Blood and serum from healthy donors were purchased from the Blood Bank. Human serum was used heat-untreated, providing an intact complement. Leighton tubes with human PMNs were purchased from Jupiter Ltd. (Ski, Norway). The PMNs were prepared according to the technique of Bøyum [25-27] using Polymorphprep^TM from Axis-Shield PoC AS (Oslo, Norway). The recovered granulocytes were washed once in Krebs Ringer Phosphate buffer enriched with 10 nM glucose and resuspended in Krebs Ringer Phosphate buffer enriched with 10 nM glucose to a final concentration of 2.5 × 10⁶ cells/mL. Aliquots of 1 mL of these suspensions, which were of at least 95% purity, were added to Leighton tubes (16 × 125 mm, Bellco Glass Inc., Vineland, NJ, USA). The tubes were placed in a horizontal position at 37˚C for 60 min. A monolayer of neutrophils with approximately 5,000 cells per square mm was established during that time.

Serum-sensitive E. coli (ATCC 11775) grown under standardized conditions to a density of 10⁹ CFU/mL, in a medium prepared according to Benacerraf et al. [28], were supplied with 1 mCi of ³²P-labeled orthophosphate from BRITATOM (Mumbai, India) per 200 mL of medium. Bacteria were incubated in a gyratory shaker for 200 min at 37˚C. The bacteria were washed twice in ice-cold Krebs Ringer Phosphate buffer enriched with 10 nM glucose to stop growth and remove surplus radioactivity, and a stock suspension of 10¹⁰ CFU/mL Krebs Ringer Phosphate buffer enriched with 10 nM glucose was prepared. This suspension was kept on ice until further use within 30 min (but not longer). At the start of opsonization or phagocytosis, the bacteria were diluted to 10⁹ CFU/mL.

Antibodies were raised in rabbits by intravenous injection of a heat-treated suspension of serum-sensitive E. coli twice weekly for 12 weeks. The IgG fraction of the serum was purified by batch adsorption on DEAESephadex A-50 (Pharmacia, Sweden) and then precipitated with ammonium sulfate. The precipitate was dissolved in phosphate-buffered saline and desalted on an Amersham Biosciences PD-10 desalting column (Amersham Biosciences AB, Uppsala, Sweden). The final preparation agglutinated the bacteria at a titer of 256. When a 10-fold dilution of the preparation was tested by gel diffusion against sheep anti-rabbit whole serum and sheep anti-rabbit IgG, only one precipitation line was found. This indicated that most of the protein in the preparation was IgG.

Opsonization of E. coli with rabbit anti-E. coli IgG was carried out with a suspension of 10⁹ bacteria/mL in a water bath shaker at 37˚C for 30 min. A fourfold dilution of the agglutinating titer of IgG was used. The bacteria were then washed in ice-cold Krebs Ringer Phosphate buffer enriched with 10 nM glucose, resuspended to 10⁹ bacteria/mL and prewarmed at 37˚C for 5 min before phagocytosis.

Opsonization with serum was performed by the addition of serum to a final concentration of 10% and 10⁹ bacteria/mL. The serum used was not heat-inactivated. This suspension was prewarmed in a water bath at 37˚C for 5 min before phagocytosis was started.

At the end of the preincubation period, the PMN monolayer was washed once with Krebs Ringer Phosphate buffer enriched with 10 nM glucose (37˚C), and the uptake was started by adding 1 mL aliquots of the suspension of non-opsonized or opsonized E. coli, with or without hyperforin, as indicated below. The tubes were placed in a horizontal position during the uptake phase, which was stopped after 30 min by washing the PMNs four times with ice-cold Krebs Ringer Phosphate buffer enriched with 10 nM glucose. For the study of the elimination of bacterial breakdown products from the PMNs, the cells were washed with prewarmed Krebs Ringer Phosphate buffer enriched with 10 nM glucose (37˚C) and incubated for another 60 min at 37˚C. The elimination was then stopped by washing the cells with ice-cold Krebs Ringer Phosphate buffer enriched with 10 nM glucose.

Hyperforin was extracted from dried material of H. perforatum purchased from Norsk Medisinaldepot (NMD, Oslo, Norway) and purified as previously described [10,13,14]. A stock solution of 50 mg/mL was stored at 4˚C in 96% ethanol, sheltered from light. Immediately before use, the solution was diluted to a final concentration of 1, 10 or 100 μg/mL in physiological solution. The drug, bacteria, opsonins and PMNs were combined in the following ways:

1) The bacteria were preincubated with hyperforin for 30 min at 37˚C and then washed twice with ice-cold Krebs Ringer Phosphate buffer enriched with 10 nM glucose before the start of phagocytosis. In the case of IgG-opsonized bacteria, opsonization and preincubation were performed simultaneously. Phagocytosis took place in the absence of hyperforin.

2) The PMN monolayer was preincubated with the drug for 30 min before the start of phagocytosis, which took place in the absence of hyperforin.

3) The PMNs were preincubated with hyperforin for 30 min and the bacteria for 5 min, after which the uptake was performed in the presence of hyperforin.

After the final washing, the tubes were dried, and 1 mL of Lowry’s alkaline copper solution was added to the PMNs to dissolve the cells overnight at room temperature. Cell protein was then determined according to the method of Oyama et al. [29], and radioactivity was counted in a Beckman LS 3801 Scintillation Counter liquid scintillation counter at 4˚C (Lab Extreme, Inc., Kent City, MI, USA). The uptake of E. coli by the PMNs was expressed as counts per mg cell protein per min. The elimination was expressed in the same way by subtracting the values found after elimination from those found at the end of the uptake period. The medium covering the PMNs during the elimination phase was also assayed for radioactivity before and after centrifugation at 10,000 g for 10 min, to discriminate between particle-bound and soluble radioactivity. Detachment of PMNs from the monolayer could thus be assessed.

All of the experiments were performed with 3 - 5 parallel runs and repeated at least five times. The uptake of E. coli in PMNs, as influenced by hyperforin, was expressed as a percentage of the control. The controls were parallel experiments, but without the antibiotic hyperforin. The elimination was expressed as the percentage of ingested radioactivity that was released from the PMNs during the elimination period both after control uptake and after uptake influenced by hyperforin. Student’s t test was used for statistical analysis and a P value < 0.05 was considered statistically significant.

When phagocytosis took place in the presence of 10% serum, the uptake of E. coli increased by a factor of 23.0 ± 3.0 (mean ± SD, n = 41), compared with uptake of non-opsonized bacteria.

Correspondingly, opsonization by anti-E. coli IgG resulted in an enhancement of the uptake by 35.8 ± 15.7 (mean ± SD, n = 35).

The uptake of E. coli pretreated with hyperforin is outlined in Table 1. With non-opsonized bacteria, a stimulation of uptake was observed with 100 μg/mL of hyperforin, whereas the uptake of serum-opsonized and IgGopsonized E. coli was inhibited by 10 and 100 μg/mL of hyperforin.

After preincubation of the PMNs, but not bacteria, with hyperforin, a similar pattern was seen (Table 2). However, the uptake of non-opsonized E. coli was more efficiently stimulated by 100 μg/mL of hyperforin after preincubation of PMNs than after preincubation of bacteria. Regarding serum-opsonized bacteria, the inhibiting effect was of the same magnitude as that seen after PMN preincubation with hyperforin; the inhibiting effect on the uptake of IgG-opsonized bacteria was less pronounced than after preincubation of the bacteria. Only a slight but significant inhibition was seen with 10 μg/mL.

Table 3 shows that preincubation of the PMNs for 30 min and the bacteria for 5 min followed by uptake in the presence of hyperforin resulted in even more pronounced effects. With 100 μg/mL, the uptake of non-opsonized E. coli increased almost 50-fold and the uptake of IgG-opsonized E. coli increased more than threefold. The uptake of serum-opsonized bacteria, however, was reduced by approximately 40% with 100 μg/mL of hyperforin.