Short Exposure of Polymorphonuclear Leucocytes to Sodium Fluoride Suppresses the Response to fMLP

JOll' n31 of

J. Trace Elements Med. BioI. Vol. 13, pp. 68-75 (1999)

Trace Elements In Medicine and Biology

Clinical Study

© 1999 Urban & Fischer

Short Exposure of Polymorphonuclear Leucocytes to Sodium Fluoride Suppresses the Response to fMLP E. KNOLL-KOHLER and G. BUSSEM* Institute of Pharmacology and *Clinic of Dentistry, Medical Centre Benjamin Franklin Free University of Berlin, Berlin, Germany (Received July/October 1998)

Summary

Fluoridated dental care products are used to prevent dental decay. Up to now, there are no data available on whether the fluoride (F) component of these products affects the bactericidal activity of salivary polymorphonuclear leucocytes, which are involved in the protection of the oral mucosa against infection. Therefore, after determining the concentration/time profile of F in mixed saliva of healthy subjects after topical application of 0.5 g of a 1.25 % Fcontaining gel, unstimulated and fMLP-stimulated polymorphonuclear leucocytes (PMNs) were shortly exposed to these F concentrations and the generation of superoxide and hypochloric acid were measured, as well as the liberation of lysomal enzymes, and correlated with the cellular Ca2+and cAMP-levels. The results show that F, at concentrations as retained in saliva, did not activate the oxidative burst in unstimulated PMNs. In tMLP-activated PMNs, P- suppressed the receptor-mediated increase in the oxidative burst and the liberation of fl-glucuronidase by reduction of the availability of extracellular Ca2+ and, thus, the influx of Ca2+necessary to couple completely the fMLP signal to effector pathways. These P- concentrations neither altered the liberation of Ca2+from internal stores nor induced a rise in cAMP. The possible clinical consequences of these results for xerostomic patients with respect to the generation of HOSCN/OSCN/SCN- in saliva, an important non-immune factor for oral health, are dicussed.

Keywords: polymorphonuclear leucocytes, fluoride, [Ca2+]i, cAMP, 02 -, dentistry. Abbreviations: Ca2+ : calcium ion; [Ca2+]i : internal Ca2+ concentration; cAMP: cyclic adenosine monophosphate; F: fluoride ion; IBMX: 3-isobutyl-l-methylxanthine; fMLP: N-formyl-L-methionyl-L-Ieucyl-L-phenylalanine; 02 -: superoxide anion; PMNs: polymorphonuclear leucocytes.

Introduction

Fluorides (F) are therapeutically employed to prevent dental decay. In addition to fluoridation of public water, fluoridated dental care products are increasingly being Reprint requests to:Prof. Dr. E. Knoll-Kohler, Medical Centre Benjamin Franklin, Free University of Berlin,Institute of Pharmacology, Thielallee 69-72, D-14195 Berlin, Germany

used. In these products, the F concentrations vary between 1 and 600 mM P-. The most important mode of action of F in caries prevention is probably the enhancement of tooth mineralization and inhibition of its demineralization (1-3). The antimicrobial action of F on plaque bacteria does not contribute to the antic aries activity in vivo even when one uses dental care products with [F] up to 600 mM, since effective concentrations are not

Exposure of polymorphonuclear leucocytes to sodium fluoride

69

maintained for a sufficient period of time (4). No data are

ers, with full dentition without tooth restorations took

available that would indicate whether topically applied

part in the study after signing a written consent. At 9 a.m.,

fluorides (F) affect the bactericidal activity of polymorphonuclear leucocytes (PMN s), which constantly migrate into the oral cavity in response to chemotactic signals (5). PMN s are part of the intraoral defense mechanism against invading microorganisms, since they are viable, functionally active and respond as primed cells to fMLP by additional O2 ' generation (6-12). Fluorides in the form of fluoroaluminate are known to trigger the oxidative burst in PMNs and to liberate lysosomal enzymes by acting on the a-subunit of the G-protein and mimicking the effect of g phosphate of GTP and thus the signal trans-

after collecting a baseline unstimulated whole saliva sample, each subject brushed his teeth with 0.5 g Elmex

duction sequence of physiological agonists (13-17). To obtain these effects, dosages of more than 10 mM F must be applied for at least 5 min. The purpose of the present study was to investigate whether F, at concentrations as measured in mixed saliva after topical application of 0.5 g of a 1.25 % Fcontaining gel (=350 mM F), has any effect on the generation of superoxide radicals and exocytosis of lysosomal enzymes in unstimulated and fMLP-activated PMNs.

Gel 0.25 % F, Wybert Company, Lorrach) for 2 min under the supervision of the dentist, spat out the gel/saliva slurry and rinsed his mouth once for 10 s with 10.0 ml of tridistilled water. Thereafter, the saliva was collected into preweighed polyethylene vials for 1 min at the intervals shown in Figure 1, sonicated for 30s and kept on ice until the [F] was measured with and without TISAB using the fluoride electrode Orion, model 96-09, at 37°C. Isolation of PMNs

Blood obtained from healthy nonsmoking members of our clinical staff (males and females, average age 25±5 years) was used for the experiments. All subjects were carefully interviewed to exclude medication that could affect PMN functions. Consent to participation was obtained in writing. PMNs were separated by Ficoll Hypaque density gradient centrifugation after dextran sedimentation and hypotonic lysis of contaminating erythro-

Materials and Methods Superoxide dismutase (SOD), catalase, N-formyl-Lmethionyl-L-Ieucyl-L-phenylalanine (fMLP) , ferricytochrome c (type III) and penta-acetoxy-methyl ester of fura-2 (fura-2/AM) were purchased from Sigma, IBMX from Calbiochem, F-127 pluronic acid and forskolin from Fluca; Ficol Hypaque from Biomol, Fico1l400 from Pharmacia, Uppsala Sweden; DMEM from Seromed and all other cel! culture reagents from Gibco BRL, Burlington. The actual F concentration in the respective assay medium, in which NaCl was reduced in proportion to the addition of NaP, was determined with the fluoride electrode Orion, model 96-09. The concentration of Ca2+ in the F- containing buffer was fluorometrically determined using fura-2 as the Ca2+ indicator. Throughout the experiments tridistilled water and plastic tubes were used to avoid contamination with AICI 3, which is known to be a component of many types of glass, especially since F in solution can etch glass. Saliva collection after F- gel application

Adult subjects (n=15), aged 25 ± 2 years, nonsmok-

cytes (8). The viability of purified PMNs was 99 % as assessed by the trypan blue exclusion. Ten minutes before starting functional measurements, PMNs were transferred to a 1.0 mM CaCI 2- containing phosphate bufferto replenish intracellular calcium stores. Assay for superoxide measurement

Dismutase-inhibitable superoxide release by activat100,00

•• - - - - - - - - - - - - - - - - - - - - ,

lD.QO

,i 0,10

0,01 ~~-----------------'

baseline gel/saliva 2 slurry

~

15

JG

lime after gel application (min)

Figurel. Concentrationltime profile of P- in mixed saliva after brushing the teeth with 0.5 g of the 1.25 % F-gel for two minutes. (n = 12 subjects)

70

E. Knoll-Kohler and G. Biissem

ed neutrophils was spectrophotometric ally quantitated as the ferricytochrome c reduction rate at 550 nm and 37 DC. The assay was carried out with 0.5 x 106 PMNs in 1.0 ml polystyrene cuvettes containing 100 llM ferricytochrome c in phosphate-buffered saline (pH=7.4) with (mM) NaCl (100), KCI (5), MgCl 2 x 6 Hp (1), CaCl 2 (l), glucose (5) (= buffer A). In some experiments the buffer contained 2 mM EGTA instead of CaCI 2• Absorbance changes at 550 nm were continuously monitored in a double-beam spec-

cator of primary (azurophilic) granule release by spectrophotometrically measuring the formation of 4-nitrophenol at 420 nm using 4-nitrophenyl-fl-D-glucuronide as substrate after a 2 hr incubation period at 38 DC. The direct effect of F on enzyme activity was tested with Triton X-100 lysed cells and was found to be negligible. cAMP measurement

trophotometer in which the content of the reference cu-

106 PMNs were pre incubated in buffer A in the ab-

vette was identical to that in the sample cuvette except for the additional presence of superoxide dismutase (SOD 0.6 mg/ml). The superoxide production rate was calculated using an extinction coefficient of29.5 mM. According to (19) Fdoes not affect the SOD enzyme at the concentrations used in our experiments.

sence or presence of the phosphodiesterase inhibitor 3isobutyl-l-methylxanthine (IBMX, 400 llM) and various concentrations of F for 5 min and 3 min, respectively, at 37 DC. After stimulation with fMLP (0.1 llM) the reactions were stopped by adding 0.1 vol. of 20 % ice-cold perchloric acid at time-intervals as shown in Figures 5 and 6; the PMN's were then incubated on ice for 20 min and centrifugated. After neutralisation of the supernatant,

Assay for hypochlorous acid measurement

cAMP was acetylated and determined in 100 llL aliquots The release of HOCl/OCI· into the medium was quantified as taurine chloramine resulting from the reaction of hypochlorous acid and taurine (20, 21). Taurine chloramine was determined by oxidation of TNB (5-thio-2nitrobenzoic acid) to DTNB (5,5'-dithiobis (2-nitrobenzoic acid), which was spectrophotometric ally detected at 412 nm and 25 DC. PMNs (0.5 x 106) were suspended in 0.5 ml buffer A, pH =7.4, with 20 mM taurine at 37 DC. The cells were stimulated with fMLP and incubated under continuous shaking for 15 min. The reaction was terminated by adding catalase (42 U/mL) and placing the tubes in melting ice for 10 min. After the cells had been pelleted (10 min at 2500 g, 4 DC) TNB (10 llM) was added to the carefully aspirated supernatants. There was no interference of F at the tested concentration range with the enzymes of the assay at pH =7.4. A 1 ml solution of TNB was prepared by raising the pH of 2 mM DTNB in 50 mM phosphate buffer from 7.4 to 12 and readjusting it to 7.4 after 5 min to promote its hydrolysis (22). Absorbance was recorded at 412 nm and the concentration of the hypochlorous acid produced was calculated using the molar extinction coefficient of 28.2. Degranulation assay

PMNs (1 x 106/ml) were incubated in buffer A and treated with cytochalasin B (2 llg/ml) for 5 min prior to exposure to F and activation by fMLP (lllM) for 2 min. After centrifugation, the released fl-glucuronidase activity was determined in the cell-free supernatant as an indi-

by the PRA 509 assay system of Amersham. Ca 2+ measurement

For Ca2+ measurements in single cells, 105 purified PMNs /ml DMEM, supplemented with 10 % heat-inactivated fetal calf serum, were immobilized onto 60 mm autoclaved coverslips in a Greiner dish and allowed to attach to it for 8 hours in a humidified atmosphere of 5 % CO/95 % air at 37 DC. To load the cells, the medium was replaced by 1.0 ml DMEM containing 1 llM Fura-2/AM and IllM pluronic acid F127 in DMSO and the cells were further incubated in darkness for 20 min at 37 DC. The coverslips were then removed from the medium and washed three times in buffer A. [Ca2+] was determined from dual wavelength measurements at excitation wavelengths of 340 and 380 nm and an emission wavelength of 510 nm using an epifluorescence microscope (Axovert 100, Zeiss Company). The fluorescence signals were recorded by a camera and digitalized with the Photomics software, program FUCAI, version 5.12 B. Experiments were calibrated and Ca2+ concentrations calculated according to the technique of (23), using ionomycin for cell lysis. The Ca2+ response to fMLP in single cells was checked with cells in suspension according to the method detailed in 23. With both methods identical values were found for unstimulated and stimulated cells (80 ± 20 and 450 ± 50 nM, respectively, M ( SD of 10 experiments).

Exposure of polymorphonuclear leucocytes to sodium fluoride

0-

Results

71

20~------------------.--,

2

[PJ in mixed saliva after topical application of the 1.25 % P gel

15 (fJ

~

43 ± 13 mM F- was measured in the gel/saliva slurry after teeth were brushed for two minutes with 0.5 g of the

'00..

o

10

'6

1.25 % F- containing gel (Figure 1). After rinsing with

E

::

10.0 mL tridistilled water for 10 s, the [F] quickly de-

5

creased with great interindividual variations. The average [F] in mixed saliva two min after application was 6.2 ±

3.8 mM. This value dropped to 1.4 ± 0.4 mM F- four min

O~--~----,---~----~---,----~

o

after application and below 1.0 mM F- during the next two min. At the end of our observation period, the average [F ] was 5.7 times above the baseline level (0.15 ± 0.08 mM

2

1

3

4

5

6

lime (min)

0-2

20.---------------,

F- vs. 0.026 ± 0.019 mM F-). 15

Effect of NaF on O2 and HOCl/OCZ- generation

(fJ

~

"-

02- generation of fMLP-activated PMNs was linear


=sE

during the first four minutes. About 18 nmoles O2- were

10

"

cumulatively generated per 106 cells. In nominally Ca2+-

5

free medium, the linear 02- generation time was shortened (Figure 2a) and the amount of generated 02 - was reD~

duced by about 50 % independent of whether EGTA was added to the cells during the preincubation period or simultaneously with fMLP. No release of 02- from unstim-

o

__

~

____

~

2

__

_ L _ _ _ _ ~_ _ _ L _ _ ~

3

4

5

6

lime (min)

0-

25~------------------~

2

ulated cells was detected. F- treatment decreased the generation of 02- in a dose-

20

and time-dependent manner. It took more than 180 and 60 sec, respectively, to suppress 02- generation when F- and fMLP were simultaneously added to the reaction vessels to reach a final F-Ievel of 2.7 or 6.2 mM (Figure 2b). Preincubation for 3 min with 2.7 mM F- (final concentration) suppressed the 02- generation to a greater extent than chelating extracellular Ca2+ with EGTA (Figure 2c).

(fJ

~ 0..

15

'6 10 E

::

Data compiled in Figure 3 show that 106 fMLP (1~M)-activated

PMNs produced about 45

±5

nmoles

HOClIOCI- within 15 min. Their generation was inhibited by adding EGTA or catalase, which demonstrated that the measured HOClIOCI was produced outside the cell and dependent on extracellular Ca2+. F- reduced HOClIOC1production by about 90 % at 2.3 mM.

Effect of NaF on intracellular and extracellular [Ca 2+J Since Ca2+ is involved in fMLP signal-effector coupling, intracellular [Ca2+] was monitored in relation to

1

2

3

4

5

6

lime (min)

Figure 2. (A) Generation of superoxide in response to fMLP (100 nM) in the presence of extracellular Ca2+
E. Knoll-Kohler and G. Biissem

72

100..-- - - - - - -

70

6.2 mM .

:0

o

10

20

30

40

50

60

Iii Figure 3. Generation of HOCLIOCL- by fMLP-activated PMNs pretreated with F (fMLP 111M; M ± SD from 4 separate experiments)

Figure 5. Effect of F on cellular cAMP during fMLP-activation of PMNs.(Data represent mean values from two separate experiments with 4 determinations each. Baseline value: 0.080 pMlI06 PMNs)

NaP-dosing. As shown in Figure 4, F reduced the rise in [Caz+]i in response to tMLP in the presence of extracellu-

Effect of NaF on cellular cAMP

lar Ca z+. At 2.7 ruM F the tMLP-mediated increase in [Caz+]i was identical to that observed in cells kept in nom-

Recent studies indicate that elevation of the intracellular cAMP level is associated with the inhibition of nu-

inally Caz+-free medium. F affected neither the tMLP-in-

merous PMN responses to tMLP (11,24-26). We there-

presence of EGTA or Mnz+. Parallel measurements of ionized [Caz+] in the extracellular medium showed a gradual drop in [Ca2+] with increasing [F).

fore measured intracellular cAMP to ascertain whether cAMP is involved in the F-induced decrease in the 0zgeneration oftMLP-activated PMNs. Figure 5 illustrates the time course of changes in cAMP after tMLP stimulation in the absence of the phosphodiesterase inhibitor

duced liberation of Ca z+ from internal stores nor the influx of Mnz+, as shown by [Caz+]i measurements in the

500 20:0~----------------------------------'

400

f

-

~

c: ....... + (\J

:ooc

-

500

nl

U

b

0

~

200

c 1tvtJ>

Figure 4. Effect of different [F) on the [Ca2+]i signal in response to fMLP-activation ofPMNs. (bar un-filled: [Ca2+] in the extracellular medium, bar diagonally filled-in from left to right: [Ca2+]i measured in the presence of I mM CaCl 2 in the buffer, bar diagonally filled-in from right to left: [Ca2+]i measured in the presence of 2 mM EGTA

F'

101 kohn

F'

Figure 6. Effect of fMLP or forskolin on cellular cAMP in PMNs pretreated with IBMX and F. (PMNs were preincubated with IBMX (ImM) and with F (bar filled-in vertically or horizontally) or without F (bar filled-in or unfilled) for 5 min and 3 min, respectively, at 37 DC and activated with 100 nM fMLP for I min or with 25 11M forskolin for 5 min. Data represent mean values from two separate experiments with 4 determinations each

Exposure of polymorphonuclear leucocytes to sodium fluoride IBMX. In contrast to controls, where the transient decrease in cellular cAMP after tMLP stimulation is fully restored within the first min after stimulation, cAMP showed no increase in 6.2 mM F-pretreated cells and only a partial recovery at 2.7 mM F. In the presence oflBMX, tMLP-activation resulted in an increase in cAMP by about 40 % above resting levels within 60 sec after stimulation. This effect was not observed in F-pretreated cells. Analogous results were obtained with forskolin, which directly activates adenylyl cyclase. It increased

73

space. The decrease in the availability of extracellular Ca2+explains the reduction of both 0 2' generation and the release of lysosomal enzymes. This is in agreement with results of (27-29) who demonstrated, in the absence of extracellular Ca2+, only a minimal activation of PLD (which generates the bulk of DAG for protein kinase C activation by fMLP) and a loss of PKC membrane adherence, which is essential for NADPH oxidase activation. Measurements of intracellular cAMP have shown that low concentrations of F did not operate through an in-

cellular cAMP four-fold above the baseline level in con-

crease in intracellular cAMP in tMLP-activated PMNs.

trols but not in cells pretreated with P- (Figure. 6).

On the contrary, in the presence of a phosphodiesterase inhibitor, a decrease in cellular cAMP was observed after fMLP-stimulation and no rise after stimulation with forskolin. The adenylyl cyclase belongs to a group of Mg2+_ dependent enzymes which bind a hydroxide ion to the Mg2+ -ion during transition in the catalytically active form. Because of its high electronegativity F displaces the hydroxide ion and, by forming a strong hydrogen bond with hydrogen donors, stabilizes the close, catalyti-

Effect of NaF on degranulation

There was a minor leakage offl-glucuronidase in resting cells, leading to the formation of less than 0.5 Ilg (0.3 ± 0.15, n =4) nitrophenol/106 cells. Nitrophenol production led to a 16.8-fold increase above baseline after stimulation with tMLP. Exposure to 6.3 mM F or 2 mM EGTA for 3 min prior to tMLP activation reduced the formation of 4-nitrophenol from 8.4 ± 1.61lg to 4.0 ± O.4llg 1106 PMNs.

Discussion Our results show that short exposure of PMNs to F- at concentrations such as those retained in saliva after toothbrushing with 0.5 g of a 1.25 % F- containing gel and one

cally inactive conformation (30-31). The substitution reaction is analogous to that observed for hydroxyapatite, which can be converted to fluoroapatite by replacing the OH ion with F. Under clinical aspects a reduction in the bactericidal activity of PMNs by F- is only to be expected by local F application, since by fluoridation of public water corresponding concentrations are not reached. In saliva, the number of PMNs increases with the degree of intraoral inflammation to that in blood (5). Although the lifespan of PMN s in saliva is short, they contribute to intraoral in-

rinse with 10.0 ml water suppressed the fMLP-mediated increase in 02 and HOCl/OCI' generation and liberation of lysosomal enzymes without affecting resting cells. Activation of PMNs by fMLP involves mobilizing of Ca2+ from intracellular stores by generation of 1,4,5-inositol trisphosphate (JP3) which, together with simultaneously generated 1,2 diacylglycerol, activates protein kinase C and, thus, various downstream events, finally

fection defense mechanisms by additionally producing 0 2' radicals upon stimulation and by providing myeloperoxidase. The myeloperoxidase molecules contribute to up to 75 % of the peroxidase activity of mixed saliva. They catalyse the oxidation of thiocyanate ion and chloride ion by HP2 to produce oxidating agents with bacteriostatic activity (32-35). According to (32) the supply of

enabling vascular adhesion of polymorphonuclear leucocytes, chemotaxis and release of both reactive oxygen metabolites and lysosomal enzymes. Since the IP3-releasable Ca2+ pool accounted for only a small portion of the Ca2+ signal, an influx of Ca2+ from the extracellular space is an absolute prerequisite for complete coupling of the signal event to effector pathways. Fdose-dependently reduced the fMLP-induced influx of Ca2+ from the extracellular space into the cytoplasm by precipitating extracellular Ca2+ without altering the release of Ca2+ from internal stores or the entry of Mn2+from the extracellular

HP2 rather than the amount of peroxidase is the limiting factor for SCN' or Cl- oxidation in saliva. Due to the low substantivity of F in the oral cavity, the clearance of F is high at a normal salivation rate. Thus, the reduction of the HP2 supply from PMNs by F is clinically insignificant for healthy subjects but might be clinically relevant for xerostomic subjects. In these patients, the clearance of F from the oral cavity is slowed (36). Moreover, in patients who are xerostomic due to radiation of a tumor in the neck region, the salivary bicarbonate concentration is near zero and thus the buffering of the F containing gel is

74

E. Knoll-Kohler and G. Biissem

insufficient since fluoride gels made of acidulated phosphate fluoride, stannuous fluoride or amine fluoride have a low pH (3.5-4.5) in addition to a high F concentration (usually at least 1-2%). At a pH value below 6.5, not only the supply of HP2 for the peroxidase enzymes but their activity as well is dose-dependently suppressed by F and thus the formation of the major non-immune defense system HOSCN/OSCN- in saliva is impared (37, 38).

Acknowledgments We are grateful to Corina Menzel for her faithful technical assistance.

References 1. FEATHERSTONE, 1. D., GLENA, R., SHARIATI. M. AND SHIELDS, C. P. (1990) Dependence of in vitro demineralization of apatite and remineralization of dental enamel on fluoride concentration. 1.Dent. Res. 69, 620-625 2. NELSON, D. G., FEATHERSTONE, J. D., DUNVAN, 1. F. AND CUTESS, T. W. (1983) Effect of carbonate and fluoride on the dissolution behaviour of synthetic apatites. Caries Res, 17,200-211 3. CATE, J. M. AND FEATHERSTONE, 1. D. (1996) Physicochemical aspects of fluoride-enamel interactions. In: Fluoride in Dentistry Eds. Fejerskow, O. Ekstrand, 1. and Burl, B. M., 252-272 4. GEHRING F. (1983) Wirkung von Aminfluoriden und Natriumfluorid auf Keime der Plaqueflora. Dtsch zahn%orztl Z 38, 36-40 5. WOOLWEAVER, D.A., KOCH, G.G., CRAWFORD, J.J. AND LUNDBLAD, R.L. (1972) Relation of the orogranulocyte migratory rate to periodontal disease and blood leukocyte count: a clinical study. J Dent Res. 51, 929-939 6. SCHlOTT, c.R. AND LOE, H. (1970) The origin and variation in number ofleukocytes in the human saliva. J Periodont Res; 5, 36-41 7. THOMAS, E.L., JEFFERSON, M.M., JOYNER, R.E., COOK, G.S. AND KING, C.c. (1994) Leukocyte myeloperoxidase and salivary lactoperoxidase: identification and quantitation in human mixed saliva. J Dent Res. 73, 544-555 8. SCULLY, C. (1982) Phagocytic and killing activity of human blood, gingival crevicular and salivary polymorphonuclear leukocytes for oral streptococci. J Dent Res. 61 , 636-639. 9. ASHKENAZI, M. AND KOHL, S. (1990) Reduced antibodydependent cellular cytotoxicity to herpes simplex virus-infected cells of salivary polymorphonuclear leukocytes and inhibition of peripheral blood polymorphonuclear leukocyte cytotoxicity by saliva. J Immunol 77, 4781-4787

10. YAMAMOTO, M., SAEKI, K. AND UTSUMI, K. (1991) Isolation of human salivary polymorphonuclear leukocytes and their stimulation-coupled responses. Arch Biochem Biophys 289,76-82 11. AL-ESSA, L., NIWA, M., KOHNO, K-I., NOZAKI, M. AND TSURUMI, K. (1995) Heterogenity of circulating and exudated polymorphonuclear leukocytes in superoxide-generating response to cyclic AMP and cyclic AMP-elevating agents. Biochern Pharmacol 49, 315-322 12. ASHKENAZI, M. AND DENNISON, D. (1989)Anew method for isolation of salivary neutrophiles and determination of their functional activity. J Dent Res 68, 1256-1261 13. DECKMYN, H., TU, S. AND MAJERUS, P. (1986) Guanine nucleotides stimulate soluble phosphoinositide-specific phospholipase C in the absence of membranes. J Bioi Chern 261, 16553-16555 14. STRNAD, c., PARENTE, 1. AND WONG,K. (1986) Use of fluoride ion as a probe for the guanine nucleotride-binding protein involved in the phosphoinositide-dependent neutrophil transduction pathway. FEBS Lett 206, 20-24 15. BIANCA, V.D., GRZESKOWIAK, M., DUSI, S. AND ROSSI, F. (1988) Fluoride can activate the respiratory burst independentl y of Ca2+ stimulation of phosphoinositide turnover and protein kinase C translocation in primed human neutrophils. Biochern Biophys Res Commun 150,955-964 16. ENGLISH, D., RIZZO, M., TRICOT, G. AND HOFFMAN, R. (1989) Involvement of guanine nucleotides in superoxide release by fluoride-treated neutrophiles. Implications for a role of a guanine nucleotide regulatory protein. J Immunol143, 16851688 17. BROM, c., BROM, J. AND K' NIG, W G. (1991) Protein activation and mediator release from human neutrophils and platelets after stimulation with sodium fluoride and receptor-mediated stimuli. Immunol 73, 287-292 18. BOYUM, A (1968) Isolation of mononuclear cells and granulocytes from human blood. Scand. J Clin Lab Invest 97, (Suppl.), 77-79 19. GABLER, W.L., CREAMER, H.R. AND BULLOCK, WW (1986) Modulation of the kinetics of induced neutrophil superoxide generation by fluoride. Dent Res 65, 1159-1165 20. WEISS, S. 1., KLEIN, R., SLIVKA, A. AND WIE, M. (1982) Chlorination of taurine by human neutrophiles: Evidence for hypochlorous acid generation. J Clin Invest 70, 598-607 21. KETTLE, AI. AND WINTERBOURN, C.c. (1991) Mechanism of inhibition of myeloperoxidase by antiinflammatory drugs. Biochem Pharmacol41 , 1485-1492 22. RIDDLES, P.W, BLAKELEY, R.L. AND ZERNER, B. (1983) Reassessment of Ellman's reagent. Methods Enzymol. 91, 4960 23. GRYNKIEWICZ, G., POENIE, M. AND TSIEN, R. (1985) A new generation on Ca2+- indicators with greatly improved fluorescence properties. Bioi Chern 260,3440-3450 24. HARVATH, L., ROBBINS, 1.D., RUSSELL, AA AND SEA-

Exposure of polymorphonuclear leucocytes to sodium fluoride

75

MON, K.B. (1991) cAMP and human neutrophil chemotaxis. Elevation of cAMP differently affects chemotactic responsiveness. J Immunol146, 224-232

32. PRUITT, K.M. AND TENOVUO, 1. (1982) Kinetics of hypothiocyanite production during peroxidase-catalyzed oxidation of thiocyanate. Biochem. Biophys. Acta 704,204-214

25. YU, H., SUCHARD, S.1., NAIM, R. AND JOVE, R. (1995) Dissociation of mitogen-activated protein kinase activation from the oxidative burst in differentiated HL-60 cells and human neutrophils. J Bioi Chern 270,15719-24

33. CAO, c.F. AND SMITH, Q.T. (1988) Crevicular fluid myeloperoxidase at healthy, gingivitis and periodontitis sites. J Clin Periodont !6, 17-20

26. WOLF, M., CUATRECASAS, P. AND SAHYOUN, N. (1985) Interaction of protein kinase C with membranes is regulated by Ca2+, phorbol esters, and ATP. J Bioi Chern 260, 15718-15722 27. KESSELS, G., KRAUSE, K. AND VERHOEVEN, A. (1993) Protein kinase C activity is not involved in N-formylmethionyl1eucyl-phenylalanine-induced phospholipase D activation in human neutrophils, but is essential for concomitant NADPH oxidase activation: Studies with a staurosporine analogue with improved selectivity for protein kinase C. Biochem J 292, 781785

34. SMITH, A.1., SMITH, G. BASU, M.K. AND WALSH, T.F. (1984) Changes in salivary peroxidase activity observed during experimentally-induced gingivitis. J Clin Periodontolll, 373378. 35. VENEZIE, R.D., JENZANO, 1.w. AND LUNDBLAD, R.L. (1991) Differentiation of myeloperoxidase and glandular peroxidase in biological fluids: application to human saliva. J Clin Lab Analyt 57-59 36. BILLINGS, R., MEYEROWITZ, c., FEATHERSTONE, 1., ESPELAND, M., COOPER, L. AND PROSKIN, H. (1988) Retention of topical fluoride in the mouths of xerostomic subjects. Caries Res 22, 306-310

28. MULLMANN, T.1., CHEEWATRAHOOLPONG, B., ANTHES, 1., SIEGE, M., EGAN, R. AND BILLAH, M. (1993) Phospholipase C and phopholipase D are activated independently of each other in chemotactic peptide-stimulated human neutrophils. J Leukocyte Bioi 53, 630-635

37. ANDERSON, M.W., IZUTUSU, K.T. AND RICE, 1.C. (1981) Parotid gland pathophysiology after mixed gamma and neutron irradation of cancer patients. Oral Surg 52, 495-504

29. DANG, P., RAIS, S., HAKIM, 1. AND PERIANIN, A. (1995) Redistribution of protein kinase C isoforms in human neutrophils stimulated by formyl peptides and phorbol myristate acetate. Biochem Biophys Res Commun 212: 664-672

38. HANNUKSELA, S., TENOVUO, 1., ROGER, v., LENANDER-LUMIKARI, M. AND EKSTRAND, 1. (1994) Fluoride inhibits the antimicrobial peroxidase systems in human whole saliva. Caries Res 28, 429-43

30. LEBIODA, L., ZHANG, E., KRZYSZTOF, L. AND BREWER, 1.M. (1993) Fluoride inhibition of yeast enolase: crystal structure of the enolase-Mg2+-F-Pi complex at 2.6 A resolution. Proteins, Structure, Function and Genetics 16,219-225 31. MURPHY, A.1. AND COLL, R.1. (1993) Formation of a stable inactive complex of the sarcoplasmic reticulum calcium ATPase with magnesium, beryllium, and fluoride. J Bioi Chern 268,23307-23310