INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

PLATELET ACTIVATING FACTOR (PAF) is a potent proinflammatory phospholipid (1-alkyl-2-acetyl-sn-glycero-3-phosphocholine) mediator produced by activated platelets, leukocytes, and endothelial cells (18). PAF exerts a variety of biological effects that include activation of platelets, leukocytes, monocytes, and macrophages and increased vascular permeability, hypotension, and contraction of smooth muscles (18). These actions of PAF are mediated through specific cell surface receptors (22). Plasma and tissue PAF levels are increased in septic shock, chronic inflammatory conditions, and allergic reactions and are modulated by regulation of enzymes involved in its synthesis and degradation (18).

PAF-acetylhydrolase (PAF-AH) catalyzes the hydrolysis of the acetyl group at the sn-2 position of PAF to produce biologically inactive lyso-PAF and acetate (33). PAF-AH also hydrolyzes oxidized phospholipids produced by the peroxidation of phosphatidylcholines containing an sn-2 polyunsaturated fatty acyl residue (36). These oxidized phospholipids also exert their effects by binding to the PAF receptor (31). PAF-AH activity is detectable in plasma as well as in the cytosol of several tissues (33). Molecular characterization and cloning studies of PAF-AH have shown that plasma and cytosolic forms are structurally distinct enzymes that are coded by different genes (15, 38). The in vivo source of plasma PAF-AH is not known; however, it has been suggested that macrophage-rich tissues are the likely source of plasma PAF-AH (5, 32, 38). In vitro studies have shown that plasma PAF-AH mRNA is induced when human monocytes differentiate into macrophages, and this is paralleled by an increase in the secretion of PAF-AH in the medium (5, 38). The enzyme secreted by macrophages is biochemically and immunologically identical to human plasma PAF-AH (32).

Plasma PAF-AH circulates in the blood as a complex with lipoproteins (33). In humans, 70% of plasma PAF-AH is associated with low-density lipoproteins (LDL) and 30% is bound with high-density lipoproteins (HDL), whereas in rodents plasma PAF-AH activity is primarily found in the HDL fraction (33). We and others have shown that the host response to infection and inflammation is associated with several alterations in lipid and lipoprotein metabolism (2, 7). Both endotoxin [lipopolysaccharide (LPS)], which induces a systemic acute phase response (APR), and cytokines, such as tumor necrosis factor (TNF), interleukin-1 (IL-1), and interleukin-6 (IL-6), which mediate the APR, produce profound changes in lipid and lipoprotein metabolism, including increased serum triglyceride and cholesterol levels, decreased HDL levels, increased hepatic fatty acid and cholesterol synthesis, increased very low density lipoprotein (VLDL) production and decreased VLDL clearance (10).

We and others have also shown that during the APR, the activity and/or mRNA levels of several proteins that are either associated with HDL or play a role in HDL metabolism are regulated. For example, the levels of serum amyloid A and apolipoprotein J are increased after LPS treatment (11, 14, 16), whereas the levels of lecithin, cholesterol acyltransferase, cholesteryl ester transfer protein, phospholipid transfer protein, hepatic lipase, and paraoxonase are decreased after LPS treatment (8, 12, 19, 21, 23). The effects of APR on plasma PAF-AH are controversial. Van Lenten et al. (39) showed that administration of croton oil, an inducer of APR (2), to rabbits decreases plasma PAF-AH activity, whereas Howard et al. (17) showed that the activity and hepatic mRNA levels of plasma PAF-AH is increased after LPS treatment in rats. On the other hand, LPS decreased PAF-AH secretion from human decidual macrophages (27) and human monocyte-derived macrophages (3) in vitro. Hence, in this study, we have examined the in vivo effects of LPS, TNF, and IL-1 on plasma PAF-AH activity and mRNA expression in several tissues in hamsters. We also studied the localization of plasma PAF-AH in lipoprotein fractions in the basal state as well as during the LPS-induced APR in hamsters. Finally, we have examined the effects of LPS, turpentine, and zymosan on plasma PAF-AH activity and mRNA expression in mice and rats to compare the effect of various APR inducers across different rodent species.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials. [-³²P]dCTP (3,000 Ci/mmol) was purchased from New England Nuclear (Boston, MA). Endotoxin (Escherichia coli 55:B5) was purchased from Difco Laboratories (Detroit, MI) and was freshly diluted to desired concentrations in pyrogen-free 0.9% saline (Kendall McGraw Laboratories, Irvine, CA). Human TNF- with a specific activity of 5 × 10⁷ U/mg was provided by Genentech (South San Francisco, CA). Recombinant human IL-1 with a specific activity of 1 × 10⁹ U/mg was provided by Immunex (Seattle, WA). The cytokines were freshly diluted to desired concentrations in pyrogen-free 0.9% saline containing 0.1% human serum albumin. Oil of turpentine (microscopy grade) was purchased from BDH Laboratories, and zymosan was purchased from Sigma (St. Louis, MO). Multiprime DNA labeling system was purchased from Amersham International, minispin G-50 columns were from Worthington Biochemical (Freehold, NJ), oligo(dT) cellulose type 77F was from Pharmacia LKB Biotechnology (Upsala, Sweden), nitrocellulose and Nytran were from Scleicher and Schuell (Keene, OH). Fuji Medical X-ray film, type Rx, was used for autoradiography. The Centricon-10 concentrator columns were purchased from Amicon (Beverly, MA). The PAF-AH assay kit was purchased from Cayman Chemicals (Ann Arbor, MI). The cDNA for plasma PAF-AH was kindly provided by Drs. D. M. Stafforini and S. M. Prescott of University of Utah (Salt Lake City, UT).

Animal procedures. Male Syrian hamsters (140-160 g) were purchased from Simonsen Laboratories (Gilroy, CA). The animals were maintained on a normal light cycle and were provided with rodent chow (Simonsen Laboratories) and water ad libitum. Animals were injected intraperitoneally with LPS, TNF, or IL-1 at the indicated doses or with 0.9% saline containing 0.1% human serum albumin. Subsequently, food was withdrawn from both control and treated animals because our previous studies showed that LPS and cytokines induce marked anorexia in Syrian hamsters (9). Animals were studied between 4 and 48 h after LPS and 16 and 24 h after cytokine administration. At the indicated times, animals were killed, blood was collected, and tissues were frozen for RNA isolation. The doses of LPS used (0.1-100 µg/100 g body wt) are far below the doses required to cause death in rodents [half-maximal lethal dose (LD₅₀) ~5 mg/100 g body wt], but have significant effects on lipid and lipoprotein metabolism in Syrian hamsters (7, 14). Similarly, the doses of TNF and IL-1 used (17 and 1 µg/100 g body wt, respectively) have marked effects on lipid and lipoprotein metabolism in Syrian hamsters (13, 14). In a separate experiment, hamsters were either splenectomized or sham operated and were allowed to recover for 2 wk. At the end of the recovery period, hamsters were injected with saline or LPS (100 µg/100 g body wt) and 24 h later, animals were killed to measure plasma PAF-AH activity and mRNA levels.

Male C57Bl/6 mice (20 g) were purchased from Jackson Laboratories (Bar Harbor, ME). The animals were maintained on a normal 12-h light cycle and were fed Purina mouse chow (Ralston Purina, St. Louis, MO) and water ad libitum. Animals were injected with either saline or LPS (5 mg/kg body wt ip), turpentine (100 µl sc), or zymosan (80 mg/kg body wt ip), and the food was withdrawn from both control and treated animals. These doses of LPS, turpentine, and zymosan were previously shown to induce APR in mice (6). Sixteen hours after the treatment, animals were killed and blood was obtained to measure plasma PAF-AH activity. In a separate experiment, Sprague-Dawley rats were injected with saline or LPS (100 µg/100 g body wt ip), food was removed from both groups, and 24 h later animals were killed to determine plasma PAF-AH activity and mRNA levels.

Plasma PAF-AH activity. Plasma PAF-AH activity was determined in control and LPS-treated hamster plasma by using a commercially available assay kit (Cayman Chemical) according to manufacturer's instructions. The assay uses 2-thio-PAF, which serves as a substrate for PAF-AH. On hydrolysis of the acetyl thioester bond by PAF-AH, free thiols are detected using 5,5'-dithiobis-2-nitrobenzoic acid (Ellman's reagent). The absorbance is read at 414 nm over a period of time using an ELISA plate reader. The detection range of the assay is from 10 to 200 nmol · min¹ · ml¹, and the assay is linear for at least 30 min. Recombinant human PAF-AH (20 ng) was used as a positive control. Absorbance values were plotted as a function of time, and the PAF-AH activity was calculated from the linear portion of the curve.

Isolation of lipoproteins. Twenty-four hours after LPS treatment, blood was collected in EDTA-containing tubes (1.5 mg/ml) and plasma was isolated. Lipoprotein fractions were isolated by using the fast protein liquid chromatography (FPLC) system (Pharmacia Biotech, Piscataway, NJ) equipped with two Superose 6 HR 10/30 columns connected in series (20), as described previously (25). Briefly, plasma was centrifuged at 12,000 g and a 0.5-ml aliquot of clear supernatant was loaded. Lipoproteins were eluted at a flow rate of 0.5 ml/min with a buffer (pH 7.4) containing 10 mM NaH₂PO₄, 150 mM NaCl, 1 mM EDTA, and 0.02% (wt/vol) NaN₃. After the initial 12 ml were eluted, 60 fractions of 0.5 ml were collected. A total of 80 ml of buffer was passed through the columns before the next sample was loaded. Cholesterol and triglyceride concentrations in the fractions were measured by commercially available enzyme assay kits (Sigma) to identify the peaks of lipoprotein fractions. VLDL, LDL, and HDL fractions were pooled and concentrated using Centricon columns with a molecular weight cutoff of 10,000. PAF-AH activity was measured using 50 µl of concentrated lipoprotein fractions.

Isolation of RNA and Northern blotting. Total RNA was isolated by a variation of the guanidinium thiocyanate method (4) as described earlier (7). Poly(A)⁺ RNA from liver, spleen, lung, small intestine, or brain was isolated using oligo(dT) cellulose and quantified by measuring absorption at 260 nm. Gel electrophoresis, transfer, and Northern blotting were performed as previously described (7). The uniformity of sample application was checked by ultraviolet visualization of acridine-orange-stained gels before transfer to Nytran membranes. We and others found that LPS increases mRNA levels of actin and cyclophilin (7, 24, 26), which are commonly used for normalizing data; therefore, the mRNA levels of these "housekeeping genes" cannot be used to study LPS-induced regulation of proteins. However, the differing direction of the changes in mRNA levels for specific proteins after LPS or cytokine treatment (increased for some proteins, decreased for some proteins, and no change for other proteins) and the magnitude of the alterations (up to 30-fold increases and 90% decreases) make it unlikely that the changes observed are due to unequal loading of mRNA. The cDNA probe hybridization was performed as described earlier (7, 24). The blots were exposed to X-ray films for various durations to ensure that measurements were carried out on the linear portion of the curve, and the bands were quantified by densitometry.

Statistics. The results are presented as means ± SE. Statistical significance between two groups was determined by using the Student's t-test. Comparisons among several groups were done by ANOVA, and statistical significance was calculated using Bonferroni's multiple-comparison test.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of LPS, TNF, and IL-1 on plasma PAF-AH activity. Our initial experiments determined the effect of LPS on plasma PAF-AH activity in Syrian hamsters. LPS (100 µg/100 g body wt) produced a 2.1-fold increase in plasma PAF-AH activity (Fig. 1A) at 16 h after treatment. The increase in PAF-AH activity after LPS administration peaked at 24 h and was sustained for at least 48 h. We next determined the dose-response curve for LPS effect on plasma PAF-AH activity after 24 h of LPS treatment (Fig. 1B). Relatively low doses of LPS (10 µg/100 g body wt) produced a maximal increase in plasma PAF-AH activity, and the half-maximal effect was seen with ~0.4 µg/100 g body wt, indicating that the increase in plasma PAF-AH activity is a very sensitive host response to LPS. The LD₅₀ for LPS-induced death in hamsters is ~5,000 µg/100 g body wt (7).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Time course (A) and dose response (B) for effect of lipopolysaccharide (LPS) on plasma platelet-activating factor (PAF)-acetylhydrolase activity. Syrian hamsters were injected intraperitoneally with either saline (Con) or 100 µg/100 g body wt LPS (A) or LPS doses indicated on x-axis (B). Blood was obtained at various time points (A) or 24 h after LPS (B). Plasma PAF-acetylhydrolase activity was determined as described in MATERIALS AND METHODS. Data are presented as means ± SE; n = 5 for each group. A: * P < 0.01, ** P < 0.001; B: * P < 0.001. BW, body wt.

We next determined the localization of plasma PAF-AH in lipoprotein fractions in control and LPS-treated (100 µg/100 g body wt, 24 h treatment) hamsters. The data presented in Fig. 2 demonstrate that, similar to other rodents (33), most of plasma PAF-AH activity in hamsters is associated with HDL. Furthermore, LPS produced a threefold increase in plasma PAF-AH activity in the HDL fraction. In the same experiment, LPS produced a 20% decrease in serum HDL-cholesterol levels (control 49 ± 2.3; LPS 39 ± 1.4 mg/dl, P < 0.001) while increasing serum total cholesterol levels by 54% (control 94 ± 5.0; LPS 145 ± 6.5 mg/dl, P < 0.001). These results indicate that the activity of plasma PAF-AH per HDL particle is increased after LPS.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of LPS on plasma PAF-acetylhydrolase in lipoprotein fractions. Animals were injected intraperitoneally with either saline or LPS (100 µg/100 g body wt), and 24 h later, blood was obtained and lipoproteins were isolated using a fast protein liquid chromatography system. Plasma PAF-acetylhydrolase activity was determined in low- and high-density (LDL and HDL, respectively) fractions as described in MATERIALS AND METHODS. Data are presented as means ± SE; n = 6 for each group; * P < 0.001.

We also determined the effect of 24 h treatment with TNF (17 µg/100 g body wt; i.e., 8.5 × 10⁵ U/100 g body wt) and IL-1 (1 µg/100 g body wt; i.e., 1 × 10⁶ U/100 g body wt) on plasma PAF-AH activity. Compared with LPS, both TNF and IL-1 produced only modest increases (33 and 46%, respectively) in plasma PAF-AH activity (Fig. 3). Higher doses of TNF and IL-1 did not produce any further increase in plasma PAF-AH activity (data not shown).

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Effect of tumor necrosis factor (TNF) and interleukin-1 (IL-1) on plasma PAF-acetylhydrolase activity. Animals were injected intraperitoneally with saline, TNF (17 µg/100 g body wt), or IL-1 (1 µg/100 g body wt), and 24 h later, blood was obtained and plasma PAF-acetylhydrolase activity was determined as described in MATERIALS AND METHODS. Data are presented as means ± SE; n = 5 for all groups; * P < 0.01, ** P < 0.002.

Effect of LPS, TNF, and IL-1 on plasma PAF-AH mRNA expression. To investigate the mechanism by which LPS increases plasma PAF-AH activity in Syrian hamsters, we determined the effect of LPS treatment on PAF-AH mRNA expression in several tissues. Syrian hamsters were injected with LPS (100 µg/100 g body wt) or saline (controls), and 16 h later, tissues were collected for poly(A)⁺ RNA isolation. Figure 4A shows a Northern blot of the effect of LPS on plasma PAF-AH mRNA expression, and the densitometric analysis of blot is presented in Fig. 4B. LPS increased PAF-AH mRNA levels in liver (7.3-fold), spleen (4.4-fold), lung (4.5-fold), small intestine (3.6-fold), and brain (45%). Plasma PAF-AH mRNA was not detectable in adipose tissue, heart, and skeletal muscle in both control and LPS-treated animals.

View larger version (90K): [in this window] [in a new window]   Fig. 4.   Effect of LPS on plasma PAF-acetylhydrolase mRNA levels in various tissues. Syrian hamsters were injected intraperitoneally either with saline or LPS (100 µg/100 g body wt), food was removed from both groups, and 16 h later, animals were killed and tissues were harvested for poly(A)+ RNA isolation. Plasma PAF-acetylhydrolase mRNA levels were determined by Northern blotting as described in MATERIALS AND METHODS. A: Northern blot showing the effect of LPS on plasma PAF-acetylhydrolase mRNA in liver, spleen, lung, small intestine (S intestine), and brain. B: data presented as %control as quantified by densitometry. Data are presented as means ± SE; n = 4 or 5 for each group in each tissue; * P < 0.001.

We next determined the time course and dose response of LPS-induced increase in plasma PAF-AH mRNA levels in liver and spleen. A maximal increase of 8.3-fold was seen in plasma PAF-AH mRNA levels in liver at 8 h after LPS treatment (Fig. 5A), which was sustained for at least 24 h. Plasma PAF-AH mRNA levels in spleen were increased by 3.1-fold at 8 h, 4-fold at 16 h, and 4.4-fold at 24 h after LPS (Fig. 5B). The dose-response experiments reveal that 10 µg/100 g body wt LPS dose was sufficient to induce a maximal increase in plasma PAF-AH mRNA levels in liver and spleen (Fig. 6). The half-maximal doses for LPS-induced increase in plasma PAF-AH mRNA in the liver (Fig. 6A) and spleen (Fig. 6B) were 0.4 and 0.1 µg/100 g body wt, respectively.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Time course of LPS effect on plasma PAF-acetylhydrolase mRNA levels in liver (A) and spleen (B). Syrian hamsters were injected intraperitoneally either with saline (controls) or LPS (100 µg/100 g body wt), and food was removed from both groups. At indicated times, animals were killed and tissues were obtained for poly(A)+ RNA isolation. Plasma PAF-acetylhydrolase mRNA levels were determined by Northern blotting as described in MATERIALS AND METHODS. Data are presented as %control as quantified by densitometry. Data are presented as means ± SE; n = 4 or 5 for each time point; A: * P < 0.005, ** P < 0.002, *** P < 0.001; B: * P <0.001 vs. saline control.

View larger version (28K): [in this window] [in a new window]   Fig. 6.   Dose response of LPS effect on plasma PAF-acetylhydrolase mRNA levels in liver (A) and spleen (B). Syrian hamsters were injected intraperitoneally either with saline or LPS doses indicated on x-axis, food was removed from all groups, and 24 h later, animals were killed and tissues were obtained for poly(A)+ RNA isolation. Plasma PAF-acetylhydrolase mRNA levels were determined by Northern blotting as described in MATERIALS AND METHODS. Data are presented as %control as quantified by densitometry. Data are presented as means ± SE; n = 4 for each dose; A: * P < 0.02, ** P < 0.001; B: * P < 0.005, ** P < 0.001 vs. saline control.

The data presented in Fig. 7 show the effect of 16 h TNF and IL-1 treatment on plasma PAF-AH mRNA levels in liver and spleen. TNF produced 56 and 60% increases in plasma PAF-AH mRNA expression in liver and spleen, respectively, whereas IL-1 increased plasma PAF-AH mRNA levels by 2-fold in liver and 2.3-fold in spleen (Fig. 7).

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Effect of TNF and IL-1 on plasma PAF-acetylhydrolase mRNA levels in liver and spleen. Syrian hamsters were injected intraperitoneally with saline, TNF (17 µg/100 g body wt), or IL-1 (1 µg/100 g body wt), and 16 h later, animals were killed and tissues were obtained for poly(A)+ RNA isolation. Plasma PAF-acetylhydrolase mRNA levels were determined by Northern blotting as described in MATERIALS AND METHODS. Data are presented as %control as quantified by densitometry. Data are presented as means ± SE; n = 5 for each group; * P < 0.05, ** P < 0.001.

Effect of splenectomy on plasma PAF-AH activity and mRNA expression in liver. Although the in vivo source of plasma PAF-AH is not known, it has been suggested that macrophage-rich tissues may be the likely source (5, 32, 38). Because the basal level of expression for plasma PAF-AH mRNA was high in spleen (Fig. 4A), we hypothesized that spleen may be a potential source of plasma PAF-AH and that surgical removal of spleen may decrease basal plasma PAF-AH activity and/or blunt the LPS-induced increase in plasma PAF-AH activity. To investigate this hypothesis, we compared plasma PAF-AH activity in sham-operated versus splenectomized hamsters that were either treated with saline or LPS (100 µg/100 g body wt, 24 h treatment). The data presented in Fig. 8A show that splenectomy had no effect on basal plasma PAF-AH activity. Moreover, LPS produced a 2.3- and 2.5-fold increase in plasma PAF-AH activity in sham-operated and in splenectomized hamsters, respectively, suggesting that spleen per se may not contribute to basal or LPS-induced increase in plasma PAF-AH activity. The basal levels of plasma PAF-AH mRNA levels in liver were 2.3-fold higher in splenectomized compared with sham-operated hamsters (Fig. 8B). Furthermore, LPS produced a 4.7-fold increase in plasma PAF-AH mRNA levels over basal levels in splenectomized hamsters compared with a 5.2-fold increase in sham-operated hamsters. These results indicate that in the absence of spleen, the basal as well as LPS-induced expression of plasma PAF-AH mRNA in liver is upregulated.

View larger version (48K): [in this window] [in a new window]   Fig. 8.   Effect of LPS on plasma PAF-acetylhydrolase activity (A) and hepatic mRNA levels (B) in sham-operated and splenectomized hamsters. Syrian hamsters were either sham-operated or splenectomized, allowed to recover for 2 wk, and then injected with saline or LPS (100 µg/100 g body wt). Twenty-four hours later, blood was obtained and livers were harvested. Plasma PAF-acetylhydrolase activity and hepatic mRNA levels were determined as described in MATERIALS AND METHODS. Data are presented as means ± SE; n = 6 for each group; A: * P < 0.001 vs. respective controls; B: * P < 0.01 vs. sham-operated control; ** P < 0.001 vs. respective controls.

Effect of LPS, turpentine, and zymosan on plasma PAF-AH in mice and rats. In addition to LPS, several other stimuli such as turpentine and zymosan are also used to induce the APR (6). To determine whether the increase in plasma PAF-AH activity is limited to LPS or can be seen with other APR inducers, we injected mice with LPS (5 mg/kg body wt ip), turpentine (100 µl sc), or zymosan (80 mg/kg body wt ip) and 16 h later obtained blood to measure plasma PAF-AH activity. These doses of LPS, turpentine, and zymosan were previously shown to induce the APR in mice (6). The data presented in Fig. 9 demonstrate that LPS produced a 78% increase in plasma PAF-AH activity, whereas turpentine and zymosan increased plasma PAF-AH activity by 70 and 37%, respectively. LPS (1 mg/kg body wt ip) also produced a 90% increase in plasma PAF-AH activity in rats after 24 h of treatment (Fig. 10A). Similar to hamsters, the basal expression of plasma PAF-AH mRNA was very low in rat liver and was readily detectable in rat spleen. LPS produced 11.6- and 3.3-fold increases in plasma PAF-AH mRNA expression in rat liver and spleen, respectively (Fig. 10B). Taken together, these results suggest that the effect of APR on plasma PAF-AH activity and mRNA expression is consistent across different rodent species.

View larger version (35K): [in this window] [in a new window]   Fig. 9.   Effect of LPS, turpentine (Turp), and zymosan (Zym) on plasma PAF-acetylhydrolase activity in mice. Animals were injected with saline, LPS (5 mg/kg body wt ip), turpentine (100 µl sc), or zymosan (80 mg/kg body wt ip), and 16 h later, blood was obtained and plasma PAF-acetylhydrolase activity was determined as described in MATERIALS AND METHODS. Data are presented as means ± SE; n = 5 for all groups; * P < 0.05, ** P < 0.001.

View larger version (42K): [in this window] [in a new window]   Fig. 10.   Effect of LPS on plasma PAF-acetylhydrolase activity (A) and mRNA levels in liver and spleen (B) in rats. Rats were injected with saline or LPS (1 mg/kg body wt ip), and 24 h later blood was obtained and tissues were harvested. Plasma PAF-acetylhydrolase activity and mRNA levels were determined as described in MATERIALS AND METHODS. Data are presented as means ± SE; n = 5 for each group; A: * P < 0.001, B: * P < 0.001.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The effect of the APR on PAF-AH has been controversial. Van Lenten et al. (39) reported that PAF-AH activity in rabbits is decreased in response to croton oil, an inducer of the APR, whereas Howard et al. (17) recently reported that LPS increases PAF-AH in rats. The results of our present study demonstrate that LPS increases plasma PAF-AH activity in Syrian hamsters. The LPS-induced increase in plasma PAF-AH activity is sustained and requires relatively low doses of LPS, indicating that the induction of plasma PAF-AH activity is a very sensitive host response. LPS also increased plasma PAF-AH activity in rats as well as mice, suggesting that the stimulatory effect of LPS on PAF-AH is consistent across different rodent species. In addition to LPS, turpentine and zymosan are also widely used to induce APR (6). Our results demonstrate that similar to LPS, turpentine and zymosan also produce a marked increase in plasma PAF-AH activity in mice, indicating that plasma PAF-AH is induced in a variety of models of APR. Thus, in conjunction with the studies of Howard et al. (17), our results indicate that PAF-AH is a positive acute phase protein. The explanation for the decrease in PAF-AH after croton oil in the studies of Van Lenten et al. (39) is not clear, but may be related to species differences. It is important to note that there are differences in the species specificity of various acute phase proteins; the kinetics, the magnitude, and even the direction of change for individual acute phase proteins may vary among different species (29). Alternatively, it is possible that croton oil induces a different pattern of proinflammatory and anti-inflammatory cytokines and/or cytokine inhibitors that may downregulate plasma PAF-AH activity.

An increase in plasma PAF-AH activity during the host response to infection and inflammation can have several beneficial effects. Recombinant plasma PAF-AH has been shown to reduce the inflammatory effects of PAF both in vivo and in vitro (38), suggesting that it plays an important role in the regulation of PAF levels in blood and tissues. Plasma PAF levels are elevated in septic shock and chronic inflammatory conditions (18). An increase in plasma PAF-AH activity will enhance the degradation of plasma PAF during infection and inflammation and will reduce the pathological effects of PAF. In addition to degradation of PAF, PAF-AH also hydrolyzes oxidatively fragmented phospholipids (36), which are mitogenic and proinflammatory. Thus an increase in plasma PAF-AH may protect the organism against the harmful effects of oxidized phospholipids that are produced during inflammatory conditions.

The in vivo source of plasma PAF-AH is not known. Previous studies have shown that macrophages secrete large amounts of PAF-AH in the media (5, 32). Cultured hepatocytes and HepG2 cells, a human hepatoma cell line, also secrete PAF-AH activity in the medium (37), although at relatively low levels compared with macrophages. The relative contribution of resident or tissue macrophages to circulating plasma PAF-AH levels is not known. Although the initial cloning study did not find detectable levels of plasma PAF-AH mRNA in several tissues (38), a recent study reported that plasma PAF-AH mRNA is widely distributed; however, most of the tissues have a very low level of expression under basal conditions (3). Our results on the mechanism of LPS-induced increase in plasma PAF-AH activity indicate that LPS increases plasma PAF-AH mRNA expression in several macrophage-rich tissues, including spleen, liver, lung, and small intestine, indicating that multiple tissues may contribute to the increased plasma PAF-AH activity during the APR. Kupffer cells have recently been shown to be the cell type responsible for LPS-induced increase in plasma PAF-AH mRNA expression in the rat liver (17). Because the basal plasma PAF-AH mRNA expression was high in spleen, we postulated that spleen may partly contribute to plasma PAF-AH activity. However, our studies in splenectomized hamsters demonstrate that surgical removal of spleen has no effect on basal or LPS-induced plasma PAF-AH activity, suggesting that spleen per se may not contribute to plasma PAF-AH activity. On the other hand, it is also possible that the liver may compensate for the absence of spleen by providing more plasma PAF-AH under basal conditions as well as during the APR. This concept is supported by the observation that the basal as well as LPS-induced expression of PAF-AH mRNA is approximately twofold higher in splenectomized compared with sham-operated hamsters (P < 0.001). The increased expression of plasma PAF-AH mRNA in livers of splenectomized animals may be related to the enhanced proliferation of Kupffer cells that is seen after splenectomy (28) and to the increased responsiveness of Kupffer cells to LPS after splenectomy (1).

TNF and IL-1 are known to mediate many of the metabolic effects of LPS (10). In the present study, we demonstrate that both TNF and IL-1 significantly increased plasma PAF-AH mRNA expression in the liver and spleen in hamsters. TNF and IL-1 also produced a modest increase in plasma PAF-AH activity. It is important to note that the magnitude of the increase in plasma PAF-AH activity induced by TNF or IL-1 is relatively small compared with the increase in plasma PAF-AH mRNA levels or activity seen with LPS. It is possible that, in addition to TNF and IL-1, other cytokines, hormones, or lipid mediators induced during the host response to infection or inflammation may play a role in mediating the effect of LPS on plasma PAF-AH. For example, corticosteroids are induced during the APR and it has been shown previously that dexamethasone, a potent glucocorticoid, increases plasma PAF-AH activity in rats (40). PAF itself, but not the degradation product lyso-PAF, also stimulates the synthesis and secretion of PAF-AH by HepG2 cells and cultured human monocyte-derived macrophages (3, 30), indicating a feedback regulatory mechanism that may serve to protect the host from the pathological effect of PAF.

Plasma PAF-AH is present in the circulation as a lipoprotein-associated enzyme. Our results demonstrate that in Syrian hamsters most of the plasma PAF-AH is associated with HDL. Moreover, the increase in plasma PAF-AH activity that is induced by LPS is found in the HDL fraction, despite a significant decrease in serum HDL levels, suggesting that the activity of plasma PAF-AH per HDL particle is substantially increased during the host response to infection and inflammation. Previous studies from our laboratory and others have shown that LPS induces profound changes in several enzymes that are associated with HDL or play a role in HDL metabolism (8, 11, 12, 14, 16, 19, 21, 23). It is likely that these changes may alter the structure and biological function of HDL during infection and inflammation.

Perspectives