Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Eosinophilic infiltration is a major characteristic of allergic airway diseases, and appears to play a key role in the development of asthmatic airway diseases. Eosinophils contain several granules associated with cationic polypeptides such as major basic protein (MBP) and eosinophilic cationic peptide (ECP), and these charged peptides are released upon activation of the eosinophils (1, 2). There is substantial experimental evidence suggesting that eosinophils and particularly their cationic proteins play an essential role in the pathogenesis of airway hyperresponsiveness in bronchial asthma (3, 4). Thus, intratracheal instillation of MBP or ECP can cause airway hyperresponsiveness and these effects can be mimicked by the synthetic cationic proteins poly-L-arginine and poly-L-lysine, suggesting that charge interactions might be important (5).

On the other hand, it has been shown that inhibition of L-arginine-dependent nitric oxide (NO) synthesis can cause airway hyperresponsiveness (8), and there is evidence that deficiency in the L-arginine/NO pathway could be one mechanism responsible for the development of airway hyperresponsiveness under pathologic conditions (9, 10).

In this context it should be noted that the intracellular availability of L-arginine can be a limiting factor for the synthesis of NO (11). One important factor determining the intracellular concentration of L-arginine is the cellular uptake of L-arginine, which is mediated via specific cationic amino-acid transport systems (12, 13). It appeared to be an interesting idea that eosinophil-derived cationic proteins could interfere with the cationic amino-acid transporter, thereby reducing the cellular availability of L-arginine, and as a consequence inhibiting L-arginine-dependent mediator synthesis, such as NO.

In the present study the effect of cationic proteins on the uptake of L-arginine in alveolar macrophages (AM) and tracheal epithelial cells was studied, because these cells can express an inducible form of NO synthase (iNOS) (e.g., 14-17). This enzyme is particularly dependent on extracellular L-arginine supply (11, 18) and, among others, appears to play a role in the control of local immune reactions (e.g., 19-21). Because biologic effects of MBP could be mimicked by synthetic cationic proteins such as poly-L-arginine and poly-L-lysine (6, 7), and because on the other hand, the available quantities of eosinophil-derived cationic proteins are very limited, most of the present experiments were carried out using synthetic cationic proteins.

A preliminary account of the present study was given at the Winter Meeting of the British Pharmacological Society (22).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Preparation and Culture of AM and Tracheal Epithelial Cells

Sprague Dawley rats (own breeding; University of Bonn, Bonn, Germany) of either sex were killed by stunning followed by exsanguination. Lung and trachea were excised en bloc, washed with calcium- and magnesium-free Dulbecco's phosphate-buffered saline (D-PBS) and lavaged three times by instilling about 15 ml of cold (4°C) D-PBS. Usually for one preparation of AM, lavage fluids of four or five lungs were pooled and centrifuged at 400 × g for 5 min. The cells were washed three times with D-PBS and thereafter resuspended in Dulbecco's modified Eagle's medium (DMEM)/Ham's F-12 medium supplemented with 5% fetal calf serum (FCS), 100 U/ml penicillin, 100 µg/ml streptomycin, and 5 µg/ml amphotericin B and plated (2.5 × 10⁶ cells/well for turnover studies and 3 × 10⁶ cells/well for uptake studies) in sterile six-well dishes or 0.5 × 10⁶ cells/well for nitrite accumulation studies in 24-well dishes (NUNC, Wiesbaden, Germany). The AM were allowed to adhere for 2 h (37°C, 5% CO₂). After 2 h the medium was renewed to remove nonadherent cells. The adherent cells consisted of more than 95% AM according to morphologic criteria (May Grünwald Giemsa staining). Thereafter, cells were cultured for 20 h in the absence or presence of lipopolysaccharides (LPS; 1 µg/ml) followed by L-arginine uptake, turnover, or nitrite accumulation studies. Cell viability assessed by trypan blue exclusion was always greater than 95%.

Tracheal epithelial cells of newborn rats were cultured using the explant technique as described previously (17), with the modification that RPMI-1640 medium with reduced calcium (60 to 80 µmol/liter) was used only during the outgrowth phase to ensure the selective outgrowth of epithelial cells from tracheal explant, but for subsequent passages DMEM/Ham's F-12 medium containing standard calcium concentration was used. The culture media also contained 5% FCS and were supplemented with 20 ng/ml epidermal growth factor, 1 µM hydrocortisone, 10 µg/ml insulin, 5 µg/ml transferrin, 100 U/ml penicillin, and 100 µg/ml streptomycin. In the present study confluent monolayers of the 8th to 12th passage were used to study L-arginine uptake. Confluency was reached usually between 7 and 10 d after dissemination of the cells.

Measurement of [³H]-L-Arginine Uptake and Turnover

In [³H]-L-arginine uptake studies, AM cultured for 20 h or confluent epithelial cultures were incubated at 37°C for 2 min in 1 ml [³H]-L-arginine (37 kBq, 0.1 µM) containing Krebs-N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid (Hepes solution [composition, in mM: NaH₂PO₄ 0.001; NaCl 118.5; KCl 5.57; CaCl₂ 1.25; MgCl₂ 1.2; Na₂-ethylenediaminetetraacetic acid (Na₂EDTA) 0.03; L-(+) ascorbic acid 0.06; Hepes 20.0 (adjusted to pH 7.4 using NaOH); and D-(+)-glucose 11.1]) in the absence or presence of polycations, followed by determination of the cellular radioactivity, as described previously (23, 24). The 2-min incubation period was chosen because [³H]-L-arginine accumulation was linear between 1 and 3 min (23, 24). [³H]-L-arginine uptake was expressed either in absolute terms (dpm/3 × 10⁶ cells) or as the percentage of the uptake observed in controls of the respective cell preparations.

In the turnover studies, AM (2.5 × 10⁶ cells, cultured for 20 h in the absence or presence of 1 µg/ml LPS) were incubated at 37°C for 1 h in 1 ml [³H]-L-arginine (37 kBq, 0.1 µM) containing Krebs-Hepes solution in the absence or presence of cationic proteins. The incubation media were collected and kept at 20°C until analyzed by high-performance liquid chromatography (HPLC) (see below).

Measurement of Nitrite Accumulation

For nitrite accumulation studies, AM, 0.5 × 10⁶ cells, cultured for 20 h in the absence or presence of 1 µg/ml LPS were washed and incubated at 37°C for 6 h in 0.5 ml Krebs-Hepes solution with or without L-arginine (0.1 mM) and additionally poly-L-arginine or poly-L-lysine. The incubation media were collected and kept at 80°C until the accumulated amounts of nitrite were determined.

Nitrite (NO₂) was determined in cell-free supernatants of the 6-h incubation period by a spectrophotometric assay based on the Griess reaction. The amount of 400 µl Griess reagent (sulfanilamide 1%, N[1-naphthyl]ethylenediamine hydrochloride 0.1% dissolved in 2.5% [wt/vol] H₃PO₄) were added to 400 µl incubation medium. After 20 min of incubation at room temperature, absorbance was measured at 540 nm. The nitrite contents given under RESULTS were calculated from a standard curve (NaNO₂) and expressed as nanomoles of nitrite (per 10⁶ cells/6 h).

HPLC Analysis

[³H]-labeled L-citrulline, L-ornithine, and L-arginine present in incubation media were separated as described previously (14) on a reverse-phase column (length, 250 mm; inner diameter, 4.6 mm; prepacked with Shadon ODS-Hypersil, 5 µm), using as mobile phase a 0.1 M sodium phosphate buffer (adjusted to pH 1.8) that contained octane sulfonic acid sodium salt (400 mg/liter), Na₂EDTA (0.3 mM), and methanol (6% vol/vol) with a flow rate of 1 ml/ min and 1.5 ml/min for the initial 30 and the following 90 min, respectively. The HPLC eluate was collected in 1-min fractions into counting vials. The retention time of amino acids was determined by the use of [¹⁴C]-labeled L-citrulline and L-ornithine or [³H]-labeled L-arginine as standards. L-Citrulline, L-ornithine and L-arginine eluted after 7 to 8 min, 12 to 13 min, and about 70 min, respectively. The amounts of radiolabeled amino acids in the media were expressed either in absolute terms (dpm/2.5 × 10⁶ cells) or as the percentage of values observed in controls of the respective cell preparations.

Drugs and Special Chemicals

Amphotericin B, L-arginine HCl, D-PBS, DMEM/Ham's F-12 medium, epidermal growth factor, hydrocortisone, insulin, LPS from Escherichia coli 0127:B8, penicillin-streptomycin solution, poly-L-arginine (MW 5,000-15,000), poly-L-aspartate (MW 5,000-15,000), poly-L-lysine (MW 4,000-15,000), pronase, protamine, and transferrin were all from Sigma, München, Germany. L-[2,3-³H]-arginine HCl (36 to 40 Ci/mmol) was from Sigma or Dupont, Dreieich, Germany. Calcium-free RPMI-1640 medium was from Serva, and FCS from Vitromex, Germany. MBP was purified from eosinophils obtained from patients with marked eosinophilia as previously described (25). In brief, eosinophils were lysed with sucrose and heparin and granules were isolated by centrifugation. Granules were lysed by exposure to 0.01 M HCl, pH 2, and by sonication. Granule extracts were applied to a 1.2 × 47 cm Sephadex G-50 column that had been equilibrated with 0.025 M sodium acetate and 0.15 M NaCl, pH 4.3. The fractions from the third peak were pooled as MBP, stored at 70°C, and thawed immediately before use. MBP was pure as judged by the banding pattern on sodium dodecyl sulfate polyacrylamide gel electrophoresis after staining with Coomassie brilliant blue R.

Statistics

All values are means ± standard error of the mean (SEM) of n experiments. Statistical significance of differences was evaluated by Student's t test. When multiple comparisons were performed, the significance of differences was evaluated by analysis of variance followed by the modified t test according to Bonferroni using the computer program GraphPad InStat (GraphPad Software, San Diego, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Cationic Proteins on [³H]-L-Arginine Uptake in AM

Under control conditions, rat AM incubated for 2 min with [³H]-L-arginine (37 kBq/ml, 0.1 µM) accumulated 20,238 ± 2,521 dpm/(3 × 10⁶ cells), corresponding to 1.30 ± 0.16 pmol L-arginine/(3 × 10⁶ cells) (n = 22). The different cationic proteins, poly-L-arginine, poly-L-lysine, protamine, and MBP, present in the incubation medium (i.e., during the uptake period) inhibited [³H]-L-arginine uptake in a concentration-dependent manner (Figure 1, Table 1). Poly-L-arginine was more potent than the other cationic proteins and caused an inhibition by about 80% at the highest concentration tested (300 µg/ml). The inhibition produced by 300 µg/ml of the other cationic proteins tested was between 40 and 50% for poly-L-lysine and protamine and 35% for MBP.

View larger version (15K): [in this window] [in a new window]   Figure 1.   Effects of poly-L-arginine on [3H]-L-arginine uptake in rat AM. AM (3 × 106 cells/well), after 20 h culture, were incubated for 2 min in amino acid-free Krebs-Hepes solution to which [3H]-L-arginine (37 kBq, 0.1 µM) and poly-L-arginine (at the concentrations indicated by the abscissa) alone or in combination with heparin was added. Ordinate: cellular radioactivity expressed as percentage of the mean value observed in controls of the respective cell preparations (for absolute values, see RESULTS), means ± SEM of six to nine experiments. Significance of differences: *P < 0.05; **P < 0.01 from the respective controls; +P < 0.01 from the corresponding value in the absence of heparin.

Heparin (100 µg/ml) present together with the cationic proteins blocked the inhibitory effect of poly-L-arginine (30 and 100 µg/ml, Figure 1) as well as that of poly-L-lysine (30 and 100 µg/ml, data not shown, each n = 6). Heparin (100 µg/ml) alone did not affect [³H]-L-arginine uptake in rat AM (n = 6, data not shown). Likewise, poly-L-aspartate (300 µg/ml), as an example of anionic protein, did not affect [³H]-L-arginine uptake in rat AM (n = 6, data not shown).

After 20 h culture in the presence of LPS (1 µg/ml) [³H]-L-arginine uptake in rat AM was enhanced to 70,031 ± 10,807 dpm/(3 × 10⁶ cells) (n = 6). Poly-L-arginine (100 and 300 µg/ml) reduced [³H]-L-arginine uptake to 38 ± 4 and 21 ± 3%, respectively (each n = 6); i.e., poly-L-arginine caused similar effects as those in AM cultured in the absence of LPS (see Table 1). On the other hand, poly-L-lysine (100 and 300 µg/ml) reduced [³H]-L-arginine uptake only to 91 ± 10 and 74 ± 5%, respectively (each n = 6); i.e., the inhibitory potency of poly-L-lysine appeared to be reduced after LPS treatment (compared with AM cultured in the absence of LPS; see Table 1).

Effects of Cationic Proteins on [³H]-L-Arginine Uptake in Tracheal Epithelial Cells

Because the characteristics of [³H]-L-arginine uptake in tracheal epithelial cells (in contrast to AM) have not been described in previous studies, some basic observations of this uptake are described first. During 2 min of incubation with [³H]-L-arginine under control conditions, confluent rat tracheal epithelial cells accumulated 34,180 ± 5,058 dpm/well (n = 29), corresponding to 2.20 ± 0.32 nmol L-arginine. The accumulation of [³H]-L-arginine showed a linear time dependency between 1 and 5 min and reached a plateau between 10 and 20 min, which was about 3- to 4-fold higher than the 2-min value (Figure 2A). The uptake of [³H]-L-arginine was not affected when extracellular sodium ions were replaced by equimolar choline and was inhibited by about 95% by 1 mM L-lysine, but was not affected by the neutral amino acid L-leucine (0.1 and 1 mM) (Figure 2B). Thus, the transport of L-arginine uptake showed typical characteristics of the cationic amino-acid transport system y⁺ (see Reference 12).

View larger version (32K): [in this window] [in a new window]   Figure 2.   Characteristics of [3H]-L-arginine uptake in rat tracheal epithelial cells in culture. Confluent monolayer of tracheal epithelial cells (six-well dishes) were incubated for 1 to 20 min (A, as indicated by the abscissa) or 2 min (B) in amino acid-free Krebs-Hepes solution to which was added [3H]-L-arginine (37 kBq, 0.1 µM) alone or in combination with L-lysine (Lys) or L-leucine (Leu) at the concentrations indicated. In some experiments, sodium in the Krebs medium was replaced by equimolar concentrations of choline (Na+, 0). Given is the cellular radioactivity, expressed as percentage of the mean value observed in controls (2 min incubation) of the respective cell preparations (for absolute values, see RESULTS), means ± SEM of six to 12 experiments. Significance of differences: *P < 0.01 from the respective controls.

As in AM, poly-L-arginine showed marked inhibitory effects on [³H]-L-arginine uptake in rat tracheal epithelial cells; at the highest concentration tested (300 µg/ml), poly- L-arginine inhibited [³H]-L-arginine uptake by 65% (Figure 3). Again, this effect of poly-L-arginine was prevented by 100 µg/ml heparin (Figure 3), which alone had no effect (data not shown). As in AM, poly-L-lysine was less effective than poly-L-arginine in inhibiting [³H]-L-arginine uptake, and the inhibition seen at the maximum concentration (300 µg/ml) tested was about 40% (data not shown).

View larger version (15K): [in this window] [in a new window]   Figure 3.   Effects of poly-L-arginine on [3H]-L-arginine uptake in rat tracheal epithelial cells in culture. Confluent monolayer of tracheal epithelial cells (six-well dishes) were incubated for 2 min in amino acid-free Krebs-Hepes solution to which [3H]-L-arginine (37 kBq, 0.1 µM) and poly-L-arginine (at the concentrations indicated by the abscissa) alone or in combination with heparin was added. Ordinate: cellular radioactivity expressed as percentage of the mean value observed in controls of the respective cell preparations (for absolute values, see RESULTS), means ± SEM of six to nine experiments. Significance of differences: **P < 0.01 from the respective controls; +P < 0.01 from the corresponding value in the absence of heparin.

Effects of Polycations on Nitrite Formation by Rat AM

The effect of poly-L-arginine and poly-L-lysine on the accumulation of nitrite (as a measure of NO synthesis) was studied. When rat AM, after culture in the absence of LPS, were incubated for 6 h in amino acid-free Krebs solution, only a small amount of nitrite (1.37 ± 0.14 nmol) accumulated in the medium and addition of 100 µM L-arginine caused an increase by 70% (Figure 4). When LPS-stimulated AM were incubated, the nitrite accumulation in the absence of L-arginine was 3.48 ± 0.29 nmol and addition of L-arginine (100 µM) increased nitrite accumulation about 4-fold (Figure 4). When poly-L-arginine (100 and 300 µg/ml) was added, nitrite accumulation in the absence of L-arginine was reduced by about 40 to 50% both in control and LPS-treated AM (Figures 5A and 5C). In the presence of L-arginine (100 µM) the inhibitory effect of poly-L-arginine on nitrite accumulation was more pronounced; an inhibition by about 60 and 85% was caused by 100 µg/ml poly-L-arginine in control and LPS-treated AM, respectively (Figures 5B and 5D). Notably, 30 µg/ ml poly-L-arginine, which caused an inhibition of [³H]-L-arginine uptake by about 50% (Table 1), inhibited nitrite accumulation in L-arginine-containing medium by about 30% in control AM but by about 80% in LPS-treated AM (Figures 5B and 5D). On the other hand, poly-L-lysine had no clear effect on nitrite accumulation in the absence of L-arginine, both in control and LPS-treated AM (Figures 5A and 5C), but in the presence of L-arginine (100 µM) we observed a significant inhibition by about 30 and 20% in control and LPS-treated AM, respectively (Figures 5B and 5D).

View larger version (12K): [in this window] [in a new window]   Figure 4.   L-Arginine dependence of the nitrite accumulation by rat AM. AM (0.5 × 106 cells/well), after 20 h culture in the absence or presence of LPS (1 µg/ml), were incubated for 6 h in amino acid-free Krebs-Hepes solution to which L-arginine (100 µM, as indicated) was added. Height of columns: nitrite accumulated in the incubation medium, expressed as nmol/106 cells/6 h, means + SEM of six experiments. *P < 0.01 versus respective controls (Ctr = not treated with LPS); +P < 0.01 versus respective value of L-arginine-free (L-Arg-free) medium.

View larger version (40K): [in this window] [in a new window]   Figure 5.   Effects of poly-L-arginine (p-L-Arg) and poly-L-lysine (p-L-Lys) on nitrite accumulation by rat AM. AM (0.5 × 106 cells/well), after 20 h culture in the absence (A and B) or presence (C and D) of LPS (1 µg/ml), were incubated for 6 h in amino acid-free Krebs-Hepes solution to which L-arginine (100 µM) and/or the cationic proteins (as indicated) were added. Height of columns: nitrite accumulated in the incubation medium, expressed as percentage of the mean value observed in the respective controls of the respective cell preparations (note the different absolute values for the different control groups, given in Figure 4), means + SEM of six experiments. *P < 0.05; **P < 0.01 versus respective controls (Ctr); +P < 0.05; ++P < 0.01 versus respective value of L-arginine-free medium. Open columns, controls; solid columns, poly-L-arginine added (amounts indicated); hatched columns, poly-L-lysine added (amounts indicated).

Effects of Cationic Proteins on [³H]-L-Arginine Metabolism by NOS and Arginase in Rat AM

As already observed in previous studies (14, 23), when rat AM that had been cultured for 20 h under control conditions were incubated for 1 h in Krebs medium containing [³H]-L-arginine (37 kBq/ml, 0.1 µM), a significant accumulation of [³H]-L-citrulline and [³H]-L-ornithine occurred in the incubation medium (Figures 6 and 7). The amount of radioactivity found in the [³H]-L-citrulline and [³H]-L-ornithine fractions corresponded to 6.7 ± 1.1 and 9.8 ± 0.7% of the total radioactivity of the incubation media, respectively. Poly-L-arginine present during the 1-h incubation period inhibited in a concentration-dependent manner the formation of [³H]-L-citrulline and [³H]-L-ornithine. However, the effects on [³H]-L-citrulline formation were more pronounced than those on [³H]-L-ornithine formation (Figure 6). When the AM were cultured in the presence of LPS, accumulation of [³H]-L-citrulline increased about 7-fold and this was accompanied by 80% reduction in [³H]-L-ornithine formation (Figure 7). Under these conditions, poly-L-arginine again reduced in a concentration- dependent manner the formation of [³H]-L-citrulline even below the level of untreated AM (not exposed to LPS) (Figure 7). This was accompanied by an increase in [³H]-L-ornithine formation, which, however, did not reach the level observed in untreated AM.

View larger version (21K): [in this window] [in a new window]   Figure 6.   Effects of poly-L-arginine on [3H]-L-arginine metabolism in rat AM by NO synthase and arginase. AM (2.5 × 106 cells/well), after 20 h culture, were incubated for 1 h in amino acid-free Krebs-Hepes solution to which [3H]-L-arginine (37 kBq, 0.1 µM) alone or in combination with poly-L-arginine (at the concentrations indicated below the columns) was added. Thereafter, [3H]-L-citrulline and [3H]-L-ornithine accumulated in the incubation medium were determined by HPLC separation. Height of columns: [3H]-amino acids expressed in absolute terms (dpm × 1,000/2.5 × 106 cells/1 h), means + SEM of six to nine experiments. *P < 0.01 versus respective values in the absence of poly- L-arginine.

View larger version (20K): [in this window] [in a new window]   Figure 7.   Effects of poly-L-arginine on [3H]-L-arginine metabolism by NO synthase and arginase in LPS-stimulated rat AM. AM (2.5 × 106 cells/well), after 20 h culture in the absence or presence of LPS (1 µg/ml), were incubated for 1 h in amino acid- free Krebs-Hepes solution to which [3H]-L-arginine (37 kBq, 0.1 µM) alone or in combination with poly-L-arginine (at the concentrations indicated below the columns) was added. Thereafter, [3H]-L-citrulline and [3H]-L-ornithine accumulated in the incubation medium were determined by HPLC separation. Height of columns: [3H]-amino acids expressed in absolute terms (dpm × 1,000/2.5 × 106 cells/1 h), means + SEM of six to nine experiments. +P < 0.01 versus respective value in the absence of poly-L-arginine of LPS-treated cells; *P < 0.001 versus respective value in cells not treated with LPS.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

It has been shown that cationic proteins, both the naturally occuring eosinophil-derived MBP and ECP as well as the synthetic peptides poly-L-arginine and poly-L-lysine, can cause airway hyperresponsiveness (5). In addition, cationic proteins can exert cytotoxic effect to airway epithelial cells (26, 27) and cause epithelial changes similar to those observed in bronchial asthma (28). Therefore, it has been suggested that epithelial damage caused by these charged proteins may be a major mechanism responsible for the development of airway hyperresponsiveness in asthma. On the other hand, there is increasing evidence that cationic proteins can also alter airway functions through "non-cytotoxic" mechanisms. Thus, hyperresponsiveness after exposure to cationic proteins is observed already within 1 h, i.e., long before any morphologic damage to the epithelium is observed, and there is additional evidence suggesting that cationic proteins may affect the release of epithelium-derived factors that modulate the responsiveness of airway smooth muscle (29). In addition, cationic proteins appear to activate sensory nerves (32), to liberate bradykinin (33), and to enhance parasympathetic neurotransmission (34).

The present experiments show that cationic proteins can inhibit the cellular uptake of L-arginine, and this effect was observed in AM and tracheal epithelial cells (i.e., in two different cell types present in the airway mucosa). Further, it may be mentioned that effects similar to those described here for rat AM and tracheal epithelial cells were also observed in AM and tracheal epithelial cells of the guinea pig (22, and data not shown), suggesting a general biologic significance of the described effects. Like other biologic effects of cationic proteins, their effects on the uptake of L-arginine were prevented by heparin, indicating that the positive charge is important for the effects on L-arginine transport.

It appears that poly-L-arginine is more potent than poly-L-lysine in inhibiting L-arginine transport. The most detailed experiments in the present studies were carried out on AM, and here it was additionally shown that MBP also reduced L-arginine uptake, as did protamine, an endogenous but commercially available cationic protein that may also be used in in vitro studies to substitute for MBP (e.g., 35). The magnitude of the effects of protamine and MBP were comparable to that of poly-L-lysine but smaller than that of poly-L-arginine. The potency of MBP in the present in vitro experiments might have been underestimated for several reasons. MBP is known to precipitate from saline at physiologic pH, but for the present uptake studies the cells have to be incubated in a physiologic salt solution. In addition, MBP may adhere to the surface of the culture dish. Such an adsorptive loss could have been minimized by pre-exposure of the culture dish to poly-L-lysine, but poly-L-lysine showed to be an effective inhibitor of L-arginine uptake on its own and therefore could not be used for this purpose.

Because poly-L-arginine was a more effective inhibitor of L-arginine uptake than were the other cationic peptides, it may be concluded that, in addition to the positive charge, specific chemical structures (related to L-arginine) appear to enhance the affinity to the transport system and the inhibitory potency. After exposure to LPS, a treatment which causes marked induction of iNOS and as a consequence increases L-arginine demand, L-arginine uptake into AM was enhanced. Most interestingly, this was accompanied by some reduction in the potency of poly-L-lysine, but not of poly-L-arginine, to inhibit L-arginine uptake. Thus, after exposure to LPS the cationic amino-acid transport system in AM appears to develop a higher degree of "L-arginine specificity."

Although in the mammalian organism L-arginine is not an essential amino acid, this amino acid can be synthesized only by some mammalian cells, whereas many other cells in the mammalian organism appear to depend on the L-arginine supply from extracellular sources. Thus, NO synthesis in rat AM, particularly after marked induction of iNOS, largely depends on an exogenous supply of L-arginine (18); this is confirmed in the present study (Figure 4). Tracheal epithelial cells may also depend on an exogenous L-arginine supply. Although we observed that rat tracheal epithelial cells could synthesize some L-arginine from L-citrulline, the turnover rate of this pathway was so small that it may not be a way to supply the cells sufficiently with L-arginine (unpublished observations), particularly not under conditions of enhanced need such as induction of NO synthesis or cell proliferation. In line with this conclusion are recent observations that the activity of the L-arginine transport in proliferating tracheal epithelial cells in culture is much higher than that in confluent cultures (36).

The cellular uptake of L-arginine is mediated via specific cationic amino-acid transporters (12, 13), which, in addition to L-arginine, also transport other cationic amino acids such as L-ornithine and L-lysine, the latter being an essential amino acid. Thus, inhibitory effects on cationic amino-acid transporters may reduce the cellular availability of several cationic amino acids and therefore may have multiple functional consequences. Because L-lysine and L-arginine are involved in protein synthesis, prolonged inhibition of cationic amino-acid transport could result in a disturbed protein synthesis. Further, L-arginine is the substrate in two pathways generating important cellular and transcellular mediators, the NO synthase and the arginase-polyamine pathways. Therefore, L-arginine deficiency may result in a reduced synthesis of NO and of ornithine that serves as a precursor for the synthesis of polyamines (37). In contrast, to effects on protein synthesis, inhibitory actions of cationic proteins on the cationic amino-acid transports may affect these mediator pathways immediately.

Indeed, the present experiments also show that poly-L-arginine and poly-L-lysine can cause an acute reduction of nitrite accumulation by rat AM. Here again, the inhibitory effect of poly-L-arginine was more pronouced than that of poly-L-lysine. As already mentioned, the NO synthesis by rat AM largely depends on the cellular uptake of L-arginine. When rat AM were incubated in amino acid-free Krebs solution, their cellular L-arginine availability appeared to be immediately at a level suboptimal for NO synthesis, particularly when the iNOS had been markedly induced by pre-exposure of the cells to LPS, and addition of L-arginine caused a marked increase in NO synthesis (Figure 4). Poly-L-lysine inhibited nitrite accumulation only in the experiments in which L-arginine was present (Figures 5B and 5D), i.e., when part of the NO synthesis depended directly on the cellular uptake of L-arginine, indicating that the inhibitory effect of poly-L-lysine on L-arginine uptake is of functional consequences for the NO synthesis. On the other side, an inhibitory effect of poly-L-arginine on nitrite accumulation was already observed in the absence of L-arginine, suggesting that this cationic protein may exert additional direct inhibitory effects on NO synthesis. The additional direct inhibitory effects of poly-L-arginine on NO synthesis appear to be of particular significance after marked induction of iNOS. Thus, in AM not exposed to LPS and incubated in L-arginine-containing medium, poly-L-arginine and poly-L-lysine in concentrations producing similar inhibiton of L-arginine uptake (30 and 300 µg/ml, respectively; Table 1) were almost equally effective in reducing nitrite accumulation (Figure 5B); whereas in LPS-treated AM, 30 µg/ml poly- L-arginine caused a substantially larger reduction of nitrite accumulation than did 300 µg/ml poly-L-lysine (Figure 5D).

The significance of a direct inhibition of NO synthesis by poly-L-arginine is further supported by the observations on the metabolism of [³H]-L-arginine (Figures 6 and 7). Rat AM cultured in the absence of LPS and incubated for 1 h in Krebs medium containing [³H]-L-arginine showed a significant formation of [³H]-L-citrulline and [³H]-L-ornithine, as measures of NOS and arginase activity, respectively. Addition of poly-L-arginine to the incubation medium resulted in a concentration-dependent reduction in the formation of [³H]-L-citrulline and [³H]-L-ornithine, the effect on [³H]-L-citrulline being more pronounced than that on [³H]-L-ornithine. Inhibitory effects of poly-L-arginine on L-arginine transport may account for the reduction of [³H]-L-ornithine and [³H]-L-citrulline, but the stronger effect on [³H]-L-citrulline formation may be explained by an additional inhibitory effect on NOS. In agreement with previous observations (14, 23), the formation of [³H]-L-citrulline was greatly enhanced and that of [³H]-L-ornithine reduced when high levels of iNOS had been induced by LPS, i.e., L-arginine was preferentially metabolized by iNOS. There is evidence that N^G-OH-L-arginine, the intermediate in the NO synthesis reaction, is released by iNOS and inhibits arginase (38). Under these conditions poly-L-arginine again caused a marked reduction in the formation of [³H]-L-citrulline, but on the contrary enhanced that of [³H]-L-ornithine, although [³H]-L-ornithine remained below the level observed in control AM. Because inhibition of high iNOS activity can favor L-arginine metabolism by arginase (14), an inhibitory effect of poly-L-arginine on iNOS is likely to be responsible for the increase of [³H]-L-ornithine in these experiments, but the additional inhibitory effect of poly-L-arginine on L-arginine transport keeps [³H]-L-ornithine formation below the level of control AM.

In conclusion, the present experiments show that different cationic proteins including MBP can exert inhibitory effects on the cellular transport of L-arginine, and this can result in a reduced substrate supply for L-arginine-dependent pathways such as NO synthesis, as shown in the present paper for AM. It appears likely that similar effects are of significance in other cells in which NO synthesis also depends on the uptake of L-arginine. There is evidence that NO synthesis in nitrergic neurons depends on the acute supply of L-arginine (39, 40), and in the airways nitrigic nerves are an important inhibitory pathway (see Reference 41). The different potencies of poly-L-lysine and poly-L-arginine suggest that besides the charge of these proteins, specific structures might also be of importance for the biologic effects of cationic proteins. Further, poly-L-arginine, but not poly-L-lysine, appears to exert additional direct inhibitory effects on NOS.