber, and N. E.Tolbert. 1978. A modification of the Lowry procedure to sim-plify protein determination in membrane and lipoprotein sam-

ples. Anal. Biochem. 87:206-210.21. Mayberry, W. R. 1984. Monohydroxy and dihydroxy fatty acid

composition of Legionella species. Int. J. Syst. Bacteriol.34:321-326.

22. Morrissey, J. H. 1981. Silver stain for proteins in polyacryl-amide gels: a modified procedure with enhanced uniform sensi-tivity. Anal. Biochem. 117:307-310.

23. Parr, T. R., and L. E. Bryan. 1984. Lipopolysaccharide com-

position of three strains of Haemophilus influenzae. Can. J.Microbiol. 30:1184-1187.

24. Raichvarg, D., C. Brossard, and J. Agneray. 1979. Chemicalcomposition and biological activities of phenol-water extract

from Haemophilus influenzae type a. Infect. Immun.26:415-421.

25. Rapp, V. J., R. F. Ross, and B. Z. Erickson. 1985. Serotyping ofHaemophilus pleuropneumoniae by rapid slide agglutinationand indirect fluorescent antibody tests in swine. Am. J. Vet.Res. 46:185-192.

26. Russa, R., and Z. Lorkiewicz. 1979. O-methylheptoses in lipo-polysaccharides. FEMS Microbiol. Lett. 6:71-74.

27. Sawardeker, J. S., J. H. Sloneker, and A. Jeanes. 1965. Quanti-tative determination of monosaccharides as their alditol acetatesby gas liquid chromatography. Anal. Chem. 37:1602-1604.

28. Sebunya, T. N. K., and J. R. Saunders. 1983. Haemophiluspleuropneumoniae infection in swine: a review J. Am. Vet.Med. Assoc. 182:1331-1337.

29. Shope, R. E. 1964. Porcine contagious pleuropneumonia. 1.Experimental transmission, etiology and pathology. J. Exp.Med. 119:357-368.

30. Wilkinson, S. G., and L. Galbraith. 1975. Studies of lipopoly-saccharides from Pseudomonas aeruginosa. Eur. J. Biochem.52:331-343.

31. Wright, B. G., and P. A. Rebers. 1972. Procedure for determin-ing heptose and hexose in lipopolysaccharides-a modificationof the cysteine-sulfuric acid method. Anal. Biochem.49:307-319.

506 MAUDSLEY ET AL.

on February 25, 2020 by guest

http://iai.asm.org/

Dow

nloaded from