Published online before print June 3, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Goldman School of Dental Medicine and
School of Medicine, Boston University, Massachusetts
1Correspondence: Department of Periodontology and Oral Biology, Boston University, Goldman School of Dental Medicine, 100 East Newton St., Boston, MA 02118. E-mail: tvandyke{at}bu.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
[Ca2+]i) and cytoplasmic pH (pHi) changes, early stimulus responses to chemotactic agents, of LAP versus control (C)-PMN and explored whether these could be modulated by sensitizing cytokines or calcium channel-blocking agents. PMN responses of LAP patients were compared with age- and gender-matched controls. [Ca2+]i and pHi were measured fluorimetrically using 1H-indole-6-carboxylic acid, 2-[4-[bis[2-[(acetyloxy)methoxy]-2-oxoethyl]amino]-3-[2-[2-[bis[2-[(acetyloxy)methoxy]-2-oxoethyl]amino]-5-methylphenoxy]ethoxy]phenyl]-1 and 2',7'-bis-(carboxyethyl)-5(6)-carboxyfluorescein as respective probes. Not only was the maximal calcium response to chemoattractants higher in LAP-PMN, but also their subsequent intracellular calcium redistribution was significantly slower. The slower calcium redistribution of LAP-PMN, but not their higher maximal calcium response, was successfully mimicked in C-PMN treated with NifedipineTM or 1-[b-[3-(4-methoxyphenyl)propoxy]-4-methoxyphenethyl]-1H-imidazole-HCl, both known to be inhibitors of membrane-associated calcium influx, but this redistribution was not affected when inhibitors of other calcium influx mechanisms, DiltiazemTM or VerapamilTM, were used. Taken together, our findings indicate that certain early stimulus responses are aberrant in LAP-PMN, that internal redistribution of cytoplasmic-free calcium is compromised, and, additionally, that a membrane-associated Ca2+ transport defect may be present.
Key Words: human polymorphonuclear neutrophils • stimulus responses • spectrofluorimetry
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Severe systemic or unusual infections are often associated with inherited defects of neutrophil function. This may be true for LAP. Furthermore, diseases such as human immunodeficiency virus/AIDS, certain forms of neutropenia or leukemia, and genetically triggered syndromes such as Down’s, Papillon–Lefèvre, Cohen, leukocyte-adhesion deficiency, Chédiak–Higashi, histiocytosis, Ehlers–Danlos, and hypophosphatasia have been related to LAP and should be considered as systemic risk factors (as reviewed by Meyle and Gonzales [4 ]). These disorders are characterized by compromised host responses to infection, particularly in the function of leukocytes, thereby impairing the clearance of oral microbial plaque and aggravating the severity and progression of associated LAP. In addition, local polymorphonuclear leukocyte (PMN) counts at sites affected by periodontal inflammation have been shown to be reduced when, in particular, their innate chemotactic and bactericidal functions are compromised [5 ].
The regulatory mechanisms that control chemotaxis, adhesion, phagocytosis, and oxidative activity involve changes in the cytoplasmic pH (pHi) and cytoplasmic calcium concentration ([Ca2+]i) [6 7 8 9 ]. Several membrane channels, cytoplasmic calcium buffer- and sensor-molecules, as well as subcellular stores control the calcium homeostasis of PMN. When monitored kinetically, the cytoplasmic calcium response to stimulation typically is biphasic: A rapid release from the intracellular stores increases [Ca2+]i to a level tenfold greater than the basal 100 nM and is followed by an influx from the surrounding medium, which re-establishes calcium homeostasis and leaves the free [Ca2+]i at approximately 200 nM [10 ]. These stimulus changes in [Ca2+]i of PMN have been shown to be modulated in part by certain calcium channel inhibitors [11 , 12 ].
Thus, the receptor-controlled mechanisms that dictate PMN function in response to specific stimuli, e.g., upon binding of the chemotactic peptide formyl-Met-Leu-Phe (fMLP), fibroblast-, macrophage-, and PMN-derived interleukin 8 (IL-8), or other ligands to their receptors on the PMN membrane, begin rapidly following engagement of the receptor by its ligand [6
7
8
9
]. This signal can be propagated through two branches of the inositol phospholipid signaling cascade. In one branch, the receptor triggers the plasma membrane-bound phospholipase C isoforms (typically PLCß) via the activated Gq-
-subunit. PLCß then cleaves inositol-1,4,5-triphosphate (IP3), leaving diacylglycerol (DAG) on the membrane. Elevated IP3 concentrations then open IP3-gated Ca2+release channels in subcellular stores, mostly in the endoplasmic reticulum. In the other branch, the membrane-bound DAG together with phosphatidylserine and cytoplasmic Ca2+ can activate protein kinase C (PKC). This enzyme has various target proteins, among them various gene regulatory proteins as well as membrane channels themselves.
Various components of these PMN stimulus-response pathways are thought to be compromised in LAP-PMN, including a lower DAG kinase activity [13 ], a diminished PKC activity [14 ], and, recently, lower DAG kinase RNA levels and DAG kinase activity [15 ]. These findings are in concordance with the reports of elevated levels of the PKC activator DAG in LAP-PMN [13 , 16 , 17 ]. More recent reports mention that an additional fMLP-mediated IP3-independent signal propagation, downstream of the receptor–ligand binding signal of the formyl peptide receptor-like 1, through CD38 and activation of the ryanodine receptor at calcium stores, initiates a cytoplasmic calcium response [18 ].
In contrast to PMN from normal controls, Daniel et al. [19 ] reported defective chemotaxis in LAP-PMN and suggested a link between impaired chemotaxis and the altered cytosolic calcium response upon stimulation with fMLP. Additionally, they found the oxidative burst to be elevated in LAP-PMN. Another report implied that the activation of the putative calcium influx factor (CIF) was depressed, providing one possible reason for an altered cytoplasmic calcium response in LAP patients [20 ].
The aim of the present study was to determine whether LAP-PMN exhibit fundamental stimulus-response signaling abnormalities. If they do, their effector functions would probably be affected, and thus, the LAP-PMN could play a role in the disease’s severity and progression. We demonstrate here that the LAP-PMN do indeed exhibit intrinsic differences in their early signaling response to chemoattractants and are undertaking an investigation of their functional responses. Additionally, the impact of commercially available calcium channel inhibitors on the changes in [Ca2+]i ([Ca2+]i) and pHi (pHi) was tested to determine mechanisms of Ca2+ homeostasis in these cells and whether the abnormal PMN functions can be modulated.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
100 nM [21
].
Patient characteristics
All patients (LAP) were recruited through the Clinical Research Center at the Boston University School of Dental Medicine (MA); healthy controls (C) were similarly recruited through the dental clinics of the school. The Institutional Review Board of the Boston University School of Medicine approved the study, and all patients and controls gave their written, informed consent for this ex vivo study. Patient samples were obtained when these individuals first came to the clinic, had been suffering from the disease without treatment for at least 2 years, and exhibited the characteristic bone loss patterns of LAP. The clinical diagnosis for LAP was made using the criteria of the American Academy of Periodontology [22
], and except for the patients’ LAP, all study participants were healthy, as determined by a detailed medical history review. No concomitant medication or systemic condition known to affect the periodontal tissues was detected.
Neutrophil preparation
Whole blood was collected by venipuncture into 15 ml vacutainers containing 1.5 ml of a 4.3% citrate solution. PMN were purified by Dextran sedimentation followed by Ficoll-Hypaque centrifugation and brief hypotonic lysis of the remaining red blood cells, as described previously [21
]. Until use, PMN were kept rocking at 4°C in phosphate-buffered saline (PBS) with glucose (125 mM NaCl, 2 mM NaH2PO4, 8 mM Na2HPO4, 5 mM KCl, 5 mM glucose), pH 7.4. All experiments were performed within 3 h of blood drawing.
Stimulus preparation
fMLP was dissolved in dry DMSO, aliquotted into glass vacutainers, and frozen at –20°C until used. Possible effects of the vehicle DMSO on pHi or [Ca2+]i were monitored at their experimental concentrations and were found to be absent. Human recombinant IL-8 was used as an alternative stimulus. Here, the carrier was PBS containing highly purified bovine serum albumin (BSA; final concentration, <0.01%), as recommended by the supplier (Sigma-Aldrich). Potential stimulation of pHi or [Ca2+]i by the vehicle was also examined and found to elicit a small [Ca2+]i (<5%) equally in LAP- and in C-PMN.
Simultaneous [Ca2+]i and pHi measurements
PMN were loaded with the AM forms of the intracellular probes Indo-1 for [Ca2+]i and BCECF for intracellular pH, respectively, as described previously [8
, 9
]. Fluorescence (Indo-1: 
ex 350 nm, em1 405, and em2 485 nm; BCECF: ex1 500 nm, ex2 450 nm, and em 530 nm) was measured continuously in a HitachiTM F-4500 fluorimeter equipped with stirring and thermostated at 37°C. The Indo-1 and BCECF fluorescences were exported into spreadsheet format for further statistical evaluation by ANOVA. The known dissociation constant of Indo-1 and the Grynkiewicz equation allowed calculation of [Ca2+]i from the Indo-1 ratios [23
]. A calibration curve for BCECF fluorescence in PMN [6
, 7
] was used to calculate pHi. We found that NifedipineTM exhibits some absorbance at 450 nm (data not shown); we therefore prepared a NifedipineTM dose curve and corrected our data for NifedipineTM absorbance at the concentration used.
Stimulus responses
PMN (2x106 per ml) were suspended in Krebs-Ringer phosphate [KRP; PBS-glucose supplemented with Ca2+ (0.9 mM) and Mg2+ (1.5 mM)] at 37°C with stirring for 2 min [21
]. After this equilibration phase, the desired volume of fMLP or IL-8 stimulus was injected using a Hamilton syringe. The concentration of vehicle did not exceed 0.1% DMSO for fMLP or a final BSA concentration <0.01% for IL-8, respectively. When applicable, Ca2+ channel inhibitors were used and added to the cuvette prior to stimulation for the indicated length of time before the baseline was measured at t = 0.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
0.06), it became statistically significant when the dose of fMLP was larger, i.e., at 10–8 M, which proved to be the saturating dose, yielding the highest response in both LAP- and C-PMN (P<0.01). The mean [Ca2+]i in response to fMLP at the saturating concentration was 1280 nM for LAP-PMN, while cells of control subjects showed a 250-nM lower response (approximately 20%).
 View larger version (14K): [in a new window] |
Figure 1. Mean cytoplasmic calcium response to fMLP: differences in the [Ca2+]i between LAP and control subjects were statistically significant at the saturating concentration of 10–8 M fMLP; *, P < 0.01 (ANOVA); n  5; error bars represent SD.
|
[Ca2+]i of five LAP subjects versus five controls stimulated with 10–11 M IL-8 were higher (P=0.07; Fig. 2
). The difference in the response between LAP- and C-PMN was statistically significant at 10–10 and 10–8 M IL-8 (P<0.01). At 10–10 M, the saturating concentration of IL-8, the maximal mean [Ca2+]i was 880 nM for the LAP-PMN versus 630 nM (28% lower) for C-PMN.
 View larger version (13K): [in a new window] |
Figure 2. Mean cytoplasmic calcium response to IL-8: differences in the [Ca2+]i between LAP and control subjects were statistically significant at 10–10 M and 10–8 M IL-8; *, P < 0.01 (ANOVA); n 5; error bars represent SD.
|
 View larger version (26K): [in a new window] |
Figure 3. (A) Stimulation of PMN with 10–8 M fMLP at 120 s in the presence of [Ca2+]out. Left ordinate: Indo-1 ratio (proportional to [Ca2+]i), LAP ( ) versus control ( ); right ordinate: percentage of initial F530 (proportional to pHi), LAP ( ) versus control ( ). (B) Stimulation of PMN with 10–8 M fMLP at 135 s; [Ca2+]out was chelated with 5 µM EGTA 15 s before this stimulation (i.e., at 120 s; not shown). Left ordinate: Indo-1 ratio (proportional to [Ca2+]i), LAP () versus control (); right ordinate: percentage of initial F530 (proportional to pHi), LAP () versus control (). Figure is representative of three independent experiments.
|
As recently shown by our group [6 ], an electrically neutral Ca2+/H+ exchanger exists in PMN, a finding here reconfirmed by the results reported for controls and extended to LAP-PMN. Thus, as shown in Figure 3B , when [Ca2+]out was chelated with 5 mM EGTA 15 s prior to fMLP stimulation, the maximal [Ca2+]i attained was slightly lower in LAP- and C-PMN, and the cytoplasmic acidification was stronger in both, in accordance with the action of the Ca2+/H+exchanger. Thus, the faster and greater acidification in the LAP-PMN is followed by their slower alkalinization and [Ca2+]i decrease after the high [Ca2+]i transient, implying a less-effective Ca2+/H+ exchange in these cells.
Modulation of PMN functions by cytokines and substance-P (SP)
A representative curve for the stimulation of LAP- and C-PMN is shown in Figure 4A
. As indicated in Figure 4B
, C-PMN were preincubated with the modulators IFN-
, IL-8, TNF-, and SP at physiological concentrations for varying lengths of time (180, 600, 600, and 60 s, respectively), as described in the literature (for review, see Condliffe et al. [27
]). Stimulation with 10–8 M fMLP then followed. Preincubation had no effect on C-PMN [Ca2+]i and pHi responses over a range of preincubation times or concentrations of these four modulators (1 and 100 ng/ml IFN-; 10–12 and 10–9 M IL-8; 10 and 50 ng/ml TNF-; and 10–9 as well as 10–7 M SP; data not shown). Figure 5
shows that these modulators also did not affect the maximal [Ca2+]i attained in LAP-PMN but that some did alter the internal redistribution that followed.
 View larger version (25K): [in a new window] |
Figure 4. PMN calcium responses to 10–8 M fMLP. (A) LAP-PMN (; a); C-PMN (•; b). (B) Preincubation of control PMN with 10 ng/ml interferon- (IFN-) for 3 min (•; c); 10–11 M IL-8 for 10 min ( ; d); 25 ng/ml tumor necrosis factor (TNF-) for 10 min (x; e); 10–8 M SP for 1 min (; f). 
is representative of five independent experiments.
|
 View larger version (13K): [in a new window] |
Figure 5. Modulation of calcium responses of LAP versus control PMN by various agents; *, P < 0.05 (ANOVA); n = 5; error bars represent SD.
|
 View larger version (19K): [in a new window] |
Figure 6. PMN calcium responses to 10–8 M fMLP after 120 s incubation with the membrane Ca2+ channel inhibitor SKF-96365-HCl in KRP buffer. (A) LAP patients. (B) Control subjects: absence of the inhibitor (; a); preincubation beginning at t = 0 with SKF-96365-HCl at 10 µM (x; b), 50 µM ( ; c), and 100 µM (; d). Figures are representative of three separate experiments.
|
 View larger version (20K): [in a new window] |
Figure 7. PMN calcium responses to 10–8 M fMLP after 120 s incubation with NifedipineTM in KRP buffer. (A) LAP patients. (B) Control subjects: absence of the inhibitor (; a); preincubation beginning at t = 0 with NifedipineTM at 10 µM (; b), 50 µM (; c), and 100 µM (x; d). Figures are representative of five separate experiments.
|
 View larger version (17K): [in a new window] |
Figure 8. PMN calcium responses to 10–8 M fMLP after 120 s incubation with VerapamilTM in KRP buffer. (A) LAP patients. (B) Control subjects: absence of the inhibitor (; a); preincubation beginning at t = 0 with VerapamilTM at 10 µM (; b), 50 µM (; c), and 100 µM (x; d). Figures are representative of three separate experiments.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our finding that the signaling process of LAP-PMN in response to chemotactic agents is altered with respect to pHi and [Ca2+]i signals agrees with and expands on previous reports of problems with the Ca2+ responses and Ca2+-related signaling pathways in LAP-PMN [19 ]. Those reports, using a calcium-buffering probe (Quin-2), suggested impaired signal transduction, leading to decreased influx of extracellular calcium. We here extend these findings to report an associated dysfunction in the cytoplasmic pH portion of PMN signal transduction [6 ]. Aberrant LAP-PMN responses have been variously attributed to later steps in the signaling pathways, including changes in the DAG/PKC pathway [13 ], calcium channel function [19 ], and abnormal influx putatively related to CIF [20 , 30 ]. We found that these dysfunctions in LAP involve Ca2+ and H+ signaling as well as the subsequent redistribution of these cations within the LAP-PMN and with their environment.
We showed here that the early, rapid stimulus responses to fMLP or IL-8 by LAP-PMN, as measured by changes in cytoplasmic [Ca2+] and pH, are larger in LAP-PMN than in C-PMN but occur at the same rate and like C-PMN, involve intracellular stores. In contrast, the redistribution of the rapidly released Ca2+ is significantly slower in LAP-PMN, implying a defect in Ca2+ passage through membrane(s) or in intracellular Ca2+ binding. The similarly slower alkalinization in LAP-PMN could reflect aberrant H+ passage through the membrane, e.g., through our recently described Ca2+/H+ channel [6 ], or to a defect in the well-documented Na+/H+ antiport [25 , 26 ].
In an attempt to delineate whether a membrane channel or the Ca2+ redistribution process was perturbed by LAP, we attempted to mimic the problem in C-PMN. A significantly slower C-PMN Ca2+ redistribution after fMLP could be documented if these control cells were preincubated with NifedipineTM, a Ca2+ channel inhibitor that is known to affect membrane channels and subcellular Ca2+ store release [11 ], or with SKF-96365-HCl, at concentrations reported to block membrane channels for Ca2+ while not varying intracellular Ca2+ store release. These drugs did not increase or decrease the rate of Ca2+ redistribution in LAP-PMN [12 ]. The combination of these observations with our finding that pHi responses are also aberrant in LAP-PMN points to but does not prove that LAP-PMN have a defective Ca2+/H+ channel in their membranes. Definitive proof must await the availability of specific inhibitors of that channel or of a patient in whom this defect has been shown by some other means.
The possibility has been raised that a sensitizing effect by "nonphysiological means" can elicit a detectable calcium response in PMN [31 ]. With respect to CIF functions, the most recent reports cite sphingosine-1 phosphate as a potential CIF, activated when intracellular calcium stores are released in human PMN [32 ]. Our studies address the question of Ca2+ signaling per se, in LAP- versus C-PMN but did not seek nor claim to identify the mediators that may be directly responsible for sensitization of the LAP-PMN (i.e., the specific factors or channels). They therefore do not address the existence or role of putative CIF.
Taken together, these findings imply that membrane-related components contribute strongly to the reported signal transduction defect in LAP-PMN. However, the elevated stimulus response in LAP-PMN has its source in an intracellular calcium release, which in view of the high DAG concentrations [16 , 17 ], may mean abnormal PLC or PLD activity or putatively slower processing of DAG itself [13 , 15 ], whether as a result of the membrane defect or independently of it. Additionally, the patient PMN appeared to have defective redistribution, a finding reinforced by our ability to create a similar impairment in C-PMN, when treated with the Ca2+ membrane channel inhibitors used here. It should be noted that the findings reported here relate specifically to the response of LAP- versus C-PMN to chemotactic agents and may not apply to the responses of PMN mediated by receptors for other stimuli. That is, although Seetoo et al. [9] showed that the calcium signal was not necessary for the oxidative burst as well as the lytic functions which are the bactericidal responsibility of PMN, these cellular functions are not initiated through chemotactic but through Fc, complement, and cytokine receptors. In summary, the experiments reported here indicate that severe infections in LAP patients may be attributable to impaired chemotactic signaling, but it does not necessarily follow that there is a defect in their capability to process and kill bacteria. Studies looking at these functions are in progress, since if there is indeed a membrane or membrane component-processing defect in LAP-PMN, it is quite likely to affect at least some of the functions mediated via Fc, complement, and cytokine receptors of PMN.
Received May 4, 2005; accepted May 5, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
receptor activation of neutrophils and provides specificity between high-valency immune complexes and fMLP signaling pathways J. Leukoc. Biol. 61,522-528[Abstract]This article has been cited by other articles:
 |
|
 |
|
  J. M. Herrmann, J. Bernardo, H. J. Long, K. Seetoo, M. E. McMenamin, E. L. Batista Jr., T. E. Van Dyke, and E. R. Simons Sequential Chemotactic and Phagocytic Activation of Human Polymorphonuclear Neutrophils Infect. Immun., August 1, 2007; 75(8): 3989 - 3998. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
