Inflammation, Vol. 15, No. 3, 1991
CONCURRENT ENHANCES HUMAN LEUKOCYTES
LIPOPOLYSACCHARIDE
CHEMOTACTIC
RESPONSE
OF
POLYMORPHONUCLEAR TO BACTERIAL
CHEMOTAXIN
H.R. CREAMER, l N. HUNTER, 3 W.W. BULLOCK, ! and W.L. GABLER 2 Departments of ~Oral Microbiology~Immunology and 2Biochemistry, School of Dentistry Oregon Health Sciences University, Portland, OR 97201-3097 3The Dental Research Institute Surry Hills, NSW, Australia
Abstract--Polymorphonuclear neutrophil (PMN) function is thought to be critical in resistance to infectiou ~ agents and this implies that the PMN must be able to migrate into, and to function in, environments that may have high levels of bacterial lipopolysaccharide (LPS). Therefore, we have evaluated the effect of LPS on the in vitro migration of PMNs. Our data reveal that the human PMN is resistant to the deleterious effects of high levels of LPS, that in high concentrations LPS is, itself, a direct chemoattractant for PMNs, and that PMN migration toward a bacterial chemotaxin is enhanced if LPS is also present. Such capabilities suggest that the PMN may be uniquely qualified to :nigrate into microenvironments that are rich in LPS.
INTRODUCTION Bacterial lipopolysaccharide (LPS) has long been a suspected virulence factor in gram-negative infections, and there is ample evidence that the polymorphonuclear neutrophil (PMN) is the primary defensive cell in infectious processes. These points together suggest that the protective roles of PMNs would have to be carried out in LPS-rich environments. Since the PMN's ability to sense and to migrate toward chernotactic agents is considered to be a critical first component of resistance, we have evaluated the ability of PMNs to migrate into environments with relatively high concentrations of LPS. We have found, employing doses of LPS that might have been predicted to inhibit cell function, that PMN migratory responses to the bacterial chemotactic signal n-formyl201
0360-3997/91/0600-0201506.50/09 1991 Plenum PublishingCorporation
202
Creamer et al.
methionyl-leucyl-phenylalanine
(FMLP)
are enhanced when presented concur-
rently with LPS and that LPS itself acts as a chemotaxin for human PMNs.
MATERIALS
AND
METHODS
PMN Preparation. Blood collected from healthy human donors was defibrinated by gentle swirling in flasks containing siliconized, stainless-steel wool pads or heparinized employing 10 units heparin/ml. A PMN-rich cell population was prepared by a modified Ficoll-Hypaque discontinuous gradient method (1). The suspensions typically contained >95% viable PMNs. Working preparations were prepared at a final concentration of 1.4 • 106 PMNs/ml in HBSS without phenol red or sodium bicarbonate but containing 5 ~g streptomycin and 5 units penicillin/ml (all from Gibco, Grand Island, New York) and 5 /zM bis Tris propane buffer (Sigma Chemical Co., St. Louis, Missouri). The final pH of the HBSS was adjusted to 7.2 for the migration studies or to pH 7.5 for the lactic dehydrogenase (LDH) assays (see below). For studies of possible chemokinetic migration, cell suspensions also were prepared to contain varying concentrations of LPS. Migration Assays. Migration studies were carried out by using our recently described assay (2) in which migrated cells are quantitated by measuring the levels of the endogenous cellular enzyme, lactic dehydrogenase (LDH). Employing heat blocks to help stabilize temperature, the lower compartments of polycarbonate blind-well chambers (Neuro Probe, Inc., Cabin John, Maryland) were loaded with 0.2 ml of HBSS, with 2 • 10 -s M FMLP (Sigma) or with various concentrations of LPS (LPS B from Salmonella typhosa #665579, Difco, Detroit, Michigan) or lipid A (from S. minnesota Re595, Calbiochem, San Diego, California) diluted in HBSS. Two polycarbonate membranes (3-/~m pore size, containing polyvinylpyrollidone; Nuclepore Corp., Pleasanton, California) were positioned over the filled, lower chambers. The upper and lower chambers were assembled in sets. Cells that had been preincubated in HBSS for an average of 30 min at 37~ were placed in the upper chambers over buffer to measure spontaneous (random) migration or over FMLP, LPS, or lipid A to measure oriented migration. To study chemokinesis, PMNs were prepared in varying concentrations of LPS so that the same concentration of LPS existed in both the upper and lower chambers. Chambers were incubated for 50 min at 37~ in humidified air. Cell migration was determined by measuring cytoplasmic LDH present on the lower membrane and in the bottom well, which were collected into 3.8 ml of pH 7.5 HBSS containing 0.02% Triton X-100 (Sigma) to lyse the PMNs. For the determination of LDH activity, 1 ml of a solution containing 1.55 mM sodium pyruvate and 0.795 mM NADH (both from Sigma) was added to the tubes containing lysates and membranes. The optical density (OD) at 340 nm was read at time zero and after 1 h of incubation at 37~ and the difference calculated (AOD). Duplicate standards containing from 104 to 105 PMNs were carried through the same assay procedure. Known PMN number was plotted against LDH activity (AOD values) of the standards and the formula of the best-fit straight line was calculated and used to determine the number of PMNs equivalent to the LDH activity measured in the test lysates. Data are expressed as the number of cell equivalents (CE) migrated (2). Superoxide Generation. PMN production of 02 was estimated by the cytochrome e reduction method previously described (3). The O2-generating system consisted of 112 nmol of cytochrome e (Sigma), 10/zmol of glucose, 2/zmol of Ca 2+, 2 • 106 PMNs, and test agents in PBS (final volume 2 ml). 0.2 ml, PBS, containing varying concentrations of LPS or PBS alone was added approximately 2 min prior to activation of the cells with either FMLP (1 • 10 -6 M) or PMA (Sigma; 20 ng/ml). After activation, the systems were incubated at 37~ for an additional 5 rain after which 0.2 ml of 40 mM iodoacetamide was added to each tube to inhibit the Oz-generating reaction. The tubes were centrifuged at 1000g for 5 min at 4~ and the nanomoles of 02 produced
Lipopolysaccharide-EnhancedMigration
203
determined by dividing the 550-nm OD of the resulting supernatants by 21.1 mM/cm (the extinction coefficient of reduced cytochrome c). The amount of non-Q-dependent reduction of cytochrome c was corrected for by using th,~ enzyme superoxide dismutase. The data are given as nanomoles of SOD-inhibitable 02 generatec per 5 min per 1 • 106 PMNs.
RESULTS Little information appears to exist concerning the effect of concurrent LPS or LPS pretreatment of PMNs on their migration toward microbial products, especially at high concentrations of LPS. As a first approximation of potential LPS effects, we tested for the possible presence of LPS-induced membrane damage by measuring the release of a cytoplasmic marker, lactic dehydrogenase (LDH). As shown in Figure 1, significant LDH release was not induced by incubating PMNs for 10 rain in up to 625 #g of LPS/ml. An additional replicated study employing 1000/zg LPS/ml also revealed no statistically significant loss of LDH from the PMNs (data not presented). The data in Figure 2 show that PMNs pretreated with up to 25 #g LPS/ml remain responsive to bolh FMLP- and PMA-induced generation of superoxide. FMLP-induced superox de synthesis was significantly inhibited at 125/zg LPS/
4 3
.J
2
0 0
1
5
25
125
62,5
LPS (ug/rnl) Fig. 1. Lactic dehydrogenase ~LDH) release as a percentage of total cellular LDH. Released LDH in PMN supernatants was measured after a 10-rain incubation at 37~ in the LPS concentrations indicated. Mean values and standard errors of the means from two duplicate experiments are shown.
2t)4
Creamer et al.
I
I HSSS
~
FM._P
PMA
~
"E" -E
50
to Z
40
o..
"' o
///
30
"/,,5
,/~
H ,
_.e o
2O
.'9-
10
X
o t,_ o.. O3
"/~
"//,
E
t-.
"/,,6,
?'/,
0
:t 0
1
5
25
125
625
LPS ( u g / m l ) Fig. 2. Effect of LPS on superoxide generation by PMNs following 2 min of preincubation in the indicated concentrations of LPS. HBSS is control buffer; FMLP was at 1 • 10 -6 M concentration and PMA at 20 ng/ml. The data points are the means and standard error bars of two duplicate experiments, a = significantly different (P < 0.01) from control response to FMLP or PMA in the absence of LPS as determined by Dunnett's test for comparison of multiple treatment means with a control.
ml and PMA-induced synthesis was inhibited at an LPS concentration of 625 t~g/ml. Since these data together indicated that PMNs are relatively resistant to possible toxic effects of LPS, we next studied the effect of LPS on spontaneous migration. PMN suspensions were prepared and preincubated for 30 min in the same concentration of LPS as was present in the bottom wells of the migration chambers (i.e., in a chemokinetic mode). We found no evidence of an effect on spontaneous migration (i.e., no chemokinetic effect) at concentrations up to 625/zg/ml (Figure 3) but did find there was an inhibitory effect on spontaneous migration at the highest concentration tested (1 mg LPS/ml; Figure 3). In contrast, a chemotactic effect was seen when the cells were preincubated in HBSS and were placed over bottom wells containing varying concentrations of LPS (i.e., the chemotactic mode). The highest concentrations of LPS tested (450, 625, and 1000/zg/ml) each induced significantly enhanced migration as compared to the spontaneous migration value (Figure 3). In an attempt to demonstrate that the chemotactic effect seen was due to LPS and not to a contaminant, migration studies were carded out using autoclaved LPS, lipid A, and combi-
205
Lipopolysaccharide-Enhanced Migration
r---]
Spontaneous
~
Chemotaxis
70 b
60 a
50 A
uJ
40
8_ 30 v
20
0 0
1
25
125 450 LPS ( u g / m l )
625
1000
Fig. 3. Direct effect of LPS oa PMN migration. LPS was present either in the bottom wells only (chemotaxis mode) or in both the cell preparations and in the bottom wells (chemokinesis mode). The means and standard error,~ of the cell equivalents (CE) migrated, calculated from the median of triplicate values from two to 12 replicate experiments, are shown, a, P < 0.05; and b, P < 0.01 as compared with migra:ion in the absence of LPS as indicated by Dunnett's test for the comparison of multiple means with a control.
nations o f LPS and polymyxin B. As shown in Figure 4, autoclaving did not affect the ability o f the LPS preparation to induce chemotactic migration, and lipid A induced significant migration when it was presented in a chemotactic mode at a concentration of 125/zg/ml. AdditionallY, polymyxin B at a concentration o f 200/xg/ml significantly inhibited P M N chemotactic migration toward LPS but not toward F M L P (Table 1). The heat stability o f LPS (4), the chemotactic effect o f lipid A, and the significant neutralization o f LPS-induced chemotaxis by p o l y m y x i a B (5) all argue strongly that LPS, not a contaminant, induced the chemotactic migration seen. In order to evaluate whether P M N s can sense a bacterial-derived chemotactic signal in the presence o f LPS, as would be necessary in foci of gramnegative infections, P M N s were prepared in HBSS and placed over bottom wells containing 2 x 10 -8 M F M L P , 125 /~g LPS/ml, or a combination o f F M L P and LPS in HBSS. The data (Table 2) reveal that P M N s can respond to F M L P presented in an environment o f LPS. In fact, it is seen that such an environment enhances the migration o f P M N s toward F M L P . Chemotactic migration toward
206
Creamer et al.
F~
Lipid A
Autoclaved LPS
80 70
a
60 A
50 40 30 20 10 0 0
5
25 ug/ml
125
450
Fig. 4. Cell equivalent chemotactic migration values (means and standard deviations of one triplicate experiment) as induced by lipid A and by autoclaved LPS; (a) values are significantly greater (P < 0.01) than spontaneous migration (0) as determined by Dunnett's test for the comparison of multiple means with a control.
Table 1. Polymyxin B Neutralization of LPS Migration 9 Bottom well contentsa
CE
SE
N
HBSS PMB LPS LPS + PMB FMLP FMLP + PMB
11,824 9,559 48,620 21,162 c 96,657 96,730
1,373 4,080 6,631 5,520 11,960 10,646
3 3 3 2 2 2
~The bottom wells contained either HBSS, or LPS at 425 #g/ml, FMLP at 2 x 10 -8 M/liter, PMB at 200 #g/ml, LPS and PMB, or FMLP and PMB all in HBSS at the indicated concentrations. i, Average of the median cell equivalents (CE) migrated in triplicated studies, standard error (SE) of the mean and number (N) of experiments. CThe value is significantly less (P < 0.01) than the control LPS-induced chemotaxis value (LPS). Statistical evaluations were performed using the Newman-Keul multiple range test of means.
Lipopolysaeeharide-Enhane,~dMigration
207
Table 2. Migration of PMNs in Presence of LPS Cell preparations b Bottom well contents ~
HBSS
LPS
HBSS
15,587 " 3,635 N=5
nd
LPS
32,924 4,013 N=3
25,087 4,270 N=4
FMLP
61,011 a 7,665 N=4
nd
LPS + FlVlLP
97,347 J'e 5,365 N=3
53,492 J 3,315 N=4
aThe botto'n wells contained either HBSS, LPS at 125 #g/ ml, FMLP at 2 x 10 _8 M/liter, or both LPS and FMLP in HBSS. hPrior to loading the chambers, 1.4 x 106 PMNs/ml were preincubaled at 37~ an average of 30 rain in either HBSS or in 125 ,zg/ml LPS in HBSS. "Average cf the median CE migrated of triplicate studies, standard e nor of the mean and the number of experiments. aValues are significantly greater (P < 0.01) than spontaneous migration values (HBSS). ~Value is significantly greater (P < 0.01) than the control chemotaxi~ value (FMLP). Statistical evaluations were performed using the Newman-Keul multiple range test of means. ND = not done. c o n c u r r e n t l y p r e s e n t e d Y M L P in L P S a v e r a g e d nearly 60% greater than the m i g r a t i o n toward F M L P alone. M i g r a t i o n e n h a n c e m e n t does not appear to h a v e b e e n due entirely to an additive response since the increased m i g r a t o r y response seen to L P S as c o m p a r e d to H B S S w o u l d a c c o u n t for less than h a l f o f the increased r e s p o n s e to F M L P that o c c u r r e d in the p r e s e n c e o f L P S (Table 1). T o o u r k n o w l e d g e migraLion e n h a n c e m e n t due to concurrent signaling with L P S has not b e e n d e s c r i b e d lzreviously.
DISCUSSION O u r data s h o w that high concentrations (up to 1 m g / m l ) o f L P S are not toxic to h u m a n P M N s as d e m o n s t r a t e d by the insignificant loss o f c y t o p l a s m i c L D H f o l l o w i n g i n c u b a t i o n in L P S ( F i g u r e 1 and text). It was o f s o m e surprise
208
Creamer et ai.
to us originally that PMNs incubated in such concentrations of LPS could be resuspended easily and that they remained viable and able to migrate (Table 1, Figure 3). The inhibition of superoxide production seen at these concentrations of LPS (Figure 2), however, indicates that some cellular modification was induced. The modification probably was not FMLP receptor-related since responses to PMA also were inhibited (albeit at a higher concentration). We are aware of no published studies of possible chemokinetic effects of LPS tested at several concentrations. Pugliese et al. (6) have reported decreased migration of human peripheral blood PMNs when the cells and the bottom wells contained Salmonella enteriditis LPS in buffer at 5/zg/ml, i.e., when LPS was presented in a chemokinetic mode, and Dahinden et al. (7) reported similar findings. We found no effects at low LPS concentrations but did find inhibition of spontaneous migration at the highest concentration of LPS tested (Figure 3). The literature concerning the effects of LPS as a possible direct inducer of chemotactic or chemokinetic PMN migration is relatively sparse and is inconsistent in content. Snyderman et al. (8) employed rabbit peritoneal PMNs and reported that 20/zg/ml Seratia marscescens LPS in buffer was not chemotaxic. Similarly, Issekutz and Bhimji (9) reported that up to 1 mg/ml of E. coli LPS did not induce chemotactic responses by rabbit peripheral PMNs employed in an under-agarose, assay and Moeller et al. (10) reported that LPS was not a chemoattractant for mouse peripheral PMNs in Boyden chamber assays. In contrast, there are recent reports that microgram quantities of LPS induced chemotactic migration by human PMNs (11, 12). The variations in LPS and PMN sources and in the migration assays employed allow little direct comparison of the data concerning the possible chemotactic effect of LPS. We found that chemotactic migration toward a bacterial chemoattractant in LPS was not depressed, but was in fact enhanced (Table 1). The basis of the enhancing effect is not known, but it appears not to be due to an additive effect since the same concentration of LPS alone (125/~g/ml) did not induce significant chemotactic migration (Figure 3). The enhancement might be viewed as an example of concurrent LPS "priming" since LPS pretreatment has been reported to enhance certain PMN responses to subsequent agonists. For example, priming of human PMNs with nanogram quantities of LPS has been reported to enhance FMLP- and PMA-induced superoxide generation (13-16), FMLPinduced lysozyme release (13, 17), and the synthesis of leukotriene B 4 following stimulation with zymosan, PMA, and the catcium-ionophore A23187 (18). However, Bremm et al. (19) have reported the opposite with respect to LPS priming and subsequent leukotriene synthesis. These contrasting results with respect to effects on subsequent leukotriene synthesis (18, 19) do not appear to be due to differences in concentrations of LPS employed or the PMN or LPS source. Likewise, there is conflicting evidence with respect to whether LPSpriming of PMNs leads to an increased expression of FMLP receptors (15, 16)
Lipopolysaccharide-EnhancedMigration
209
and Haslett et al. (13) found that LPS-priming resulted in a significant reduction in subsequent FMLP-induced migration. Thus it is not clear what accounts for our finding that LPS presented concurrently with FMLP in a gradient mode induces significantly greater migration than either FMLP or LPS alone. The possibilities appear to be (1) an up-regulation of FMLP receptors, (2) the recruitment of a larger pool of FMLP-responsive ceils, or (3) a facilitation of the FMLP-receptor signal-transduction pathway. Our data and that c f others (20, 21) clearly indicate that the PMN is resistant to deleterious effecls of LPS. We show here that in vitro PMN migration toward a bacterial chemotactic signal is significantly enhanced when both agonist and LPS are presenled in gradient modes. In addition we have demonstrated that high concentrations of LPS presented in a gradient mode act as a direct chemoattractant for human PMNs. Our findings are consistent with certain known clinical situations in the human in which very large numbers of PMNs migrate into LPS-rich environments. From the steady-state data of Klinkhamer (22), it can be calculated that more than 1 x 10 l~ PMNs/day migrate through inflamed tissues associated with severe periodontitis into microbe-rich environments (periodontal pockets) that are known to contain claemoattractants (23) and LPS (24-26). The local concentration of LPS in such areas could be quite high, at least on the leading face of the migrating PMNs, and similar to the conditions presented in the in vitro assays reported here. Our data indicate that PMNs are quite resistant to deleterious effects of LPS and that LPS, in fact, enhances the oriented migration of PMNs toward bacterial attractants. These findings, coupled with the demonstration that PMNs ca~ enzymatically deacylate and detoxify LPS (27, 28), support the contention that the PMN may be uniquely qualified to migrate into and to carry out the peripheral neutralization of LPS in foci of gram-negative infections.
REFERENCES
1. BOYUM,A. 1968. Isolationof mononuclearcells and granulocytesfrom human blood. Scand. J. Lab. Invest. 21(suppl):77-89. 2. CREAMER, H.R., W.L. GABLER,and W.W. BULLOCK.1983. Endogenous component chemotactic assay (ECCA). rnflammation 7:321-329. 3. GABLER, W.W., H.R. C:~EAMER,and W.W. BULLOCK.1986. Modulation of the kinetics of induced neutrophil superoxide generation by fluoride. J. Dent. Res. 65:1159-1165. 4. MILNER,K.C., J.A. RUDBACH,and E. Rml. 1971. General characteristics. In Microbial Tox-
ins, Vol. IV, Bacterial Endotoxins. G. Weinbaum, S. Kadis and S.J. Ajl, editors. Academic Press, New York. 1-65.
210
Creamer et al.
5. ISSEKUTZ,A.C. 1983. Removal of gram-negative endotoxin from solutions by affinity chromatography. J. Immunol. Methods 61:275-281. 6. PUGLIESE, C., M.D. LASALLE, and V.A. DEBARI. 1988. Relationships between the structure and function of lipopolysaccharide chemotypes with regard to their effects on the human polymorphonuclear neutrophil. Mol. lmmunol. 25:631-637. 7. DAHINDEN,C., C. GALANOS,and J. FEHR. 1983. Granulocyte activation by endotoxin I. Correlation between adherence and other granulocyte functions, and role of endotoxin structure on biologic activity. J. Immunol. 130:857-862. 8. SNYDERMAN,R., H. GEWURZ,and S.E. MERGENHAGEN. 1968. Interactions of the complement system with endotoxic lipopolysaccharide. J. Exp. Med. 128:259-275. 9. ISSEKUTZ,A.C., and S. BHIMJI. 1982. Role for endotoxin in the leukocyte infiltration accompanying Escherichia coli inflammation. Infect. Immun. 36:558-566. 10. MOELLER, G.R., L. TERRY, and R. SNYDERMAN. 1978. The inflammatory response and resistance to endotoxin in mice. J. lmmunol. 120:116-123. 11. KOTANI, S., H. TAKADA, M. TSUJIMOTO, et al. 1985. Synthetic lipid A with endotoxic and related biological activities comparable to those of a natural lipid A from an Escherichia coli Re-mutant. Infect. Immun. 49:225-237. 12. BOGAR, L., Z. MOLNAR, and M. TEKERES. 1988. A new simplified assay for evaluation of motile activity of human polymorphonuclear leucocytes. Immunol. Lett. 17:211-216. 13. HASLETT, C., L.A. GUTHRIE, M.M. KOPANIAK, R.B. JOHNSTON, and P.M. HENSON. 1985. Modulation of multiple neutrophil functions by preparative methods or trace concentrations of bacterial lipopolysaccharide. Am. J. Pathol. 119:101-110. 14. WRIGHT, G.G., and G.L. MANDELL. 1986. Anthrax toxin blocks priming of neutrophils by lipopolysaccharide and by mummyl dipeptide. J. Exp. Med. 164:1700-1709. 15. FOREHAND, J.R., M.J. PABST, W.A. PHILIPS, and R.B. JOHNSTON, JR. 1989. Lipopolysaccharide priming of human neutrophils for an enhanced respiratory burst--role of intracellular free calcium. J. Clin. Invest. 83:74-83. 16. VOSBECK, K., P. TOBIAS, H. MUELLER, et al. 1990. Priming of polymorphonuclear granulocytes by lipopolysaccharides and its complexes with lipopolysaccharide binding protein and high density lipoprotein. J. Leukocyte Biol. 47:97-104. 17. MURPHY, P., and D.A. HART. 1987. Regulation of enzyme release from human polymorphonuclear leukocytes: Further evidence for the independent regulation of granule subpopulations. Biochem. Cell Biol. 65:1007-1015. 18. DOERFLER, M.E., R.L. DANNER, J.H. SHELHAMER,and J.E. PARRILLO. 1989. Bacterial lipopolysaccharides prime human neutrophils for enhanced production of leukotriene B4. J. Clin. Invest. 83:970-977. 19. BREMM,K.D., W. KONIG, M. THELESTAM,and J.E. ALOUF. 1987. Modulation of granulocyte functions by bacterial exotoxins and endotoxins. Immunology 62:363-371. 20. ISSEKUTZ,A.C., and W.D. BIGGAR. 1977. Influence of serum-derived chemotactic factors and bacterial products on human neutrophil chemotaxis. Infect. Immun. 15:212-220. 21. HENRICKS, P.A.J., M.E. VANDER TOL, R.W.W.M. THYSSEN, B.S. VAN ASBECK,and J. VERHOEF. 1983. Escherichia coli lipopolysaccharides diminish and enhance cell function of human polymorphonuclear leukocytes. Infect. lmmun. 41:294-301. 22. KLINKHAMER,J.M. 1968. Quantitative evaluation of gingivitis and periodontal disease I. The orogranulocytic migratory rate. Periodontics 6:207-211. 23. TEMPEL, T.R., R. SNYDERMAN, H.V. JORDAN, and S.E. MERGENHAGEN. 1970. Factors from saliva and oral bacteria, chemotactic for polymorphonuclear leukocytes: Their possible role in gingival inflammation. J. Periodontol. 41:71-80. 24. SHAPmO, L., F.M. LODATO, P.R. COURANT, and R.E. STALLARD. 1972. Endotoxin determinations in gingival inflammation. J. Periodontol. 43:591-596,
Lipopolysaccharide-EnhancedMigration
211
25. SIMON, B., H. GOLDMA?,!,M. RUBEN, and E. BAKER. 1971. The role of endotoxin in periodontal disease. III. Correlation of the quantity of endotoxin in human gingival exudate with the clinical degree of inf ammation. J. Periodontol. 42:210-216. 26. McCoY, S.A., H.R. CREAMER,M. KAWANAMI,and D.F. ADAMS. 1987. The concentration of lipopolysaccharide on individual root surfaces at varying times following in vivo root planing. J. Periodontol. 58:2~93-399. 27. HALL, C.L., and R.S. MUNFORD. 1983. Enzymatic deacylation of the lipid A moeity of Salmonella typhimurium lip()polysaccharides by human neutrophils. Proc. Natl. Acad. Sci. U.S.A. 80:6671-6675. 28. DAL NOGARE, A.R., ant W.C. YARBROUGH,JR. 1990. A comparison of the effects of intact and deacylated lipopolysaccharide on human polymorphonuclear leukocytes. J. Immunol. 144:1404-1410.
CONCURRENT ENHANCES HUMAN LEUKOCYTES
LIPOPOLYSACCHARIDE
CHEMOTACTIC
RESPONSE
OF
POLYMORPHONUCLEAR TO BACTERIAL
CHEMOTAXIN
H.R. CREAMER, l N. HUNTER, 3 W.W. BULLOCK, ! and W.L. GABLER 2 Departments of ~Oral Microbiology~Immunology and 2Biochemistry, School of Dentistry Oregon Health Sciences University, Portland, OR 97201-3097 3The Dental Research Institute Surry Hills, NSW, Australia
Abstract--Polymorphonuclear neutrophil (PMN) function is thought to be critical in resistance to infectiou ~ agents and this implies that the PMN must be able to migrate into, and to function in, environments that may have high levels of bacterial lipopolysaccharide (LPS). Therefore, we have evaluated the effect of LPS on the in vitro migration of PMNs. Our data reveal that the human PMN is resistant to the deleterious effects of high levels of LPS, that in high concentrations LPS is, itself, a direct chemoattractant for PMNs, and that PMN migration toward a bacterial chemotaxin is enhanced if LPS is also present. Such capabilities suggest that the PMN may be uniquely qualified to :nigrate into microenvironments that are rich in LPS.
INTRODUCTION Bacterial lipopolysaccharide (LPS) has long been a suspected virulence factor in gram-negative infections, and there is ample evidence that the polymorphonuclear neutrophil (PMN) is the primary defensive cell in infectious processes. These points together suggest that the protective roles of PMNs would have to be carried out in LPS-rich environments. Since the PMN's ability to sense and to migrate toward chernotactic agents is considered to be a critical first component of resistance, we have evaluated the ability of PMNs to migrate into environments with relatively high concentrations of LPS. We have found, employing doses of LPS that might have been predicted to inhibit cell function, that PMN migratory responses to the bacterial chemotactic signal n-formyl201
0360-3997/91/0600-0201506.50/09 1991 Plenum PublishingCorporation
202
Creamer et al.
methionyl-leucyl-phenylalanine
(FMLP)
are enhanced when presented concur-
rently with LPS and that LPS itself acts as a chemotaxin for human PMNs.
MATERIALS
AND
METHODS
PMN Preparation. Blood collected from healthy human donors was defibrinated by gentle swirling in flasks containing siliconized, stainless-steel wool pads or heparinized employing 10 units heparin/ml. A PMN-rich cell population was prepared by a modified Ficoll-Hypaque discontinuous gradient method (1). The suspensions typically contained >95% viable PMNs. Working preparations were prepared at a final concentration of 1.4 • 106 PMNs/ml in HBSS without phenol red or sodium bicarbonate but containing 5 ~g streptomycin and 5 units penicillin/ml (all from Gibco, Grand Island, New York) and 5 /zM bis Tris propane buffer (Sigma Chemical Co., St. Louis, Missouri). The final pH of the HBSS was adjusted to 7.2 for the migration studies or to pH 7.5 for the lactic dehydrogenase (LDH) assays (see below). For studies of possible chemokinetic migration, cell suspensions also were prepared to contain varying concentrations of LPS. Migration Assays. Migration studies were carried out by using our recently described assay (2) in which migrated cells are quantitated by measuring the levels of the endogenous cellular enzyme, lactic dehydrogenase (LDH). Employing heat blocks to help stabilize temperature, the lower compartments of polycarbonate blind-well chambers (Neuro Probe, Inc., Cabin John, Maryland) were loaded with 0.2 ml of HBSS, with 2 • 10 -s M FMLP (Sigma) or with various concentrations of LPS (LPS B from Salmonella typhosa #665579, Difco, Detroit, Michigan) or lipid A (from S. minnesota Re595, Calbiochem, San Diego, California) diluted in HBSS. Two polycarbonate membranes (3-/~m pore size, containing polyvinylpyrollidone; Nuclepore Corp., Pleasanton, California) were positioned over the filled, lower chambers. The upper and lower chambers were assembled in sets. Cells that had been preincubated in HBSS for an average of 30 min at 37~ were placed in the upper chambers over buffer to measure spontaneous (random) migration or over FMLP, LPS, or lipid A to measure oriented migration. To study chemokinesis, PMNs were prepared in varying concentrations of LPS so that the same concentration of LPS existed in both the upper and lower chambers. Chambers were incubated for 50 min at 37~ in humidified air. Cell migration was determined by measuring cytoplasmic LDH present on the lower membrane and in the bottom well, which were collected into 3.8 ml of pH 7.5 HBSS containing 0.02% Triton X-100 (Sigma) to lyse the PMNs. For the determination of LDH activity, 1 ml of a solution containing 1.55 mM sodium pyruvate and 0.795 mM NADH (both from Sigma) was added to the tubes containing lysates and membranes. The optical density (OD) at 340 nm was read at time zero and after 1 h of incubation at 37~ and the difference calculated (AOD). Duplicate standards containing from 104 to 105 PMNs were carried through the same assay procedure. Known PMN number was plotted against LDH activity (AOD values) of the standards and the formula of the best-fit straight line was calculated and used to determine the number of PMNs equivalent to the LDH activity measured in the test lysates. Data are expressed as the number of cell equivalents (CE) migrated (2). Superoxide Generation. PMN production of 02 was estimated by the cytochrome e reduction method previously described (3). The O2-generating system consisted of 112 nmol of cytochrome e (Sigma), 10/zmol of glucose, 2/zmol of Ca 2+, 2 • 106 PMNs, and test agents in PBS (final volume 2 ml). 0.2 ml, PBS, containing varying concentrations of LPS or PBS alone was added approximately 2 min prior to activation of the cells with either FMLP (1 • 10 -6 M) or PMA (Sigma; 20 ng/ml). After activation, the systems were incubated at 37~ for an additional 5 rain after which 0.2 ml of 40 mM iodoacetamide was added to each tube to inhibit the Oz-generating reaction. The tubes were centrifuged at 1000g for 5 min at 4~ and the nanomoles of 02 produced
Lipopolysaccharide-EnhancedMigration
203
determined by dividing the 550-nm OD of the resulting supernatants by 21.1 mM/cm (the extinction coefficient of reduced cytochrome c). The amount of non-Q-dependent reduction of cytochrome c was corrected for by using th,~ enzyme superoxide dismutase. The data are given as nanomoles of SOD-inhibitable 02 generatec per 5 min per 1 • 106 PMNs.
RESULTS Little information appears to exist concerning the effect of concurrent LPS or LPS pretreatment of PMNs on their migration toward microbial products, especially at high concentrations of LPS. As a first approximation of potential LPS effects, we tested for the possible presence of LPS-induced membrane damage by measuring the release of a cytoplasmic marker, lactic dehydrogenase (LDH). As shown in Figure 1, significant LDH release was not induced by incubating PMNs for 10 rain in up to 625 #g of LPS/ml. An additional replicated study employing 1000/zg LPS/ml also revealed no statistically significant loss of LDH from the PMNs (data not presented). The data in Figure 2 show that PMNs pretreated with up to 25 #g LPS/ml remain responsive to bolh FMLP- and PMA-induced generation of superoxide. FMLP-induced superox de synthesis was significantly inhibited at 125/zg LPS/
4 3
.J
2
0 0
1
5
25
125
62,5
LPS (ug/rnl) Fig. 1. Lactic dehydrogenase ~LDH) release as a percentage of total cellular LDH. Released LDH in PMN supernatants was measured after a 10-rain incubation at 37~ in the LPS concentrations indicated. Mean values and standard errors of the means from two duplicate experiments are shown.
2t)4
Creamer et al.
I
I HSSS
~
FM._P
PMA
~
"E" -E
50
to Z
40
o..
"' o
///
30
"/,,5
,/~
H ,
_.e o
2O
.'9-
10
X
o t,_ o.. O3
"/~
"//,
E
t-.
"/,,6,
?'/,
0
:t 0
1
5
25
125
625
LPS ( u g / m l ) Fig. 2. Effect of LPS on superoxide generation by PMNs following 2 min of preincubation in the indicated concentrations of LPS. HBSS is control buffer; FMLP was at 1 • 10 -6 M concentration and PMA at 20 ng/ml. The data points are the means and standard error bars of two duplicate experiments, a = significantly different (P < 0.01) from control response to FMLP or PMA in the absence of LPS as determined by Dunnett's test for comparison of multiple treatment means with a control.
ml and PMA-induced synthesis was inhibited at an LPS concentration of 625 t~g/ml. Since these data together indicated that PMNs are relatively resistant to possible toxic effects of LPS, we next studied the effect of LPS on spontaneous migration. PMN suspensions were prepared and preincubated for 30 min in the same concentration of LPS as was present in the bottom wells of the migration chambers (i.e., in a chemokinetic mode). We found no evidence of an effect on spontaneous migration (i.e., no chemokinetic effect) at concentrations up to 625/zg/ml (Figure 3) but did find there was an inhibitory effect on spontaneous migration at the highest concentration tested (1 mg LPS/ml; Figure 3). In contrast, a chemotactic effect was seen when the cells were preincubated in HBSS and were placed over bottom wells containing varying concentrations of LPS (i.e., the chemotactic mode). The highest concentrations of LPS tested (450, 625, and 1000/zg/ml) each induced significantly enhanced migration as compared to the spontaneous migration value (Figure 3). In an attempt to demonstrate that the chemotactic effect seen was due to LPS and not to a contaminant, migration studies were carded out using autoclaved LPS, lipid A, and combi-
205
Lipopolysaccharide-Enhanced Migration
r---]
Spontaneous
~
Chemotaxis
70 b
60 a
50 A
uJ
40
8_ 30 v
20
0 0
1
25
125 450 LPS ( u g / m l )
625
1000
Fig. 3. Direct effect of LPS oa PMN migration. LPS was present either in the bottom wells only (chemotaxis mode) or in both the cell preparations and in the bottom wells (chemokinesis mode). The means and standard error,~ of the cell equivalents (CE) migrated, calculated from the median of triplicate values from two to 12 replicate experiments, are shown, a, P < 0.05; and b, P < 0.01 as compared with migra:ion in the absence of LPS as indicated by Dunnett's test for the comparison of multiple means with a control.
nations o f LPS and polymyxin B. As shown in Figure 4, autoclaving did not affect the ability o f the LPS preparation to induce chemotactic migration, and lipid A induced significant migration when it was presented in a chemotactic mode at a concentration of 125/zg/ml. AdditionallY, polymyxin B at a concentration o f 200/xg/ml significantly inhibited P M N chemotactic migration toward LPS but not toward F M L P (Table 1). The heat stability o f LPS (4), the chemotactic effect o f lipid A, and the significant neutralization o f LPS-induced chemotaxis by p o l y m y x i a B (5) all argue strongly that LPS, not a contaminant, induced the chemotactic migration seen. In order to evaluate whether P M N s can sense a bacterial-derived chemotactic signal in the presence o f LPS, as would be necessary in foci of gramnegative infections, P M N s were prepared in HBSS and placed over bottom wells containing 2 x 10 -8 M F M L P , 125 /~g LPS/ml, or a combination o f F M L P and LPS in HBSS. The data (Table 2) reveal that P M N s can respond to F M L P presented in an environment o f LPS. In fact, it is seen that such an environment enhances the migration o f P M N s toward F M L P . Chemotactic migration toward
206
Creamer et al.
F~
Lipid A
Autoclaved LPS
80 70
a
60 A
50 40 30 20 10 0 0
5
25 ug/ml
125
450
Fig. 4. Cell equivalent chemotactic migration values (means and standard deviations of one triplicate experiment) as induced by lipid A and by autoclaved LPS; (a) values are significantly greater (P < 0.01) than spontaneous migration (0) as determined by Dunnett's test for the comparison of multiple means with a control.
Table 1. Polymyxin B Neutralization of LPS Migration 9 Bottom well contentsa
CE
SE
N
HBSS PMB LPS LPS + PMB FMLP FMLP + PMB
11,824 9,559 48,620 21,162 c 96,657 96,730
1,373 4,080 6,631 5,520 11,960 10,646
3 3 3 2 2 2
~The bottom wells contained either HBSS, or LPS at 425 #g/ml, FMLP at 2 x 10 -8 M/liter, PMB at 200 #g/ml, LPS and PMB, or FMLP and PMB all in HBSS at the indicated concentrations. i, Average of the median cell equivalents (CE) migrated in triplicated studies, standard error (SE) of the mean and number (N) of experiments. CThe value is significantly less (P < 0.01) than the control LPS-induced chemotaxis value (LPS). Statistical evaluations were performed using the Newman-Keul multiple range test of means.
Lipopolysaeeharide-Enhane,~dMigration
207
Table 2. Migration of PMNs in Presence of LPS Cell preparations b Bottom well contents ~
HBSS
LPS
HBSS
15,587 " 3,635 N=5
nd
LPS
32,924 4,013 N=3
25,087 4,270 N=4
FMLP
61,011 a 7,665 N=4
nd
LPS + FlVlLP
97,347 J'e 5,365 N=3
53,492 J 3,315 N=4
aThe botto'n wells contained either HBSS, LPS at 125 #g/ ml, FMLP at 2 x 10 _8 M/liter, or both LPS and FMLP in HBSS. hPrior to loading the chambers, 1.4 x 106 PMNs/ml were preincubaled at 37~ an average of 30 rain in either HBSS or in 125 ,zg/ml LPS in HBSS. "Average cf the median CE migrated of triplicate studies, standard e nor of the mean and the number of experiments. aValues are significantly greater (P < 0.01) than spontaneous migration values (HBSS). ~Value is significantly greater (P < 0.01) than the control chemotaxi~ value (FMLP). Statistical evaluations were performed using the Newman-Keul multiple range test of means. ND = not done. c o n c u r r e n t l y p r e s e n t e d Y M L P in L P S a v e r a g e d nearly 60% greater than the m i g r a t i o n toward F M L P alone. M i g r a t i o n e n h a n c e m e n t does not appear to h a v e b e e n due entirely to an additive response since the increased m i g r a t o r y response seen to L P S as c o m p a r e d to H B S S w o u l d a c c o u n t for less than h a l f o f the increased r e s p o n s e to F M L P that o c c u r r e d in the p r e s e n c e o f L P S (Table 1). T o o u r k n o w l e d g e migraLion e n h a n c e m e n t due to concurrent signaling with L P S has not b e e n d e s c r i b e d lzreviously.
DISCUSSION O u r data s h o w that high concentrations (up to 1 m g / m l ) o f L P S are not toxic to h u m a n P M N s as d e m o n s t r a t e d by the insignificant loss o f c y t o p l a s m i c L D H f o l l o w i n g i n c u b a t i o n in L P S ( F i g u r e 1 and text). It was o f s o m e surprise
208
Creamer et ai.
to us originally that PMNs incubated in such concentrations of LPS could be resuspended easily and that they remained viable and able to migrate (Table 1, Figure 3). The inhibition of superoxide production seen at these concentrations of LPS (Figure 2), however, indicates that some cellular modification was induced. The modification probably was not FMLP receptor-related since responses to PMA also were inhibited (albeit at a higher concentration). We are aware of no published studies of possible chemokinetic effects of LPS tested at several concentrations. Pugliese et al. (6) have reported decreased migration of human peripheral blood PMNs when the cells and the bottom wells contained Salmonella enteriditis LPS in buffer at 5/zg/ml, i.e., when LPS was presented in a chemokinetic mode, and Dahinden et al. (7) reported similar findings. We found no effects at low LPS concentrations but did find inhibition of spontaneous migration at the highest concentration of LPS tested (Figure 3). The literature concerning the effects of LPS as a possible direct inducer of chemotactic or chemokinetic PMN migration is relatively sparse and is inconsistent in content. Snyderman et al. (8) employed rabbit peritoneal PMNs and reported that 20/zg/ml Seratia marscescens LPS in buffer was not chemotaxic. Similarly, Issekutz and Bhimji (9) reported that up to 1 mg/ml of E. coli LPS did not induce chemotactic responses by rabbit peripheral PMNs employed in an under-agarose, assay and Moeller et al. (10) reported that LPS was not a chemoattractant for mouse peripheral PMNs in Boyden chamber assays. In contrast, there are recent reports that microgram quantities of LPS induced chemotactic migration by human PMNs (11, 12). The variations in LPS and PMN sources and in the migration assays employed allow little direct comparison of the data concerning the possible chemotactic effect of LPS. We found that chemotactic migration toward a bacterial chemoattractant in LPS was not depressed, but was in fact enhanced (Table 1). The basis of the enhancing effect is not known, but it appears not to be due to an additive effect since the same concentration of LPS alone (125/~g/ml) did not induce significant chemotactic migration (Figure 3). The enhancement might be viewed as an example of concurrent LPS "priming" since LPS pretreatment has been reported to enhance certain PMN responses to subsequent agonists. For example, priming of human PMNs with nanogram quantities of LPS has been reported to enhance FMLP- and PMA-induced superoxide generation (13-16), FMLPinduced lysozyme release (13, 17), and the synthesis of leukotriene B 4 following stimulation with zymosan, PMA, and the catcium-ionophore A23187 (18). However, Bremm et al. (19) have reported the opposite with respect to LPS priming and subsequent leukotriene synthesis. These contrasting results with respect to effects on subsequent leukotriene synthesis (18, 19) do not appear to be due to differences in concentrations of LPS employed or the PMN or LPS source. Likewise, there is conflicting evidence with respect to whether LPSpriming of PMNs leads to an increased expression of FMLP receptors (15, 16)
Lipopolysaccharide-EnhancedMigration
209
and Haslett et al. (13) found that LPS-priming resulted in a significant reduction in subsequent FMLP-induced migration. Thus it is not clear what accounts for our finding that LPS presented concurrently with FMLP in a gradient mode induces significantly greater migration than either FMLP or LPS alone. The possibilities appear to be (1) an up-regulation of FMLP receptors, (2) the recruitment of a larger pool of FMLP-responsive ceils, or (3) a facilitation of the FMLP-receptor signal-transduction pathway. Our data and that c f others (20, 21) clearly indicate that the PMN is resistant to deleterious effecls of LPS. We show here that in vitro PMN migration toward a bacterial chemotactic signal is significantly enhanced when both agonist and LPS are presenled in gradient modes. In addition we have demonstrated that high concentrations of LPS presented in a gradient mode act as a direct chemoattractant for human PMNs. Our findings are consistent with certain known clinical situations in the human in which very large numbers of PMNs migrate into LPS-rich environments. From the steady-state data of Klinkhamer (22), it can be calculated that more than 1 x 10 l~ PMNs/day migrate through inflamed tissues associated with severe periodontitis into microbe-rich environments (periodontal pockets) that are known to contain claemoattractants (23) and LPS (24-26). The local concentration of LPS in such areas could be quite high, at least on the leading face of the migrating PMNs, and similar to the conditions presented in the in vitro assays reported here. Our data indicate that PMNs are quite resistant to deleterious effects of LPS and that LPS, in fact, enhances the oriented migration of PMNs toward bacterial attractants. These findings, coupled with the demonstration that PMNs ca~ enzymatically deacylate and detoxify LPS (27, 28), support the contention that the PMN may be uniquely qualified to migrate into and to carry out the peripheral neutralization of LPS in foci of gram-negative infections.
REFERENCES
1. BOYUM,A. 1968. Isolationof mononuclearcells and granulocytesfrom human blood. Scand. J. Lab. Invest. 21(suppl):77-89. 2. CREAMER, H.R., W.L. GABLER,and W.W. BULLOCK.1983. Endogenous component chemotactic assay (ECCA). rnflammation 7:321-329. 3. GABLER, W.W., H.R. C:~EAMER,and W.W. BULLOCK.1986. Modulation of the kinetics of induced neutrophil superoxide generation by fluoride. J. Dent. Res. 65:1159-1165. 4. MILNER,K.C., J.A. RUDBACH,and E. Rml. 1971. General characteristics. In Microbial Tox-
ins, Vol. IV, Bacterial Endotoxins. G. Weinbaum, S. Kadis and S.J. Ajl, editors. Academic Press, New York. 1-65.
210
Creamer et al.
5. ISSEKUTZ,A.C. 1983. Removal of gram-negative endotoxin from solutions by affinity chromatography. J. Immunol. Methods 61:275-281. 6. PUGLIESE, C., M.D. LASALLE, and V.A. DEBARI. 1988. Relationships between the structure and function of lipopolysaccharide chemotypes with regard to their effects on the human polymorphonuclear neutrophil. Mol. lmmunol. 25:631-637. 7. DAHINDEN,C., C. GALANOS,and J. FEHR. 1983. Granulocyte activation by endotoxin I. Correlation between adherence and other granulocyte functions, and role of endotoxin structure on biologic activity. J. Immunol. 130:857-862. 8. SNYDERMAN,R., H. GEWURZ,and S.E. MERGENHAGEN. 1968. Interactions of the complement system with endotoxic lipopolysaccharide. J. Exp. Med. 128:259-275. 9. ISSEKUTZ,A.C., and S. BHIMJI. 1982. Role for endotoxin in the leukocyte infiltration accompanying Escherichia coli inflammation. Infect. Immun. 36:558-566. 10. MOELLER, G.R., L. TERRY, and R. SNYDERMAN. 1978. The inflammatory response and resistance to endotoxin in mice. J. lmmunol. 120:116-123. 11. KOTANI, S., H. TAKADA, M. TSUJIMOTO, et al. 1985. Synthetic lipid A with endotoxic and related biological activities comparable to those of a natural lipid A from an Escherichia coli Re-mutant. Infect. Immun. 49:225-237. 12. BOGAR, L., Z. MOLNAR, and M. TEKERES. 1988. A new simplified assay for evaluation of motile activity of human polymorphonuclear leucocytes. Immunol. Lett. 17:211-216. 13. HASLETT, C., L.A. GUTHRIE, M.M. KOPANIAK, R.B. JOHNSTON, and P.M. HENSON. 1985. Modulation of multiple neutrophil functions by preparative methods or trace concentrations of bacterial lipopolysaccharide. Am. J. Pathol. 119:101-110. 14. WRIGHT, G.G., and G.L. MANDELL. 1986. Anthrax toxin blocks priming of neutrophils by lipopolysaccharide and by mummyl dipeptide. J. Exp. Med. 164:1700-1709. 15. FOREHAND, J.R., M.J. PABST, W.A. PHILIPS, and R.B. JOHNSTON, JR. 1989. Lipopolysaccharide priming of human neutrophils for an enhanced respiratory burst--role of intracellular free calcium. J. Clin. Invest. 83:74-83. 16. VOSBECK, K., P. TOBIAS, H. MUELLER, et al. 1990. Priming of polymorphonuclear granulocytes by lipopolysaccharides and its complexes with lipopolysaccharide binding protein and high density lipoprotein. J. Leukocyte Biol. 47:97-104. 17. MURPHY, P., and D.A. HART. 1987. Regulation of enzyme release from human polymorphonuclear leukocytes: Further evidence for the independent regulation of granule subpopulations. Biochem. Cell Biol. 65:1007-1015. 18. DOERFLER, M.E., R.L. DANNER, J.H. SHELHAMER,and J.E. PARRILLO. 1989. Bacterial lipopolysaccharides prime human neutrophils for enhanced production of leukotriene B4. J. Clin. Invest. 83:970-977. 19. BREMM,K.D., W. KONIG, M. THELESTAM,and J.E. ALOUF. 1987. Modulation of granulocyte functions by bacterial exotoxins and endotoxins. Immunology 62:363-371. 20. ISSEKUTZ,A.C., and W.D. BIGGAR. 1977. Influence of serum-derived chemotactic factors and bacterial products on human neutrophil chemotaxis. Infect. Immun. 15:212-220. 21. HENRICKS, P.A.J., M.E. VANDER TOL, R.W.W.M. THYSSEN, B.S. VAN ASBECK,and J. VERHOEF. 1983. Escherichia coli lipopolysaccharides diminish and enhance cell function of human polymorphonuclear leukocytes. Infect. lmmun. 41:294-301. 22. KLINKHAMER,J.M. 1968. Quantitative evaluation of gingivitis and periodontal disease I. The orogranulocytic migratory rate. Periodontics 6:207-211. 23. TEMPEL, T.R., R. SNYDERMAN, H.V. JORDAN, and S.E. MERGENHAGEN. 1970. Factors from saliva and oral bacteria, chemotactic for polymorphonuclear leukocytes: Their possible role in gingival inflammation. J. Periodontol. 41:71-80. 24. SHAPmO, L., F.M. LODATO, P.R. COURANT, and R.E. STALLARD. 1972. Endotoxin determinations in gingival inflammation. J. Periodontol. 43:591-596,
Lipopolysaccharide-EnhancedMigration
211
25. SIMON, B., H. GOLDMA?,!,M. RUBEN, and E. BAKER. 1971. The role of endotoxin in periodontal disease. III. Correlation of the quantity of endotoxin in human gingival exudate with the clinical degree of inf ammation. J. Periodontol. 42:210-216. 26. McCoY, S.A., H.R. CREAMER,M. KAWANAMI,and D.F. ADAMS. 1987. The concentration of lipopolysaccharide on individual root surfaces at varying times following in vivo root planing. J. Periodontol. 58:2~93-399. 27. HALL, C.L., and R.S. MUNFORD. 1983. Enzymatic deacylation of the lipid A moeity of Salmonella typhimurium lip()polysaccharides by human neutrophils. Proc. Natl. Acad. Sci. U.S.A. 80:6671-6675. 28. DAL NOGARE, A.R., ant W.C. YARBROUGH,JR. 1990. A comparison of the effects of intact and deacylated lipopolysaccharide on human polymorphonuclear leukocytes. J. Immunol. 144:1404-1410.
