L-4F Alters Hyperlipidemic (but not Normal) Mouse Plasma to Reduce Platelet Aggregation

Abstract

Objective

Hyperlipidemia is associated with platelet hyper-reactivity. We hypothesized that L-4F, an apoA-I mimetic peptide, would inhibit platelet aggregation in hyperlipidemic mice.

Methods and Results

Injecting L-4F into apoE null and LDL receptor null mice resulted in a significant reduction in platelet aggregation in response to agonists but there was no reduction in platelet aggregation after injection of L-4F into wild-type (WT) mice. Consistent with these results, injection of L-4F into apoE null mice prolonged bleeding time but not in WT mice. Incubating L-4F in vitro with apoE null platelet rich plasma also resulted in decreased platelet aggregation. However, incubating washed platelets from either apoE null or WT mice with L-4F did not alter aggregation. Compared to wild-type mice, unstimulated platelets from apoE null mice contained significantly more 12-HETE, thromboxane A₂ (TXA₂), prostaglandins D₂ (PGD₂) and E₂ (PGE₂). In response to agonists, platelets from L-4F treated apoE null mice formed significantly less TXA₂, PGD₂ PGE₂, and 12-HETE.

Conclusions

By binding plasma oxidized lipids that cause platelet hyper-reactivity in hyperlipidemic mice, L-4F decreases platelet aggregation.

Hyperlipidemia is a risk factor associated with oxidative stress, generation of oxidized lipoproteins, platelet hyper-reactivity and thrombogenic potential^1,2. Increasing evidence indicates that interactions between platelets and oxidized lipoproteins play a major role in the initiation, development and progression of atherosclerosis^3–5. Apolipoprotein A-I (ApoA-I)⁶, apoA-I _Milano ⁷ or high-density lipoprotein (HDL)⁸ have been shown to inhibit platelet hyper-reactivity and reverse the pro-thrombotic effects of hyperlipidemia. The atheroprotective and antithrombotic activity of HDL is generally attributed to the ability of HDL to promote reverse cholesterol transport and to the antioxidant, antiinflammatory properties of the lipids and proteins associated with HDL⁹. We have recently reported¹⁰ that the apoA-I mimetic 4F peptides bind oxidized lipids with much higher affinity than apo-AI thus, explaining their ability to be effective in animal models and in preliminary human studies in vivo or in vitro at very low concentrations.

Patients afflicted with hyperlipoproteinaemia show enhanced platelet reactivity, increased production of the arachidonic (AA) metabolites thromboxane A₂ (TXA₂) and 12-hydroxyeicosatetraenoic acid (12-HETE), secretion of mediators from platelet dense granules and a reduction of prostacyclin receptors on the platelet membrane that might prevent the anti-platelet action of prostacyclin^11,12.

Based on the reported evidence on the role of HDL in platelet activation in hyperlipidemia and the knowledge that hyperlipidemic mice have elevated levels of oxidized lipids¹³ we hypothesized that L-4F would exert an inhibitory effect on increased platelet aggregability in hyperlipidemic mice.

METHODS

Materials

The apoA-I mimetic peptide L-4F was synthesized as described¹⁴. For other Materials please see supplemental material (available online at http://atvb.ahajournals.org).

Mice

For details on the mice used in these studies please see supplemental material (available online at http://atvb.ahajournals.org). The mice were injected subcutaneously with L-4F at doses ranging between 0.01 to 1 mg/kg/day in 0.1 ml of vehicle (ABCT) containing 50 mM ammonium bicarbonate, pH 7.0 and 0.1 mg/ml Tween-20 or vehicle only for various time periods and one hour after the last injection, blood was collected. All animal procedures were approved by the UCLA, Animal Research Committee.

Methods

Platelet preparation and aggregation

Blood was collected in 3.8% sodium citrate (9:1, v:v) from the retro-orbital plexus of isoflurane-anaesthetized mice using heparin-coated capillary tubes. Blood from 8–10 mice was pooled for the experiments. The citrated blood was centrifuged at 150g for 10 min at room temperature to obtain platelet-rich plasma (PRP), and centrifuged again at 1800g for 15 min to obtain platelet-poor plasma (PPP)¹⁵,. Contaminating red blood cells and leukocytes in the PRP were removed by a two minute centrifugation at 180 g. The number of platelets in PRP was counted manually by light microscopy using a hemocytometer at a magnification of 400×¹⁶. Five PRP samples were counted by one investigator and recounted by two additional observers blinded to treatment. The coefficient of variation for inter-observer measurements was found to be 12 ± 1%. Platelet number in the PRP was adjusted to 1–5 × 10⁸ cells/mL with PPP as a diluent. For the preparation of washed platelets blood collected in ACD (acid-citrate-dextrose) was processed as described without any modifications¹⁷. Briefly, PRP containing prostaglandin E1 (1 µg/mL) to prevent activation during washing, was sedimented by centrifugation at 1800 g for ten minutes and gently washed twice with the platelet wash buffer (pH 6.5). After centrifugation the platelet pellet was resuspended in a modified calcium-free Hepes -Tyrode buffer (pH 7.4) and diluted to the final concentration of 1–3 × 10⁸ platelets/mL¹⁷. Platelets suspended either in plasma or in Hepes – Tyrode buffer were incubated at room temperature for 30 min with gentle agitation. Murine fibrinogen (1 mg/mL) and CaCl₂ (1 mM) were added to the washed platelet suspension 30 seconds prior to agonist addition. Platelet aggregation was conducted at 37°C at a constant stirring rate of 1000 rpm in a 4-channel PAP-4 Platelet Aggregation Profiler (Bio/Data Corporation, Horsham, PA). Aggregation of platelets was elicited by addition of the following agonists: adenosine diphosphate (ADP, 0.5–20 µM), collagen (0.048–0.190 mg/mL) and arachidonic acid (AA, 25–500 µg/mL). Sub-optimal concentrations of agonists were used after establishing the concentrations that caused minimal and maximal aggregation for each experiment¹⁸. All experiments were repeated at least three times using platelets from different mice. The resulting aggregation measured as the change in light transmission was recorded until a plateau was reached. The following platelet aggregation parameters were used to evaluate the effects of L-4F on platelets: amplitude or the extent of aggregation represented by the total decrease in the optical density (OD) and expressed as % aggregation, the slope or V_max (the maximum sustained rate of aggregation) determined from the steepest slope of the aggregation curve and expressed as a change in the OD per second, and lag time (for PRP only) representing the elapsed time (sec) between agonist addition and the start of aggregation^{19, 20}.

For in vitro studies of L-4F (0.01, 0.1 and 1 µg/mL) or vehicle alone was added to PRP or to washed platelets and incubated 60 min at 37°C with gentle stirring in the aggregometer and platelet aggregation was determined in response to the indicated agonists.

Measurement of TXA₂ formation

Mice were injected subcutaneously with L-4F at a dose of 1 mg/kg/day in 0.1 ml of vehicle (ABCT) or vehicle only on day one and were injected twice more every 24hr. One hour after the last injection, blood was collected for platelet preparation and aggregation studies. Formation of TXA₂ in PRP containing 1×10⁸ platelets/mL from L-4F or vehicle injected mice was measured by determining TXB₂, the stable metabolite of TXA₂ in the samples used for platelet aggregation in the presence or the absence of the COX-1 inhibitor, SC-560 (1µM) which was added in DMSO and preincubated with the PRP for 30 min at room temperature. The reaction initiated by addition of ADP (20 µM), collagen (0.190 mg/mL) or AA (500 µg/mL) for 5 minutes at 37°C at a stirring rate of 1000 rpm was terminated by addition of 80 µM aspirin and 10 mM EDTA followed by rapid freezing and storing at −80°C²⁰. The amount of TXB₂ in medium was determined by using the TXB₂ EIA kit, according to the procedure described by the manufacturer (Assay Designs, Inc. (Ann Arbor, MI).

LC-MS/MS analysis

Please see supplemental material (available online at http://atvb.ahajournals.org).

Measurement of tail bleeding time

Mice were injected subcutaneously with L-4F at a dose of 1 mg/kg/day in 0.1 mL of vehicle or vehicle only on day one and were injected twice more every 24hr. One hour after the last injection, tail bleeding time was measured in isoflurane anesthetized mice. Aspirin (10 mg/kg) administered by stomach gavage 24 hr prior to bleeding was used as a positive control. For details of the method used to measure tail bleeding time please see supplemental material (available online at http://atvb.ahajournals.org).

Other procedures

Plasma lipoprotein and lipid levels were determined as described previously²¹.

Statistical Analyses

For methods used for statistical analyses please see supplemental material (available online at http://atvb.ahajournals.org). Differences were considered statistically significant at a value of p<0.05 or less.

RESULTS

L-4F inhibits agonist-induced platelet aggregation ex vivo in apo E null and LDL receptor null mice

L-4F (0.01, 0.1 and 1 mg/kg) injected subcutaneously dose-dependently inhibited the ex vivo aggregation of platelets (Figure 1) stimulated with: (a) ADP (IC₅₀ values of 2.6, 6.69, 7.32**and 7.56* µM for administration of ABCT or L-4F at doses of 0.01, 0.1, and 1 mg/kg respectively; *p<0.05; **p<0.01); (b) collagen (IC₅₀ values of 0.0035, 0.013, 0.049 and 0.082* mg/mL respectively); (c) AA with IC₅₀ values of 91.4, 95.6, 551** and 685** µg/mL respectively. In addition, L-4F significantly reduced the slope and increased the lag time in response to AA, collagen and ADP (data not shown).

Figure 1.

ApoE null mice were injected subcutaneously with L-4F 0.01, 0.1 and 1 mg/kg or with vehicle ABCT for 48 hours. One hour after the last dose the mice were bleed, PRP was prepared and the percent of platelet aggregation in response to increasing concentrations of ADP (panel A), collagen (panel B), and arachidonic acid (AA) (panel C), was determined as described in Methods. Blood from 8–10 mice was pooled for each experiment. The data shown are Mean ± SEM (n=5). *p<0.05; ** p<0.01.

Similar results were obtained when L-4F (but not vehicle) was injected intoLDLR null mice fed an atherogenic diet. Because of the marked hyperlipidemia associated with turbid plasma that was induced in these mice it was necessary to use washed platelet suspensions rather than PRP. Aggregation elicited by collagen (0.096 mg/mL) was significantly decreased (p = 0.003) in the washed platelet suspensions from L-4F injected LDLR null mice (72.00 ± 3.61 percent aggregation) but not in those from vehicle injected LDLR null mice (101.33 ± 2.85 percent aggregation). AA (250 µg/mL) induced aggregation was also significantly inhibited (p = 0.006) in the washed platelet suspensions from L-4F injected LDLR null mice (52.33 ± 5.89 percent aggregation) but was not significantly inhibited in the platelets from vehicle-injected LDLR null mice (89.33 ± 3.67 percent aggregation).

L-4F inhibits agonist-induced platelet aggregation in vitro in the presence of plasma from apoE null mice but not in the absence of plasma

Incubation of PRP from apoE null mice with L-4F (500 ng/mL) (but not vehicle) significantly reduced platelet aggregation in response to ADP (Figure 2, panel A) and collagen (Figure 2, panel B). In contrast to these results, adding L-4F in vitro to washed platelets from either apoE null (Figure 3, panels A, B and C) or WT mice did not inhibit platelet aggregation (Figure 3, panels D, E and F).

Figure 2.

L-4F or vehicle alone was added to PRP containing 3 × 10⁸ platelets/mL from untreated apoE null mice and incubated and platelet aggregation was determined in response to ADP (Panel A) and collagen (Panel B) as described in Methods. Blood from 8–10 mice was pooled for each experiment. The data shown are Mean ± SEM from three different experiments and are expressed as the percent of platelets aggregating (% Aggregation). *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 3.

Washed platelets (3 × 10⁸ platelets/mL) from untreated apoE null mice were prepared as described in Methods, and incubated for 60 min under gentle agitation with L-4F (1 µg/ml) or vehicle. Platelet aggregation was determined in response to ADP (Panel A), collagen (Panel B) and arachidonic acid (AA) (Panel C) as described in Methods. Washed platelets (3 × 10⁸ platelets/mL) from untreated type C57BL/6 mice were prepared as described in Methods, and incubated for 60 min under gentle agitation with L-4F (0.01, 0.1 µg/ml and 1 µg/ml or vehicle. Platelet aggregation was determined in response to ADP (Panel D), collagen (Panel E) and arachidonic acid (AA) (Panel F) as described in Methods. Blood from 8–10 mice was pooled for each experiment. The data shown are Mean ± SEM from three different experiments and are expressed as the percent of platelets aggregating (% Aggregation).

These results indicate that L-4F does not have a direct effect on platelets but acts through some component of plasma that influences platelet function. As previously reported²², there was no change in plasma total cholesterol, triglycerides, HDL-cholesterol or apoB containing-cholesterol levels after injection of L-4F in these experiments (data not shown).

A single subcutaneous injection of L-4F in apoE null mice significantly inhibited platelet aggregation in response to ADP, collagen or AA for up to seventy-two hours after injection (data not shown). Ninty-six hours after injection there was no significant difference in platelet aggregation between mice injected with L-4F or vehicle alone in response to any of the agonists (data not shown).

While a single daily injection subcutaneously of L-4F of 0.01 mg/kg day was without a significant effect (Figure 1) administration of the peptide by Alzet osmotic pumps delivering L-4F at a dose of 0.01 mg/kg/day for two weeks significantly reduced platelet aggregation in response to ADP or collagen or AA compared to mice implanted with pumps delivering the same amount of vehicle without L-4F (data not shown).

L-4F inhibits agonist-induced thromboxane B₂ (TXB₂) production

Thromboxane B₂ (TXB₂) is the stable hydrolysate of TXA₂, which is generated from the AA released from platelet plasma membranes following the sequential activities of the cytosolic phospholipase A₂ (cPLA₂) – cyclooxygenase 1 – (COX-1) - TXA₂ synthase pathway²³. Following treatment with L-4F, platelet stimulation with AA resulted in significantly less TXB₂ production, ~ 30% less (data not shown). Under these conditions in the same samples aggregation was reduced by more than 40% (data not shown). Addition of the COX-1 inhibitor SC-560, further inhibited platelet TXB₂ formation and aggregation in platelets taken from both vehicle and L-4F treated mice (data not shown). Similarly, the slope exhibited significant reductions that were comparable with the reduction in TXB₂ production (data not shown). Conversely, the lag times were increased in the platelets collected from the L-4F injected mice and addition of SC-560 to these platelets further increased the lag time, although it did not reach statistical significance, whereas pre-incubation of platelets from the vehicle treated mice with SC-560 significantly prolonged the lag time (data not shown). Similar results were obtained when the platelets were stimulated with collagen (data not shown). In the platelets stimulated with ADP, the addition of SC-560 did not significantly inhibit these parameters beyond that achieved with L-4F suggesting that L-4F may inhibit the same segment of the AA-COX-1-TXA₂ pathway that is antagonized by SC-560 under these conditions (data not shown).

L-4F treatment inhibits agonist-induced formation of TXB₂, PGD₂, PGE₂ and 12-HETE without changing the concentration of plasma lipids

In addition to TXB₂, a number of other eicosanoids derived from the AA cascade via COX-1 and 12-Lipoxygenase (12-LO) enzymatic pathways were analyzed by LC-MS/MS. Platelet rich plasma obtained from L-4F treated or vehicle treated mice was stimulated with the following agonists: collagen (0.05 mg/mL), ADP (20 µM), AA (125 µg/ml), calcium ionophore A23187 (2.5 µM), and the thromboxane A2 mimetic U44619 (1µM)²⁴.

As shown in Table 1 except for U44619 which directly activates platelet TXA₂ receptors²⁴, the production of of AA metabolites 12-HETE, TXB₂, PGD₂ and PGE₂ was significantly diminished in platelets from L-4F treated mice versus vehicle controls. The rank order for AA metabolites calculated as ng/3×10⁸ platelets was 12-HETE > TXA₂ > PGE₂ > PGD₂ or calculated as a percent of the most abundant metabolite: 12-HETE (100%) > TXA₂ (12%) >PGE₂ (0.6%) > PGD₂ (0.3%). In stimulated platelet suspensions preincubated with SC-560 production of TXB₂ was significantly reduced in an agonist dependent manner with the potency rank order of AA > collagen > A23187 > U46619 > ADP, and was associated with a marked increase of 12-HETE formation (Table 2) suggesting that in the presence of COX-1 inhibitor the AA is diverted towards the 12-LO pathway to generate additional 12-HETE²⁵.

Table 1.

ApoE null mice were injected with L-4F at a dose of 1 mg/kg/day for 48 hours or were injected with vehicle alone. One hour after the last injection the mice were bled. Blood from 8–10 mice was pooled for each experiment. Platelet rich plasma (PRP) was prepared and exposed to the agonists shown at the concentrations indicated as described in Methods. The formation of 12-HETE, TXB₂, PGD₂, and PGE₂ was determined as described in Supplemental Methods. The values obtained in platelets taken from the vehicle treated mice were normalized to 1.0 and the values obtained in platelets taken from the L-4F treated mice were expressed as the percent change from control (i.e. percent of the value obtained in the vehicle treated mice). The data shown are Mean ± SEM (n = number of experiments; ns = not significant).

Agonist	12-HETE L-4F-injected mice % change from vehicle	TXB2 L-4F-injected mice % change from vehicle	PGD2 L-4F-injected mice % change from vehicle	PGE2 L-4F-injected mice % change from vehicle
Collagen (0.05 mg/mL)	58.7 ± 5.4 p<0.001 n=7	22.2 ± 11.7 p<0.0001 n=7	27.8 ± 7.1 p<0.0001 n=7	28.8 ± 10.5 p<0.0001 n=7
ADP (20 µM)	75.8 ± 3.3 p<0.001 n=5	76.2 ± 12.1 P<0.05 n=5	83.4 ± 5.0 P<0.05 n=5	57.4 ± 8.9 p<0.001 n=5
A23187 (2.5 µM)	53.9 ± 17.7 p<0.05 n=4	39.6 ± 20.1 p<0.05 n=4	39.1 ± 16.3 p<0.01 n=4	44.5 ± 15.3 p<0.05 n=4
Arachidonic acid (125 µg/mL)	73.4 ± 9.42 p<0.05 n=5	67.4 ± 16.8 p<0.05 n=5	64.2 ± 15.3 p<0.05 n=5	69.4 ± 16.7 p<0.05 n=5
U46619 (1 µM)	95.7 ± 9.2 ns n=6	93.0 ± 21.2 ns n=6	104.7 ± 25.8 ns n=6	77.3 ± 19.0 ns n=6

Table 2.

A specific inhibitor of COX-1 (SC-560) reduces the formation of TXB₂ and increases the formation of 12-HETE in platelets obtained from apoE null mice. The platelets obtained from vehicle treated or L-4F treated apoE null mice following the treatment protocol described in Table 1 were pre-incubated with SC-560 (1 µM) or the vehicle used to add the SC-560 (DMSO) for 30 min at room temperature prior to being stimulated with the agonists shown at the indicated concentrations as described in Methods. The values obtained for TXB₂ or 12-HETE with addition of DMSO were normalized to 1.0 and the values obtained for TXB₂ or 12-HETE in platelets pre-incubated with SC-560 were expressed as a percent of the DMSO values (i.e. % of DMSO). The data shown are Mean ± SEM (n = number of experiments; ns = not significant).

Agonist	TXB2 Vehicle + SC-560 % of DMSO	TXB2 L-4F+SC-560 % of DMSO	12-HETE Vehicle + SC-560 % of DMSO	12-HETE L-4F+SC-560 % of DMSO
Collagen (0.05 mg/mL)	2.3 ± 0.7 p<0.001 n=3	1.7± 0.7 p<0.001 n=3	113.3 ± 27.5.0 ns n=3	108.7±33.4 ns n=3
ADP (20 µM)	56.75±15.8 ns n=4	51.0 ±17.2 ns n=4	166.3±23.6 p<0.05 n=4	151.8±18.7 ns n=4
A23187 (2.5 µM)	6.3 ± 0.9 p<0.01 n=3	2.7 ± 0.7 p<0.01 n=3	109.7±15.7 ns n=3	82.7±7.9 ns n=3
Arachidonic acid (125 µg/mL)	1.0 ± 0.0 p<0.001 n=3	1.0 ± 0.0 p<0.001 n=3	212.3±19.9 p<0.01 n=3	190.7±11.6 p<0.05 n=3
U46619 (1 µM)	25.2 ± 1.7 p<0.001 n=6	21.3 ±1.6 p<0.001 n=6	231±20.2 p<0.001 n=6	182.7±30.7 p<0.05 n=6

Arachidonic acid metabolite levels are higher in platelets from apoE null mice than in platelets from wild-type mice

Unstimulated platelets from apoE null mice contained significantly more 12-HETE, TXB₂, PGD₂ and PGE₂ than unstimulated platelets from wild-type C57BL/6 mice (Table 3).

Table 3.

Untreated (i.e. no injections) wild-type C57BL/6 and apoE null mice were bled and blood was pooled from 8–10 mice for each experiment. Platelet rich plasma (PRP) was prepared as described in Methods and the levels of 12-HETE, TXB₂, PGD₂, PGE₂ in the platelets were determined as described in Supplemental Methods. The data shown are Mean ± SEM and are expressed as nanograms (ng) per 3 × 10⁸ platelets. The p values indicate the level of significance for the difference between the mean values in the apoE null platelets compared to the C57BL/6 platelets. (n = number of experiments).

C57BL/6 Platelets (Unstimulated) ng/3×108 platelets				ApoE null Platelets (Unstimulated) ng/3×108 platelets
12-HETE	TXB2	PGD2	PGE2	12-HETE	TXB2	PGD2	PGE2
512.00 ±4.04 n=3	1.327 ±0.07 n=3	0.017 ±0.009 n=3	0.245 ±0.015 n=3	2050.8a ±525.6 p = 0.0327 n=6	5.672a ±1.256 p = 0.0182 n=6	0.245a ±0.087 p = 0.0479 n=6	0.738a ±0.172 p = 0.0356 n=6

^asignificantly different from unstimulated C57BL/6 platelets

L-4F treatment results in prolonged bleeding time in apoE null mice but not in wild-type mice

Treatment of apoE null mice with L-4F resulted in a significant (p<0.0001) increase in the bleeding time compared to vehicle treated mice (L-4F treated apoE null mice: 3.76±0.73 minutes vs. vehicle treated apoE null mice: 1.68±0.25 minutes). Addition of oral aspirin significantly (p<0.01) prolonged the bleeding time of vehicle treated apoE null mice from 1.68±0.25 minutes to 4.65±0.65 minutes. However, in L-4F treated apoE null mice the increase in bleeding time after L-4F treatment alone (3.76±0.73 minutes) was not significantly increased by addition of aspirin (4.14±0.48 minutes). Wild-type C57BL/6 mice did not show a significant increase in bleeding time with L-4F treatment (2.06±0.24 minutes) compared to vehicle alone treatment (1.84±0.21 minutes). However the bleeding time in response to aspirin in wild-type mice increased significantly (p<0.01) to 3.93±0.55 minutes. Untreated (i.e. no injections) wild-type C57BL/6 mice had a slightly but significantly longer bleeding time (2.10±0.21 minutes) than untreated apoE null mice (1.62± 0.10 minutes) (p=0.049).

DISCUSSION

In the present study, we investigated the effects of the apoA-I mimetic peptide L-4F on platelet aggregation in wild-type mice, LDLR null and in apoE null mice on the same genetic background. Since no information regarding the effects of apoA-I mimetic peptides on platelet function was available, we chose to use several physiologically active agonists (collagen, ADP, AA, A23187 and U46619) with distinct platelet activation and aggregation signaling pathways.

Administration of L-4F in vivo significantly reduced ex vivo the percent of aggregating platelets (Figure 1), accompanied by the expected changes in slope and lag time (data not shown) in response to increasing concentrations of ADP, AA or collagen compared to vehicle treated mice. L-4F treatment in vitro of platelet rich plasma produced similar results but incubation of washed platelets with L-4F was without effect, suggesting that L-4F acts on plasma components that modulate platelet function.

The plasma half life of L-4F after a single subcutaneous injection in mice is on the order of only 1 hour²⁶. The fact that a dose of 0.01 mg/kg of L-4F administered by a single subcutaneous injection was not effective (Figure 1) but the same dose administered by continuous infusion by Alzet pumps for two weeks was effective (data not shown) suggests that the time during which the peptide is present in plasma may be more important than the peak plasma concentration.

TXA₂, one of the major metabolites of AA and a potent endogenous platelet agonist and vasoconstrictor, plays an important role in platelet-rich thrombus formation^{27, 28}. TXA₂ is produced in platelets from AA via the COX – TXA₂ synthase pathway²⁹. The stable hydrolysate of TXA₂, TXB₂, was significantly reduced in agonist stimulated platelets obtained from L-4F treated mice compared to vehicle treated mice. The inhibition of TXB₂ formation paralleled the effects observed for platelet aggregation suggesting that the inhibition of the COX- TXA₂ synthase pathway was at least in part responsible for the reduction in platelet aggregation.

Formation of TXB₂, PGD₂ PGE₂ and 12-HETE was significantly inhibited in platelets obtained from L-4F injected mice and stimulated with collagen, A23187, ADP or AA but not with U46619. U46619 addition in the absence of COX-1 inhibition, continues the cycle of TXA₂ receptor activation - AA liberation -TXA₂ formation, generating an excess of TXA₂ (Table 1). Inhibition of COX-1 further reduced TXB₂, (Table 2) PDG₂ and PGE₂ formation (data not shown) and significantly increased 12-HETE accumulation (Table 2), confirming that AA utilized for both 12-HETE and TXB₂ biosynthesis is derived from a single phospholipid pool that can be shunted from one pathway to the other²⁵. In addition, the use of U46619 in the presence of COX-1 blockade also demonstrated that L-4F lacks any inhibitory effect on TXA₂ synthase activity or on the TXA₂ receptor (Table 2).

The decreased platelet aggregation and TXB₂ formation following treatment with L-4F was associated with an increased bleeding time in the apoE null mice but not in the wild type C57BL/6 mice. While the difference in bleeding time between the two treatment groups was statistically significant, no spontaneous bleeding was observed in the L-4F injected mice and no additive or synergistic effects were observed between aspirin and L-4F. The small differences in the bleeding times in vehicle treated mice versus untreated mice may indicate a slight effect of the vehicle.

Our results indicate that L-4F decreases the hypersensitivity of platelets to agonist stimulation in hyperlipidemic mice but not in normolipidemic C57BL/6 mice. L-4F does not have a direct effect on platelets but appears to work through a plasma component as noted above. Forte et al.¹³ reported that the type of oxidized lipids that have been shown to have a particularly high affinitity for the 4F peptide¹⁰ are significantly increased in the plasma of apoE null mice compared to wild-type mice. Our results are consistent with L-4F binding and removing these oxidized lipids from plasma in hyperlipidemic mice resulting in altered platelet function. The precise identity and mechanism(s) by which such oxidized lipids influence platelet function will require further studies.

Abbreviations

ADP

adenosine diphosphate

AA

archidonic acid

A23187

calcium ionophore

U46619

thromboxane A₂ mimetic

COX

cyclooxygenase

TX

thromboxane

PG

prostaglandin

12-LO

12-lypoxygenase

12 –HETE

12-hydroxy 5,8,10, 14-eicosatetraenoic acid