Effect of Pentoxifylline In Vitro on Neutrophil Reactive Oxygen Production and Phagocytic Ability Assessed by Flow Cytometry**

Authors: Christoph Wenisch, Konstantin Zedtwitz-Liebenstein, Bernhard Parschalk, Wolfgang Graninger, Department of Infectious Diseases, Internal Medicine I, University Hospital of Vienna, Vienna, Austria

**Please Note: This article has been altered from the original…no figures are included and some text and images have been omitted. The original paper can be seen and downloaded from Medscape………….. Chick Newman

Summary: Neutrophil granulocytes have been described as agents of defense and destruction. The effect of pentoxifylline on the phagocytic ability and generation of reactive oxygen radicals of neutrophils was studied at concentrations of 1, 10 and 100 mg/L. Flow cytometry was used to study phagocytic ability by measuring uptake of fluorescein-labelled bacteria. The generation of reactive oxygen intermediates was estimated by the quantification of the intracellular conversion of dihydrorhodamine 123 to rhodamine 123. In vitro pentoxifylline treatment diminished neutrophil reactive oxygen production (at 10 mg/L -45% and at 100 mg/L -63%; p < 0.001 for both) and reduced neutrophil phagocytic ability (at 100 mg/L -23%; p < 0.05). Both effects were rapidly reversible after plasma exchange. We conclude that pentoxifylline could decrease oxidative tissue damage by neutrophils in septicaemia or after IV granulocyte transfusion. [Clin. Drug Invest. 13(2):99-104, 1997. © 1997 Adis International Limited]



Contents


Introduction

Damage to the vascular system during septicaemia and its sequelae are caused, in part, by the recruitment and adhesion of neutrophils to the endothelium and their release of destructive molecules.[1]

Several agents with anti-inflammatory activity, including nonsteroidal anti-inflammatory drugs,[2] methotrexate[3] and corticosteroids,[4] were shown to inhibit endothelial adhesion of neutrophils by different mechanisms.[5] Pentoxifylline was shown to inhibit the inflammatory action of interleukin-1 and tumour necrosis factor alpha (TNFa),[6] diminish endothelial inflammation, and clinically reduce mortality in patients with septic syndromes.[7]

In this study we analysed the effect of both permanent and transient pentoxifylline treatment on neutrophil phagocytic ability and reactive oxygen production ex vivo using flow cytometry.[8-13]

Materials and Methods

We performed a flow cytometry assessment of neutrophil function using heparinised whole blood from 8 healthy volunteers (laboratory personnel, 4 female, 4 male, aged 21 to 34 years) as described below.[11-13] The samples were incubated with pentoxifylline (Hoechst, Frankfurt, Germany) at final concentrations of 1, 10 and 100 mg/L at ambient temperature for 0, 30, 120, 150, 210 and 270 minutes. There is no difference between pentoxifyllineincubation at ambient temperature versus incubation at 37°C. To examine whether potential effects are reversible, pentoxifylline-containing plasma was removed and pentoxifylline-free plasma was substituted after 1 hour's incubation at room temperature.

Phagocytic capacity was assessed by adding 10µl of fluorescein isothiocyanate (FITC)-labelled Escherichia coli ATCC 25922 (10 8 /ml) to 100µl of heparinised whole blood and incubating for 10 minutes at 37°C. Thereafter, 100µl of icecold quenching solution (Orpegen, Heidelberg, Germany) was added and the samples were washed twice in phosphate-buffered-saline (pH 7.4). Finally, 2ml fluorescence-activated cell sorter (FACS)-lysing solution (Becton Dickinson, California, USA) was added. After 20 minutes, the samples were washed again and resuspended in 100ul of phosphate-buffered saline containing propidium iodide at a concentration of 50 mg/L for DNA staining and kept on ice until analysis.

For analysis of reactive oxygen (ROI) production blood samples (100ul) were stimulated with 25µl of E. coli (10 8 /ml) [ATCC 25922, not labelled] at 37°C. After 10 minutes, 25µl of the dihydrorhodamine 123 (DHR) solution was added. After another 10 minutes at 37°C, 2ml of FACS-lysing solution was added and incubated for 20 minutes at room temperature. Thereafter, the samples were washed with phosphate-buffered saline and resuspended with 100ul of phosphate-buffered saline containing propidium iodide at a final concentration of 50 mg/L for DNA staining.

The cells were analysed on a standard FACScan flow cytometer (Becton Dickinson, California, USA). For each measurement, 10 000 events were collected. Granulocytes were separated by setting a gate on the population. To set the gate, the forward scatter (size) and side scatter (granularity) of the cells were determined and recorded. The granulocyte population of the whole blood sample was then identified as medium-sized cells with high granularity and separated from monocytes and lymphocytes. To exclude cell debris and non-phagocytised bacteria, a gate was set on leucocytes during acquisition in FL2 (red fluorescence).

For analysis of the ROI production, the shift to the right in FL1 (green fluorescence) was determined. The amount of cleaved substrate was estimated by the mean fluorescence using the statistical option of the FACScan software. Similarly, the amount of phagocytised bacteria was assessed by a shift in mean fluorescence to the right (FL1). The mean fluorescence of both assays was compared with unstimulated controls. Daily alignment and calibration of the instrument was done using fluorescence beads (Calibrite, Becton Dickinson, California, USA). The beads were put into the same histogram channel every day.

Statistical Analysis

For statistical analysis nonparametric tests were used. Values were compared with the Kruskall-Wallis test and Wilcoxon rank sum test. Spearman correlation was used. Two-sided p values < 0.05 were considered to be significant.

Results

The effect of pentoxifylline (0 to 100 mg/L) on neutrophil phagocytic ability and reactive oxygen production is depicted (in figures 1 and 2---not included on this web site) and table 1. Pentoxifylline (10 to 100 mg/L) diminished neutrophil phagocytic ability in a dose- and time-dependent manner. After 270 minutes of pentoxifylline exposure at concentrations of 10 and 100 mg/L, respectively, the phagocytic ability was <= 50% of pre-exposure levels. The reduction of pentoxifylline-driven impairment of neutrophil phagocytosis was significant at 1 mg/L after 270 minutes, at 10 mg/Lafter 150 minutes and at 100 mg/L after 120 minutes. Plasma exchange after 60 minutes' pentoxifylline incubation prevented the impairment of neutrophil phagocytosis .

   

Neutrophil reactive oxygen production was significantly decreased at concentrations of 10 and 100 mg/L after 30 minutes' incubation time. Lower concentrations (1 mg/L) did not significantly affect neutrophil reactive oxygen production. Plasma exchange (pentoxifylline-laden plasma versus pentoxifylline-free plasma) resulted in a rapid restoration of normal reactive oxygen production.

Discussion

Once septic shock develops, the role of neutrophils as mediators of inflammation may predominate over their role as the principal line of cellular defence against bacterial infection.[8] Activated neutrophils adhere to each other and to the endothelium. This results in capillary leakage, release of toxic oxygen species and lytic proteins, and recruitment of other inflammatory cells to the site of inflammation. Neutrophil-derived oxidants, proteinases, and antiproteinases interact in sepsis to maximise their ability to damage host tissues.[8]

In our study, pentoxifylline reduced oxidative product formation in neutrophils in a dose-dependent manner. This could be due to a reduction of cellular TNF expression in vitro via the inhibition of phosphodiesterase and a consecutive increase of intracellular cyclic adenosine monophosphate, and correlates with previous findings.[14] Reduced TNF levels after E. coli stimulation could result in decreased neutrophil phagocytic ability and decreased reactive oxygen production.[7] However, pentoxifylline could also directly inhibit the action of E. coli on neutrophil oxidative activity before TNF production is affected at the early time-points.

Oral pentoxifylline treatment in humans (400mg) resulted in a reduced superoxide anion production similar to our findings even at concentrations as low as < 0.6 mg/L.[15] However, these authors described a strong correlation between 3 metabolites of pentoxifylline (not the parent compound), and the suppression of reactive oxygen production. These findings were corroborated by in vitro investigations showing 2 of the 4 methylxanthines being most effective in reducing the polymorphonuclear neutrophil leucocyte respiratory burst.[15] Furthermore, the inhibitory activity was seen at concentrations achievable in vivo.[15] The principal active metabolites were shown to be 10 times more active than the parent compound.[15,16] This corresponds with the higher concentrations of the parent compound used in the in vitro assay.

Flow cytometry provides a powerful tool for analysing effects of immunomodulators on bacterium- host cell interactions.[13] In this study whole blood was used, avoiding neutrophil isolation and purification steps, which could potentially affect neutrophil functional behaviour. Results obtained with assays using isolated neutrophils cannot be extended to the reactivity of polymorphonuclear neutrophil leucocytes of septic patients.[13] Although anticoagulation with heparin was required for the experiments, it is unlikely that addition of heparin preferentially influences the phagocytic or oxidative radical-producing capacity of pentoxifylline- treated neutrophils as opposed to normal donor leucocytes.

In septic shock hyperreagible neutrophils are implicated in widespread and uncontrolled tissue destruction.[8] However, in sepsis a decreased neutrophil function has been described ex vivo and clinical improvement correlated with a normalisation of neutrophil function.[11] This could be due to auto-oxidation of neutrophils (i.e. overproduction of intercellular reactive oxygen and functional impairment). This mechanism for altered function of these cells may also lead to a diminished neutrophil reactive oxygen production and phagocytic ability in vivo.[9]

In vivo the antagonism of intracellular oxygen radical production in hyperreagible cells with pentoxifylline could abrogate the pro-inflammatory effects of cytokines on neutrophil toxic radical production. In `hyper-cytokine states' like septic shock with extensive neutrophil-mediated tissue damage, pentoxifylline might be useful.[12]



Table

Table 1 - Effect of pentoxifylline on polymorphonuclear neutrophil leucocyte function [n = 8, median (range) fluorescence channel]

Time
(min)
Phagocytosis
(pentoxifylline mg/L)
Reactive oxygen production (pentoxifylline mg/L)
0
1
10
100
0
1
10
100
0 1208
(1002-1312)
1198
(980-1348)
1206
(1102-1298)
1240
(1098-1331)
98
(82-115)
95
(86-106)
106
(81-119)
102
(86-112)
30 1256
(881-1380)
1125
(1002-1456)
1150
(878-1221)
1059
(822-1111)
105
(81-110)
103
(75-109)
58
(35-74)
46
(29-71)
60 1201
(952-1945)
1123
(641-1345)
1056
(527-1121)
987
(522-1012)
116
(96-128)
108
(99-131)
45
(33-52)
34
(27-43)
120 1298
(853-1380)
1105
(859-1231)
928
(789-1102)
855
(775-1099)
105
(89-121)
100
(87-110)
50
(34-66)
34
(20-49)
150 1158
(902-1258)
1001
(758-1056)
643
(459-845)
580
(456-689)
110
(86-128)
117
(89-131)
55
(36-65)
39
(27-51)
210 1222
(958-1308)
987
(724-1022)
632
(472-1192)
528
(479-817)
100
(76-112)
95
(66-121)
56
(21-61)
32
(21-43)
270 1142
(769-1453)
939
(743-1370)
598
(443-701)
480
(380-556)
91
(71-113)
82
(46-100)
53
(44-58)
33
(28-39)
Plasma exchange at 60 minutes
0 1253
(980-1304)
1280
(995-1456)
1198
(945-1501)
1243
(976-1369)
99
(81-121)
112
(87-125)
104
(85-124)
110
(92-119)
30 1248
(986-1331)
1135
(956-1458)
1140
(878-1401)
1004
(802-1202)
102
(71-110)
101
(75-109)
61
(35-74)
43
(29-49)
60 1240
(956-1398)
1112
(879-1256)
1043
(888-1152)
978
(814-1179)
106
(83-116)
95
(76-105)
46
(31-72)
32
(27-43)
90 1131
(802-1357)
1297
(929-1448)
1283
(970-1406)
1250
(865-1318)
114
(99-128)
108
(88-129)
121
(90-141)
89
(58-102)
120 1141
(802-1354)
1138
(769-1471)
1077
(874-1343)
1256
(860-1308)
104
(92-131)
109
(100-132)
111
(85-131)
94
(69-105)
150 1100
(892-1250)
1034
(764-1316)
1091
(708-1210)
1178
(805-1278)
117
(84-132)
92
(79-113)
123
(87-138)
102
(82-113)
210 1005
(982-1098)
1254
(969-1369)
1073
(965-1244)
1109
(869-1289)
107
(86-122)
99
(93-128)
100
(91-115)
90
(72-102)
270 1058
(930-1300)
1062
(848-1351)
1088
(835-1268)
1049
(789-1214)
116
(95-121)
102
(74-110)
101
(81-114)
90
(77-112)

Correspondence and reprints: Dr Christoph Wenisch, Department of Infectious Diseases, Internal Medicine I, University Hospital of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria.


 

**Please Note: This article has been altered from the original…no figures are included and some text (referring to figures and Table of Contents) and images have been omitted. The original paper can be seen and downloaded from Medscape………….. Chick Newman