Life Sciences, Vol. 47, pp. 1023-1029 Printed in the U.S.A.
Pergamon Press
PENTOXIFYLLINE INHIBITS LIPOPOLYSACCHARIDE-INDUCED SERUM TUMOR NECROSIS FACTOR AND MORTALITY Phillip Noel, Steve Nelson, Ronald Bokulic, Gregory Bagby, Howard Lippton, Gary Lipscomb, Warren Summer Departments of Pulmonary/Critical Care Medicine, Physiology, and Pathology. Louisiana State University Medical Center New Orleans, Louisiana (Received in final form July 17, 1990) Summary Tumor necrosis factor, a mononuclear phagocyte-derived peptide produced in response to lipopolysaccharide, has been shown to mediate certain aspects of septic shock and multiple organ failure resulting from gram-negative septicemia. In the present investigation, pretreatment of animals with pentoxifylline inhibited lipopolysaccharide-induced serum tumor necrosis factor in a dose-dependent fashion. Pentoxifylline prevented the sequestration of neutrophils seen in animals given intravenous lipopolysaccharide. Furthermore, pentoxifylline protected animals from the lethal effects of an intravenous challenge with lipopolysaccharide. These data indicate that pentoxifylline inhibits lipopolysaccharide-induced tumor necrosis factor and may be an effective agent in mitigating the lethal consequences of sepsis and other disease processes mediated by this cytokine. Tumor necrosis factor-alpha (TNF), a cytokine elaborated by lipopolysaccharide (LPS)-stimulated mononuclear phagocytes, is a potent mediator in a variety of physiologic and immunologic processes (1). Circulating TNF has been detected in the serum of normal healthy human subjects in response to intravenous LPS (2, 3). The strong association between TNF production and LPS stimulation suggests that TNF is a central mediator of the inflammatory cascade activated by sepsis. TNF is thought to be a proximal mediator of the inflammatory response and most likely triggers the release of other secondary mediators including other cytokines and arachidonic acid metabolites. TNF has been implicated as an important factor in a wide variety of acute and chronic inflammatory disease states such as the adult respiratory distress syndrome and rheumatoid arthritis (I). Studies indicate that the disseminated and/or excessive stimulation of TNF biosynthesis during septicemia is a critical step in triggering endotoxin-induced shock and multiple organ injury (4, 5). Furthermore, it has recently been shown that baboons pretreated with monoclonal antiTNF antibodies were protected from a lethal challenge with live Escherichia coli (6).
0024-3205/90 $3.00 + .oo Copyright (c) 1990 Pergamon Press plc
1024
Pentoxifylline and Tumor Necrosis Factor
Vol. 47, No.
12, 1990
Pentoxifylline, a methylxanthine, has been shown to increase animal survival in lethal models of infection although the mechanism underlying this effect remains unknown (7, 8). Pentoxifylline has been demonstrated in vitro to suppress LPS-induced macrophage-derived TNF, and to block the stimulatory action of TNF on neutrophils(9, 10). The purpose of the present investigation was to determine in vivo i f the protective effect of pentoxifylline during lethal sepsis was associated with a direct suppression of TNF released in response to a systemic challenge with E. coli LPS. Methods Male Sprague-Dawley rats (200-225 gm) were l i g h t l y anesthetized with ether and two catheters were surgically placed under aseptic conditions by previously described methods, one into the internal jugular vein and the other into the carotid artery (11). The rats were allowed to recover for 24 hours post-catheterization and had access to food and water. The next day 5.0 mg/kg of E. coli LPS (026:B6, Difco Laboratories) was administered intravenously. Serum blood samples (0.3 ml) were collected to determine serum TNF levels at 15, 30, 60, 120, 180, and 240 min post-LPS administration. After the time course of LPS-induced serum TNF in normal animals was determined, separate groups of rats were pretreated with either saline or pentoxifylline (Hoechst-Roussel Pharmaceuticals) 50 mg/kg or 100 mg/kg intravenously 60 min prior to intravenous E. coli LPS challenge (5 mg/kg). Serum samples were taken 90 min after LPS administration for assaying TNF a c t i v i t y which corresponded to the time of peak serum TNF a c t i v i t y . Blood samples were also obtained prior to and 240 min post-LPS for total and d i f f e r e n t i a l peripheral blood counts. Selected saline and pentoxifylline treated rats were observed for 24 hours post-LPS administration to determine survival. TNF-alpha a c t i v i t y in serum specimens was determined with a cytot o x i c i t y assay u t i l i z i n g L929 cells obtained from~American Type Culture Collection, Rockville, Maryland. Briefly, 2 x 10~ L929 cells were planted in the wells of microtiter plates and allowed to adhere overnight. The following day wells were treated with serial dilutions of the specimens followed by addition of ~ctinomycin D (1.0 ug/ml, final conc.). After an 18 hour incubation at 37 in 5% CO~, the wells were washed and surviving cells indexed with a chromogenic a~say for hexosaminidase. One unit of TNF a c t i v i t y was defined as the reciprocal of the dilution required to k i l l 50% of the cells which was normalized against murine recombinant TNF (Genentech, Inc.). The 50% endpoint was calculated by a four-parameter logistic analysis. To verify that the cell k i l l i n g was due specifically to TNF-alpha, cell l y t i c a c t i v i t y was completely neutralized by the addition of rabbit antibody against murine recombinant TNF-alpha (Genentech, Inc.). Rats pretreated with saline or 100 mg/kg pentoxifylline were sacrificed 4 hours after LPS challenge for histological determinations. The animals were exsanguinated by cardiac puncture and the lungs surgically removed en bloc. The lungs were gently inflated with 10% formalin in phosphatebuffered saline solution and processed by routine histological methods for hematoxylin-eosin stained sections to be examined by l i g h t microscopy. Data were analyzed using analysis of variance and a Student's two tailed t - t e s t where indicated. All data were reported as mean ± standard error.
Vol. 47, No.
12, 1990
Pentoxifylline
and Tumor Necrosis Factor
1025
Results TNF was not detected in the serum of unchallenged rats. Intravenous LPS challenge, however, resulted in the appearance of significant TNF activity in the serum. Serum TNF peaked 90 min after intravenous LPS with levels in the 10,000 U/ml range (Figure I ) . By 3 hours after intravenous LPS, serum TNF had decreased to levels near the non-detectable level of the assay (40 U/ml). 12 10
(/) 2
1
2
3
4
Hours FIG. 1 Serum TNF activity in rats following LPS administration. Histologic sections of the lungs following intravenous LPS showed a marked sequestration of polymorphonuclear leukocytes (PMN) within the vasculature of the lung (Figure 2). This sequestration of PMN within the lung vasculature was associated with a significant decrease in the3number of PMN circulating within the peripheral b,ood from 1.7 + 0.2 x 10 PMN/mm 3 in control animals to 0.45 ± 0.02 x 103 PMN/mm3 in animaTs receiving intravenous LPS. Pretreatment with pentoxifylline prevented this LPSinduced decrease in circulatory PMN (Table I).
FIG. 2 Photomicrograph of a hematoxylin-eosin stained lung section following LPS challenge.
1026
Pentoxifylline and Tumor Necrosis Factor
Vol. 47, No. 12, 1990
TABLE I Number of Circulating PMN in Rats Injected with Saline, LPS alone, or Pentoxifylline Prior to LPS Groups
PMN Cell Count (PMN/mm3)
Saline
1.7±0.2 x 103
LPS
0.45±0.02 x 103*
LPS + Pentoxifylline 100 mg/kg
1.5±0.3 x 103
Each data point equals the mean ± SEM of 8-10 determinations; *p!O.01 compared to saline control Pretreatment with pentoxifylline 60 min prior to LPS administration inhibited serum TNF in a dose-dependent fashion (Figure 3). The highest dose of pentoxifylline reduced serum TNF to less than 5 percent of the levels seen in the saline treated animals (p~O.01 compared to saline control).
PTX 100 mg/k 9
PTX 50
mg~g
Saline o
i
,
4
6
8
,
,
lO
12
Serum TNF (xlO00
units)
FIG. 3 Serum TNF activity in rats pretreated with either saline or pentoxifylline (PTX) prior to LPS administration. The lungs of rats pretreated with 100 mg/kg pentoxifylline prior to intravenous LPS did not develop histologic evidence of PMN sequestration within the pulmonary vasculature (Figure 4).
Vol. 47, No. 12, 1990
Pentoxifylline
and Tumor Necrosis Factor
1027
FIG. 4 Histologic section of the lung of a rat pretreated with pentoxifylline prior to LPS challenge. Pentoxifylline caused a significant increase in the 24 hour survival of animals challenged with an intravenous dose of LPS. Rats in both treatment groups were completely protected against the lethal consequencesof systemic LPS administration. TABLE I I Survival in LPS-Injected Rats Pretreated with Saline or Pentoxifylline Groups
24 Hour Survival
Saline
4/10
Pentoxifylline 50 mg/kg
10/10"
Pentoxifylline 100 mg/kg
I0/I0"
Each data point equals the mean ± SEM of I0-12 determinations; *p~O.01 compared to saline control Discussion Tumor necrosis factor is emerging as an important immunoregulatory mediator of septic shock and multiple organ failure. Its production by mononuclear phagocytes is stimulated by LPS, gram-postive and gramnegative bacteria, and other cytokines. Administration of recombinant TNF to animals results in pathophysiologic abnormalities similar to those of septic shock, including tissue injury involving the lung, kidney, and bowel (4). Furthermore, studies have demonstrated that passive immunization with antibodies directed against TNF protects animals from the lethal consequences of intravenous LPS or challenge with live E. coli (6).
1028
Pentoxifylline and Tumor Necrosis Factor
Vol. 47, No. 12, 1990
Pentoxifylline has been reported to significantly improve survival in animal models of lethal infection (7). Several reports in humans have correlated mortality from severe infections with elevated systemic TNF levels (2). We, therefore, hypothesized that one possible mechanism of pentoxifylline's action may be an inhibition of TNF release in response to a systemic infectious stimulus. These in vivo data demonstrate that pentoxifylline suppressed LPS-induced TNF in a dose-dependent fashion and significantly improved animal survival. Furthermore, pentoxifylline prevented LPS-induced sequestration of PMN within the pulmonary vasculature and systemic neutropenia. Intravenous lipopolysaccharide is known to cause a dose-dependent sequestration of neutrophils within the pulmonary vascular bed (12, 13). Pulmonary PMN sequestration and increased pulmonary permeability have also been reported in guinea pigs injected intravenously with human recombinant TNF (14). TNF is also known to stimulate PMN adhesion to vascular endothelial surfaces in vitro (15). In the present investigation, pretreatment with pentoxifylline prevented LPS-induced neutrophil aggregation within the lung vasculature. Neutrophils l i k e l y play an important role in TNF-induced lung injury. TNF is known to be a potent activator of PMN functions including chemotaxis, phagocytosis, and superoxide anion production (16). Pentoxifylline has been shown to i n h i b i t the stimulatory action of TNF on PMN functions in vitro and to protect animals from LPS and TNF-induced non-cardiogenic pulmonary edema (14, 1 5 ) . Recently, pentoxifylline has been demonstrated to suppress LPS-induced mononuclear phagocyte-derived TNF in vitro at the level of both TNF mRNAaccumulation and TNF bioactivity (9). The presumed mechanism of action is secondary to the phosphodiesterase inhibitory effects of pentoxifylline resulting in an increase in intracellular cAMP. Pretreatment of mononuclear phagocytes with dibutyryl cAMP has been shown to have similar suppressive effects on TNF expression (9). The immunologic effects of pentoxifylline may, therefore, have important therapeutic implications in clinical disease states characterized by excessive or persistent TNF activity. Pentoxifylline appears to be efficacious in both directly suppressing LPS-induced TNF release and in modifying the actions of this cytokine on phagocytic cells. Therefore, i t may prove useful not only as a prophylactic agent in preventing the development of septic shock in high risk patients, but also as a therapeutic agent in patients where the inflammatory-cytokine pathway has already been ignited.
Acknowledgements This work was supported in part by Grant #Z0739 from the Cystic Fibrosis Foundation. We wish to sincerely thank Amy Weinberg, Rhonda Martinez, and Amad Jawda for their expert technical support, as well as Janel Furlong and Mamie Kinder for secretarial assistance in preparing the manuscript. The authors also wish to thank Dr. William J. Novick, Jr., Hoechst-Roussel Pharmaceuticals, for his support in this project.
Vol. 47, No. 12, 1990
Pentoxifylline
and Tumor Necrosis Factor
1029
REFERENCES i. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
B. BEUTLER and A. CERAMI, N. Engl. J. Med. 316 379-385 (1987). H.R. MICHIE, K.R. MANOGUE, D.R. SPRIGGS, A. REVHAUG, S. O'DWYER, C.A. DINARELLO, A. CEP~AMI, S.M. WOLFF and D.W. WILMORE, N. Engl. J. Med. 318 1481-1486 (1988). E. ZIEGLER, N. Engl. J. Med. 318 1533-1534 (1988). K.J. TRACEY, B. BEUTLER, S.F. LOWRY, J. MERRYWEATHER, S. WOLPE, I.W. MILSARK, R.J. HARIRI, T.J. FAHEY I l l , A. ZENTELLA, J.D. ALBERT, G.T. SHIRES and A. CERAMI, Science. 234 470-474 (1986). D.N. MANNEL, H. NORTHOFF, F. BAUSS and W. FALK, Rev. Inf. Dis. 602-606 (1987). K.J. TRACEY, Y. FONG, D.G. HESSE, K.R. MANOGUE, A.T. LEE, G.C. KUO, S.F. LOWRYand A. CERAMI, Nature. 330 662-664 (1987). G.L. MANDELL, Am. Rev. Respir. Dis. 138 1103-1105 (1988). G.E. CHALKIADAKIS, A. KOSTAKIS, P.E. KARAYANNACOS, M.E. CHALKIADAKIS, S. SGOUROMALI, H. GIOAMARELLOU, and G.D. SKALKEAS, Arch. Surg. 120 1141-1144 (1985). R.Mo STRIETER, D.G. REMICK, P.A. WARD, R.N. SPENGLER, J.P. LYNCH I l l , J. LARRICK and S.L. KUNKEL, Biochem. Biophys. Res. Com. 155 1230-1236 (1988). G.W. SULLIVAN, H.T. CARPER, W.J. NOVICK JR. and G.L. MANDELL, Inf. Immune. 56 1722-1729 (1988). V. POPOVIC and P. POPOVIC, J. Appl. Physiol. I__5727-728 (1960). G.S. WORTHEN, C. HASLETT, A.J. REES, R.S. GUMBAY, J.E. HENSON, P.M. HENSON, Am. Rev. Respir. Dis. 136 19-28 (1987). C. HASLETT, G.S. WORTHEN, P.C. GICLAS, D.C. MORRISON, J.E. HENSON, P.M. HENSON, Am. Rev. Respir. Dis. 136 9-18 (1987). C.M. LILLY, J.S. SANDHU, A. ISHIZAKA, H. HARADA, M. YONEMARU,J.W. LARRICK, T. SHI, P.T. O'HANLEY, and T.A. RAFFIN, Am. Rev. Respir. Dis. 139 1361-1368 (1989). C.H. WELSH, D. LIEN, G.S. WORTHENand J.V. WEIL, Am. Rev. Respir. Dis. 138 1106-1114 (1988). H. BESSLER, R. GILGAL, M. DJALDETTI and I. ZAHAVl, J. Leuk. Biol. 4._00747-754 (1986).
Pergamon Press
PENTOXIFYLLINE INHIBITS LIPOPOLYSACCHARIDE-INDUCED SERUM TUMOR NECROSIS FACTOR AND MORTALITY Phillip Noel, Steve Nelson, Ronald Bokulic, Gregory Bagby, Howard Lippton, Gary Lipscomb, Warren Summer Departments of Pulmonary/Critical Care Medicine, Physiology, and Pathology. Louisiana State University Medical Center New Orleans, Louisiana (Received in final form July 17, 1990) Summary Tumor necrosis factor, a mononuclear phagocyte-derived peptide produced in response to lipopolysaccharide, has been shown to mediate certain aspects of septic shock and multiple organ failure resulting from gram-negative septicemia. In the present investigation, pretreatment of animals with pentoxifylline inhibited lipopolysaccharide-induced serum tumor necrosis factor in a dose-dependent fashion. Pentoxifylline prevented the sequestration of neutrophils seen in animals given intravenous lipopolysaccharide. Furthermore, pentoxifylline protected animals from the lethal effects of an intravenous challenge with lipopolysaccharide. These data indicate that pentoxifylline inhibits lipopolysaccharide-induced tumor necrosis factor and may be an effective agent in mitigating the lethal consequences of sepsis and other disease processes mediated by this cytokine. Tumor necrosis factor-alpha (TNF), a cytokine elaborated by lipopolysaccharide (LPS)-stimulated mononuclear phagocytes, is a potent mediator in a variety of physiologic and immunologic processes (1). Circulating TNF has been detected in the serum of normal healthy human subjects in response to intravenous LPS (2, 3). The strong association between TNF production and LPS stimulation suggests that TNF is a central mediator of the inflammatory cascade activated by sepsis. TNF is thought to be a proximal mediator of the inflammatory response and most likely triggers the release of other secondary mediators including other cytokines and arachidonic acid metabolites. TNF has been implicated as an important factor in a wide variety of acute and chronic inflammatory disease states such as the adult respiratory distress syndrome and rheumatoid arthritis (I). Studies indicate that the disseminated and/or excessive stimulation of TNF biosynthesis during septicemia is a critical step in triggering endotoxin-induced shock and multiple organ injury (4, 5). Furthermore, it has recently been shown that baboons pretreated with monoclonal antiTNF antibodies were protected from a lethal challenge with live Escherichia coli (6).
0024-3205/90 $3.00 + .oo Copyright (c) 1990 Pergamon Press plc
1024
Pentoxifylline and Tumor Necrosis Factor
Vol. 47, No.
12, 1990
Pentoxifylline, a methylxanthine, has been shown to increase animal survival in lethal models of infection although the mechanism underlying this effect remains unknown (7, 8). Pentoxifylline has been demonstrated in vitro to suppress LPS-induced macrophage-derived TNF, and to block the stimulatory action of TNF on neutrophils(9, 10). The purpose of the present investigation was to determine in vivo i f the protective effect of pentoxifylline during lethal sepsis was associated with a direct suppression of TNF released in response to a systemic challenge with E. coli LPS. Methods Male Sprague-Dawley rats (200-225 gm) were l i g h t l y anesthetized with ether and two catheters were surgically placed under aseptic conditions by previously described methods, one into the internal jugular vein and the other into the carotid artery (11). The rats were allowed to recover for 24 hours post-catheterization and had access to food and water. The next day 5.0 mg/kg of E. coli LPS (026:B6, Difco Laboratories) was administered intravenously. Serum blood samples (0.3 ml) were collected to determine serum TNF levels at 15, 30, 60, 120, 180, and 240 min post-LPS administration. After the time course of LPS-induced serum TNF in normal animals was determined, separate groups of rats were pretreated with either saline or pentoxifylline (Hoechst-Roussel Pharmaceuticals) 50 mg/kg or 100 mg/kg intravenously 60 min prior to intravenous E. coli LPS challenge (5 mg/kg). Serum samples were taken 90 min after LPS administration for assaying TNF a c t i v i t y which corresponded to the time of peak serum TNF a c t i v i t y . Blood samples were also obtained prior to and 240 min post-LPS for total and d i f f e r e n t i a l peripheral blood counts. Selected saline and pentoxifylline treated rats were observed for 24 hours post-LPS administration to determine survival. TNF-alpha a c t i v i t y in serum specimens was determined with a cytot o x i c i t y assay u t i l i z i n g L929 cells obtained from~American Type Culture Collection, Rockville, Maryland. Briefly, 2 x 10~ L929 cells were planted in the wells of microtiter plates and allowed to adhere overnight. The following day wells were treated with serial dilutions of the specimens followed by addition of ~ctinomycin D (1.0 ug/ml, final conc.). After an 18 hour incubation at 37 in 5% CO~, the wells were washed and surviving cells indexed with a chromogenic a~say for hexosaminidase. One unit of TNF a c t i v i t y was defined as the reciprocal of the dilution required to k i l l 50% of the cells which was normalized against murine recombinant TNF (Genentech, Inc.). The 50% endpoint was calculated by a four-parameter logistic analysis. To verify that the cell k i l l i n g was due specifically to TNF-alpha, cell l y t i c a c t i v i t y was completely neutralized by the addition of rabbit antibody against murine recombinant TNF-alpha (Genentech, Inc.). Rats pretreated with saline or 100 mg/kg pentoxifylline were sacrificed 4 hours after LPS challenge for histological determinations. The animals were exsanguinated by cardiac puncture and the lungs surgically removed en bloc. The lungs were gently inflated with 10% formalin in phosphatebuffered saline solution and processed by routine histological methods for hematoxylin-eosin stained sections to be examined by l i g h t microscopy. Data were analyzed using analysis of variance and a Student's two tailed t - t e s t where indicated. All data were reported as mean ± standard error.
Vol. 47, No.
12, 1990
Pentoxifylline
and Tumor Necrosis Factor
1025
Results TNF was not detected in the serum of unchallenged rats. Intravenous LPS challenge, however, resulted in the appearance of significant TNF activity in the serum. Serum TNF peaked 90 min after intravenous LPS with levels in the 10,000 U/ml range (Figure I ) . By 3 hours after intravenous LPS, serum TNF had decreased to levels near the non-detectable level of the assay (40 U/ml). 12 10
(/) 2
1
2
3
4
Hours FIG. 1 Serum TNF activity in rats following LPS administration. Histologic sections of the lungs following intravenous LPS showed a marked sequestration of polymorphonuclear leukocytes (PMN) within the vasculature of the lung (Figure 2). This sequestration of PMN within the lung vasculature was associated with a significant decrease in the3number of PMN circulating within the peripheral b,ood from 1.7 + 0.2 x 10 PMN/mm 3 in control animals to 0.45 ± 0.02 x 103 PMN/mm3 in animaTs receiving intravenous LPS. Pretreatment with pentoxifylline prevented this LPSinduced decrease in circulatory PMN (Table I).
FIG. 2 Photomicrograph of a hematoxylin-eosin stained lung section following LPS challenge.
1026
Pentoxifylline and Tumor Necrosis Factor
Vol. 47, No. 12, 1990
TABLE I Number of Circulating PMN in Rats Injected with Saline, LPS alone, or Pentoxifylline Prior to LPS Groups
PMN Cell Count (PMN/mm3)
Saline
1.7±0.2 x 103
LPS
0.45±0.02 x 103*
LPS + Pentoxifylline 100 mg/kg
1.5±0.3 x 103
Each data point equals the mean ± SEM of 8-10 determinations; *p!O.01 compared to saline control Pretreatment with pentoxifylline 60 min prior to LPS administration inhibited serum TNF in a dose-dependent fashion (Figure 3). The highest dose of pentoxifylline reduced serum TNF to less than 5 percent of the levels seen in the saline treated animals (p~O.01 compared to saline control).
PTX 100 mg/k 9
PTX 50
mg~g
Saline o
i
,
4
6
8
,
,
lO
12
Serum TNF (xlO00
units)
FIG. 3 Serum TNF activity in rats pretreated with either saline or pentoxifylline (PTX) prior to LPS administration. The lungs of rats pretreated with 100 mg/kg pentoxifylline prior to intravenous LPS did not develop histologic evidence of PMN sequestration within the pulmonary vasculature (Figure 4).
Vol. 47, No. 12, 1990
Pentoxifylline
and Tumor Necrosis Factor
1027
FIG. 4 Histologic section of the lung of a rat pretreated with pentoxifylline prior to LPS challenge. Pentoxifylline caused a significant increase in the 24 hour survival of animals challenged with an intravenous dose of LPS. Rats in both treatment groups were completely protected against the lethal consequencesof systemic LPS administration. TABLE I I Survival in LPS-Injected Rats Pretreated with Saline or Pentoxifylline Groups
24 Hour Survival
Saline
4/10
Pentoxifylline 50 mg/kg
10/10"
Pentoxifylline 100 mg/kg
I0/I0"
Each data point equals the mean ± SEM of I0-12 determinations; *p~O.01 compared to saline control Discussion Tumor necrosis factor is emerging as an important immunoregulatory mediator of septic shock and multiple organ failure. Its production by mononuclear phagocytes is stimulated by LPS, gram-postive and gramnegative bacteria, and other cytokines. Administration of recombinant TNF to animals results in pathophysiologic abnormalities similar to those of septic shock, including tissue injury involving the lung, kidney, and bowel (4). Furthermore, studies have demonstrated that passive immunization with antibodies directed against TNF protects animals from the lethal consequences of intravenous LPS or challenge with live E. coli (6).
1028
Pentoxifylline and Tumor Necrosis Factor
Vol. 47, No. 12, 1990
Pentoxifylline has been reported to significantly improve survival in animal models of lethal infection (7). Several reports in humans have correlated mortality from severe infections with elevated systemic TNF levels (2). We, therefore, hypothesized that one possible mechanism of pentoxifylline's action may be an inhibition of TNF release in response to a systemic infectious stimulus. These in vivo data demonstrate that pentoxifylline suppressed LPS-induced TNF in a dose-dependent fashion and significantly improved animal survival. Furthermore, pentoxifylline prevented LPS-induced sequestration of PMN within the pulmonary vasculature and systemic neutropenia. Intravenous lipopolysaccharide is known to cause a dose-dependent sequestration of neutrophils within the pulmonary vascular bed (12, 13). Pulmonary PMN sequestration and increased pulmonary permeability have also been reported in guinea pigs injected intravenously with human recombinant TNF (14). TNF is also known to stimulate PMN adhesion to vascular endothelial surfaces in vitro (15). In the present investigation, pretreatment with pentoxifylline prevented LPS-induced neutrophil aggregation within the lung vasculature. Neutrophils l i k e l y play an important role in TNF-induced lung injury. TNF is known to be a potent activator of PMN functions including chemotaxis, phagocytosis, and superoxide anion production (16). Pentoxifylline has been shown to i n h i b i t the stimulatory action of TNF on PMN functions in vitro and to protect animals from LPS and TNF-induced non-cardiogenic pulmonary edema (14, 1 5 ) . Recently, pentoxifylline has been demonstrated to suppress LPS-induced mononuclear phagocyte-derived TNF in vitro at the level of both TNF mRNAaccumulation and TNF bioactivity (9). The presumed mechanism of action is secondary to the phosphodiesterase inhibitory effects of pentoxifylline resulting in an increase in intracellular cAMP. Pretreatment of mononuclear phagocytes with dibutyryl cAMP has been shown to have similar suppressive effects on TNF expression (9). The immunologic effects of pentoxifylline may, therefore, have important therapeutic implications in clinical disease states characterized by excessive or persistent TNF activity. Pentoxifylline appears to be efficacious in both directly suppressing LPS-induced TNF release and in modifying the actions of this cytokine on phagocytic cells. Therefore, i t may prove useful not only as a prophylactic agent in preventing the development of septic shock in high risk patients, but also as a therapeutic agent in patients where the inflammatory-cytokine pathway has already been ignited.
Acknowledgements This work was supported in part by Grant #Z0739 from the Cystic Fibrosis Foundation. We wish to sincerely thank Amy Weinberg, Rhonda Martinez, and Amad Jawda for their expert technical support, as well as Janel Furlong and Mamie Kinder for secretarial assistance in preparing the manuscript. The authors also wish to thank Dr. William J. Novick, Jr., Hoechst-Roussel Pharmaceuticals, for his support in this project.
Vol. 47, No. 12, 1990
Pentoxifylline
and Tumor Necrosis Factor
1029
REFERENCES i. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
B. BEUTLER and A. CERAMI, N. Engl. J. Med. 316 379-385 (1987). H.R. MICHIE, K.R. MANOGUE, D.R. SPRIGGS, A. REVHAUG, S. O'DWYER, C.A. DINARELLO, A. CEP~AMI, S.M. WOLFF and D.W. WILMORE, N. Engl. J. Med. 318 1481-1486 (1988). E. ZIEGLER, N. Engl. J. Med. 318 1533-1534 (1988). K.J. TRACEY, B. BEUTLER, S.F. LOWRY, J. MERRYWEATHER, S. WOLPE, I.W. MILSARK, R.J. HARIRI, T.J. FAHEY I l l , A. ZENTELLA, J.D. ALBERT, G.T. SHIRES and A. CERAMI, Science. 234 470-474 (1986). D.N. MANNEL, H. NORTHOFF, F. BAUSS and W. FALK, Rev. Inf. Dis. 602-606 (1987). K.J. TRACEY, Y. FONG, D.G. HESSE, K.R. MANOGUE, A.T. LEE, G.C. KUO, S.F. LOWRYand A. CERAMI, Nature. 330 662-664 (1987). G.L. MANDELL, Am. Rev. Respir. Dis. 138 1103-1105 (1988). G.E. CHALKIADAKIS, A. KOSTAKIS, P.E. KARAYANNACOS, M.E. CHALKIADAKIS, S. SGOUROMALI, H. GIOAMARELLOU, and G.D. SKALKEAS, Arch. Surg. 120 1141-1144 (1985). R.Mo STRIETER, D.G. REMICK, P.A. WARD, R.N. SPENGLER, J.P. LYNCH I l l , J. LARRICK and S.L. KUNKEL, Biochem. Biophys. Res. Com. 155 1230-1236 (1988). G.W. SULLIVAN, H.T. CARPER, W.J. NOVICK JR. and G.L. MANDELL, Inf. Immune. 56 1722-1729 (1988). V. POPOVIC and P. POPOVIC, J. Appl. Physiol. I__5727-728 (1960). G.S. WORTHEN, C. HASLETT, A.J. REES, R.S. GUMBAY, J.E. HENSON, P.M. HENSON, Am. Rev. Respir. Dis. 136 19-28 (1987). C. HASLETT, G.S. WORTHEN, P.C. GICLAS, D.C. MORRISON, J.E. HENSON, P.M. HENSON, Am. Rev. Respir. Dis. 136 9-18 (1987). C.M. LILLY, J.S. SANDHU, A. ISHIZAKA, H. HARADA, M. YONEMARU,J.W. LARRICK, T. SHI, P.T. O'HANLEY, and T.A. RAFFIN, Am. Rev. Respir. Dis. 139 1361-1368 (1989). C.H. WELSH, D. LIEN, G.S. WORTHENand J.V. WEIL, Am. Rev. Respir. Dis. 138 1106-1114 (1988). H. BESSLER, R. GILGAL, M. DJALDETTI and I. ZAHAVl, J. Leuk. Biol. 4._00747-754 (1986).
