Suppression of monocyte and neutrophil function by recombinant IL-2

Little IS known about the influence of IL-2 on phagocytes. We now describe the effects of human recombinant IL-2 on human neutrophil and monocyte functions related to mobility, phagocytosis, glucose uptake, respiration and degranulation. Neutrophil adherence and hexose monophosphate shunt activities were both suppressed after incubation with IL-2. IL-2 had no effect on neutrophil migration, phagocytosis, deoxyglucose uptake or degranulation, ionocytes demonstrated a greater sensitivity to IL-2 with suppression of monocyte adherence, random and stimulated migration, glucose uptake and hexose monophosphate shunt activity, even after addition of phorbol myristate acetate. Monocyte phagocytosis and degranulation were not affected. All of the effects observed were dose-dependent within a biologically active range for IL-2. These studies suggest that IL-2 may have an important down-regulatory role across a broad range of monocyte functions including movement, deoxyglucose uptake and respiration. However, its role in regulation of neutrophil function is limited to adherence and respiration. IL-2 may be a more versatile cytokine than has previously been appreciated.


Introduction
Interleukin-2 (IL-2) is a glycoprotein cytokine with a broad range of immunoregulatory functions. It has been shown to play a role in the differentiation and proliferation of both Band T-lymphocytes by upor down-regulation of various subsets; [1][2][3][4] it can directly augment the cytotoxic activity of natural killer cells and large granular lymphocytes, 5-7 but may also indirectly suppress this function, IL-2 has also been implicated in monocyte/macrophage regulation as these cells have been shown to bind IL-2 via specific membrane receptors [9][10][11] and it has been shown to enhance monocyte cytotoxic activity 2 and induce macrophage resistance to infection. 13 The most widely studied of the membrane receptors through which the IL-2 molecule mediates its effects is the Tac protein, which is expressed on primed or preactivated cells. 13 There are, however, a number of other cell membrane proteins on the surfaces of resting and stimulated cells, which bind or are phosphorylated by IL-2. [4][5][6] While resting monocytes are devoid of the Tac receptor protein, expression has been shown to occur following culture with a number of classical stimuli including lipopolysaccharide and interferon-2.9 Addition of IL-2 to such stimulated ceils causes an augmentation of microbicidal activity. 9'1 '12 Polymorphonuclear leukocytes have not been demonstrated to display the Tac protein and, consequently, possible modulation of their functions by IL-2 has not been as extensively investigated. Furthermore, neutrophils have only recently been recognized as target cells for immunoregulatory cytokines. 7 '8 In the light of evidence that IL-2 modulates the activities of various cell types via pathways other than the Tac protein receptor, 3'15'6 we set out to compare the effect of IL-2 on two immunocyte subsets classically devoid of Tac protein expression: the human peripheral blood mononuclear phagocytes (monocytes) and polymorphonuclear phagocytes (neutrophils).

Methods and Materials
Interleukin-2: Human recombinant IL-2 (AIa125) from an Escherichia coli host was manufactured by Amgen Biologicals, CA, USA and obtained through Amersham International, UK, with a specific activity of 2.6-2.8 103 unit//l. Purity was at least 96% and lipopolysaccharide contamination was less than 0.09pg/ml at the highest concentration of IL-2 used under the final assay conditions. The alanine substitution for cysteine at position 125 provides an increased specific activity of up to 1 x 107 unit/mg protein in the T-cell growth assay and improves stability. Apart from this, the biological properties of this product are essentially indistinguishable from cell culture derived material. One unit of activity is defined as the amount of IL-2 which induces 50% of maximal 3H thymidine incorporation by IL-2 dependent T-cells in culture (Amersham Product Technical Information). A stock IL-2 solution of 1 000 U/ml was prepared in 0.1% bovine serum albumin (BSA; Sigma Chemical Co., St Louis, MO) in Dulbecco's phosphate buffered saline (DPBS; Flow Laboratories, McLean, VA), and stored in aliquots at -70C for up to 3 months. Concentrations of 2, 20 and 200 U/ml were used for these experiments since, classically, 2-50 U/ml is required for the maintenance of IL-2 dependent T-cell lines. Leukocyte preparation: Blood collected by venepuncture from healthy volunteer donors was anticoagulated with heparin. The blood was layered over Mono Poly Resolving Medium (Flow Laboratories) and centrifuged at 400 x g for 30-40min. Neutrophils of 97% purity were removed from the lower of the two leukocyte bands formed at the interface. 19 The upper band, containing mononuclear cells and platelets, was enriched for monocytes using preformed continuous density gradients made from Percoll (Pharmacia, South Seas) as described previously. 2 Briefly, Percoll was aseptically made up to 325 mosM with a density of 1.070 g/cm 3 with phosphate buffered saline and 0.1% BSA; 10 ml of the solution was spun at 21 000 x g in a Beckman Ultracentrifuge (21C, fixed angle rotor) for 20min. The mononuclear cell suspension, depleted of platelets by centrifugation through autologous plasma, was layered over the Percoll gradient and centrifuged at 1 000 x g for 15 min. The upper of the three bands formed was collected using a hypodermic syringe and needle and routinely yielded 70-90% monocytes. The percentage of monocytes was assessed using 0.01% Methylene Blue vital staining and the accuracy of this method was periodically checked with fluorescein-conjugated MO2 monoclonal antibody (Coulter Electronics, FL, USA) staining.
Leukocytes were washed twice and resuspended in Medium 199 (M199; Commonwealth Serum Laboratories, Australia) or in DPBS as dictated by assay requirements.
Mobility assays: Adherence of leukocytes was measured after passage through a dacron fibre (100/.zg Olympic General Products, Queensland, Australia) microcolumn as described previously. 21 Neutrophils were incubated with IL-2 for 30 min prior to a 5 min contact with the fibre at 37C in a humidified 5% CO2 atmosphere. Monocytes were in contact with the fibre microcolumn for 30 min in the presence of varying concentrations of IL-2.

Adherence was expressed as
Co C x 100% Co where Co is the original cell concentration and Ce is the cell concentration in the effluent. All monocyte suspensions were differentially stained to exclude contaminant adherent lymphocytes from the calculation. The under-agarose migration assay 22 was used to measure random and stimulated migration using autologous serum activated by Candida albicans as the chemoattractant. IL-2 was incorporated into 3 ml M199 and 10% FCS and 2% agarose at the required concentrations. Triplicate wells were punched 2.5 cm apart, once the agarose had set in 60 mm diameter plastic tissue culture plates. The centre well received the cell suspension, the outer well received the chemoattractant (for stimulated migration) and the inner well received M199 (for random migration). Neutrophils were allowed to migrate for 2.5 h at 37C in a humidified CO2 atmosphere while monocytes required 18 h incubation. The distance travelled towards either well was measured with the aid of an eye piece grid on an inverted microscope.
Phagocytosis assay: Phagocytosis was measured by the inhibition of uridine uptake by C. albicans; once phagocytosed, yeast cells do not incorporate uridine. 23 Neutrophils or monocytes were preincubated with IL-2 for 15 min in M199 and 2.5% autologous serum. Live, washed C. albicans cells were added and the mixture incubated for 30 min at 37C with rotational mixing. Triplicate 0.1 ml aliquots of the suspensions were dispersed in a 96-well microtitre tray and 5/ll 14C uridine (Amersham) was added to achieve a final concentration of 0.93/.zCi/ml. This was incubated at 37C for 1 h, centrifuged and measured for yeast uridine incorporated (as loss of 14C from the supernatant) by liquid scintillation counting (LSC). This value was expressed as a proportion of the value obtained for suspensions of yeast without leukocytes to give a percentage of yeast phagocytosis. Cell respiration assays: Glucose transport was assayed indirectly by measuring the comparative concentrations of 3H-2-deoxyglucose (3H-DOG; Amersham, UK) in cell suspension supernatants, as described previously. 24 Briefly, cells were incubated together with varying concentrations of IL-2, in 96-well round-bottom microtitre plates (Flow Laboratories) to which 3H-DOG was added to give a final concentration of 0.78/iCi/ml in a total of 0.2 ml DPBS (Flow Laboratories). After allowing uptake for 30 or 60min at 37C for neutrophils and monocytes, respectively, the plates were centrifuged at 800 x g for 5 min and 0.05 ml of supernatant was removed for LSC. Uptake of 3H-DOG was Suppression of phagocytes by 1L-2 calculated as the total disintegration/ml added minus the disintegration/ml remaining in the supernatant. A standard assay for hexose monophosphate shunt (HMPS) activity which measures the conversion of D-(1-14C) glucose (Amersham) 2s to 14CO2 in the oxidative pathway of energy generation, was adapted to microassay form using two tube caps in small sealed jars. In one of the caps, 0.4 ml cell suspensions in DPBS containing varying concentrations of IL-2 and 14C glucose at a final concentration of 0.25 #Ci/ml were incubated at 37C for 45 min. Then 1 M hydrochloric acid was added to drive out dissolved 14CO. which was captured in 0.2 ml 5 M sodium hydroxide contained in the second cap and quantitated by LSC. Phorbol myristate acetate (PMA, Sigma) was added in some experiments after preliminary incubation of cells with IL-2 for 30 min.
Degranulation assays: A standard optical density assay for lysozyme activity based on lysis of dried Micrococcus lysodeiktius (Sigma reagents) was used to gauge specific granule release in the cell supernatants. Cells were incubated in 96-well microtitre plates with IL-2 in DPBS and 1% glucose for 30 min at 37C. 26 Beta-glucuronidase activity was used as a marker for azurophilic granule release in supernatants as for lysozyme activity and was assayed spectrophotometrically using phenolphthalein-glucuronic acid substrate (Sigma). 26 For both assays data is represented as a percentage of total cell enzyme content determined from sonicated suspensions. In some experiments the calcium ionophore A23187 (calimycin, 7.5 mM; Sigma) was added after cells were preincubated with IL-2 for 30 min.
Statistical anasis" All experimental data are representative of two to five similar experiments performed in triplicate with individual standard deviations (SD) of less than 10%. Values are expressed as mean _-t-SD and statistical significance was measured with the Student's t-test.

Results
Mobility: Adherence of both neutrophils and monocytes to dacron fibre was decreased following incubation with IL-2 (Fig. 1). The efl:ect on migration under agarose differed for the two phagocyte types: neither neutrophil random nor stimulated migration was altered as a result of exposure to IL-2, while both monocyte random and stimulated migration decreased significantly with increasing concentrations of IL-2 ( Fig. 2(a) and (b)).
Chemotactic indices for the monocytes were calculated from the ratio of stimulated migration/random migration. The indices for the  Phagocytosis: Typically, under our conditions, the phagocytosis assay yielded 40-50% uptake of C. albicans over a 30 min incubation period, neutrophils demonstrated no response to IL-2 in this assay and monocyte phagocytosis remained similarly unaffected (Table 1) cells. Addition of PMA (0.1/,zg/ml) to neutrophils elicited an equivalent stimulated response in both control cells (no IL-2 preincubation) and those preincubated with 1L-2 at 200 U/ml. However, the addition of PMA to monocytes which had been preincubated with IL-2 induced a response significantly less than that in control monocytes ( Fig. 4(b)).
Degranulation: Specific granule release was not significantly affected by incubation of neutrophils or monocytes with any concentration of IL-2 assessed. Similarly, azurophilic granule release did not vary significantly from the control for resting or stimulated phagocytes (Table 2(a) and (b)). Results expressed as per cent of total fl-glucuronidase content. Results expressed as per cent of total lysozyme count.  (Table 2) of neutrophil HMPS activity and adherence by IL-2 was significantly reduced, but not totally abrogated by heating. Heat treatment also significantly reversed the suppressive effect of IL-2 on monocyte HMPS activity. The suppressive effect on monocyte migration was totally abrogated by heat treatment of IL-2. Similarly, expired IL-2 (stored at 4C for at least 6 months) was no longer able to vary monocyte and neutrophil adherence values from that of control M199 (data not shown).

Discussion
Under our in vitro conditions we have shown that recombinant IL-2 exerts a suppressive effect on selected functions of freshly isolated resting and stimulated phagocytic cells. As summarized in Table 3, neutrophil adherence was reduced while migration and phagocytosis were not. Hexose monophosphate shunt activity was reduced by IL-2, however, stimulation of the cells with PMA negated this effect. Glucose uptake and cytoplasmic granule release were not aected. Compared to neutrophils, monocyte sensitivity to IL-2 was more pronounced with both a larger number of functions being affected and more profoundly so. Monocyte adherence and migration were reduced as were glucose uptake and HMP shunt activity. Phagocytosis and granule release were not affected.
The greater diversity of monocyte functions affected by IL-2 as compared to neutrophils may not be surprising since IL-2 is classically a cytokine of the lymphoid mononuclear cell line, with which monocytes are known to interact closely in various phases of the immune response. However, the activity of IL-2 is not restricted to these cells and its eects may be diverse, for example, IL-2 is able to stimulate the release of prostacyclin by endothelial cells, 27 so again it is not unexpected that neutrophils could also be targeted. The more profound effect of IL-2 on monocytes than on neutrophils was noticed particularly at the respiratory level. Membrane transport of deoxyglucose, a non-metabolizable glucose analogue and HMP shunt activity were both reduced in monocytes. The decreased availability of intracellular glucose correlates with the decreased activity state of the cell. As demonstrated by the migration assays, monocyte kinetic energy was reduced. The monocyte chemotactic indices (chemotaxis:random migration) for different IL-2 concentrations did not vary significantly, indicating that IL-2 did not affect the cells' ability to detect and respond to a chemotactic gradient. However, both random migration and chemotaxis values were reduced, indicating a slower response and decreased cell kinetic energy level. In contrast, neutrophil HMP shunt activity, but not deoxyglucose uptake, was suppressed by IL-2. Thus, glucose may have still been available via other respiratory pathways to support more of the neutrophils' activities.
We observed that the effect of IL-2 was not globally suppressive, but was restricted to certain areas of cell functions as has been previously noted by others. while phagocytosis and antibody-dependent cellular cytoxicity were minimally affected. 28 Klempner et al. 29 also observed suppression of neutrophil chemotaxis, but found that other functions, including superoxide production were not decreased. Kowanko and Ferrante 3 have also demonstrated suppression of neutrophil migration, but stimulation of the respiratory burst and degranulation. The contradictory effects observed by others and ourselves may be due to differences in the state of priming or deactivation of neutrophils under different study conditions. 31 Freshly harvested cells from normal donors may not be fully dormant if they have been exposed to even 'trivial' quantities of endotoxin or have been triggered by adherence to surfaces during preparation. 2 Inconsistencies in the reported effects of IL-2 may indicate that the modulatory effects of IL-2 on neutrophils are highly sensitive to the state of activation of the target cell and to the interaction of other mediators. This may be the explanation for our observation that HMP shunt activity is significantly reduced by IL-2 in resting neutrophils, but not in PMA stimulated neutrophils. Similarly the state of activation of the monocyte may alter the observed effects of IL-2.
Hellstrand and Hermodsson found that IL-2 induced suppression of natural killer cell cytotoxicity occurs indirectly via a p55IL-2R locus on monocytes. They showed that the primary IL-2-monocyte interaction leading to killer cell suppression could only occur in monocytes prior to adherence, demonstrating the importance of the activation state.
The variation in the observed effects of IL-2 on neutrophil chemotaxis may also be due to the different sources of IL-2 or its purity. The IL-2 Ala125 product used in this study is a recombinant molecule which contains an extra methionine residue at the N-terminus of the natural sequence molecule. This addition has had no demonstrable effect on the classic biological activity of the molecule as determined by various cell culture assay systems. The cysteine residue at position 125 has also been substituted by alanine which has resulted in a five-fold increase in specific activity as well as improving stability (Amersham Technical Information). Preliminary experiments comparing purified rat IL-2 with the Amersham product demonstrated a similar effect on neutrophil adherence (data not shown). However, the possibility remains that this difference in structure may affect the neutrophil response in the chemotaxis assay. The effects of IL-2 on neutrophils and monocytes that we observed were consistently suppressive.
Although IL-2 first became recognized because of its requirement for T-cell proliferation and, more recently, augmentation of cytotoxicity in natural killer cells and macrophages, s-9 it has been shown to exert several down-regulatory effects, e.g. myelopoiesis, specifically granulocyte-macrophage colony formation '4 and T-helper cell activation. 1'4 It is unclear whether the suppressive effect of IL-2 on the myelopoietic progenitor cell is mediated via regulatory T-cells 3 or acts directly. 4 It would seem logical that any direct effect of IL-2 on the product of the myelopoietic progenitor cell, i.e. the circulating neutrophil and monocyte, may also be correspondingly suppressive under similar conditions. An indirect T-cell mediated effect is unlikely in our neutrophil experiments as the total lymphocyte contamination was only 1-2%. Burdach et al. did observe a T-cell mediated effect, but their T-cell versus effector cell ratio was 1:5. The short assay incubation times also favour a direct effect.
However, contamination of the monocyte preparation by lymphocytes ranged from 10 to 30%, which may have allowed interaction with T-cells. The assay times again favoured a direct effect in all, but the random migration and chemotaxis experiments.
Once cells migrate into tissue compartments their activation state and response may be altered, depending on the presence of other cytokines and activators. Thus, Wahl et al. 9 found that IL-2 augmented monocyte microbicidal activity, but only after the monocytes had been activated and cultured to express the IL-2 Tac receptor. The appearance of these receptors on the cells during the process of macrophage maturation occurs as the ceils become localized in the tissues. Hall et al. 27 also suggested that IL-2 may regulate the inductive phase of the immune response by inhibiting the proliferative response of T-cells to mitogens and preventing further attachment of leukocytes via stimulation of prostacyclin production by endothelial cells, which may be another indirect down-regulatory effect on circulating phagocytes. Modulation of leukocyte function by IL-2 has been demonstrated to operate via the Tac protein which was, until recently, believed to be expressed only by activated T-cells. Experimental evidence has extended the modulatory role of IL-2 to include a much wider range of cells as mentioned previously. More recently, cell membrane proteins distinct from the Tac receptor protein have been found which bind IL-2 and cause activation in ceils including natural killer cells and' certain B-cell lines. s Thus, as freshly isolated peripheral blood monocytes and neutrophils do not possess the Tac IL-2 receptor protein, l an interaction with IL-2 via other membrane receptors is probable. Our findings support this, as the only assays we used which allow suflqcient time for new Tac protein synthesis and expression were the monocyte random migration and chemotaxis assays.
We are confident that the effects we describe here 64 Mediators of Inflammation. Vol 2.1993 are due to the activity of IL-2. We have previously demonstrated that endotoxin is unlikely to be implicated as Polymyxin B does not affect these assays 21 and the effects are depressive rather than stimulatory which would be expected of endotoxin. Furthermore, we have shown the activity to be sensitive to heat, which endotoxin is not and to fade after prolonged storage. Interestingly, the effect of IL-2 was not totally abrogated by heating as measured by the HMP shunt and adherence assays. This is consistent with the observation that the Ala125 residue substitution in our IL-2 preparation increases heat stability of the recombinant molecule over that of purified IL-2 (Amersham Technical Information). Heating of this preparation to 80C for 24 h at the production laboratories demonstrated retention of up to 25% of original biological activity as measured by standard assay (personal communication with Amersham Technical Advisor). After 6 months storage at 4C, IL-2 activity was totally lost as measured in the monocyte and neutrophil chemotaxis assay (data not shown).
Finally, the effects observed occurred over a concentration range similar to that shown to have biological effects on other cells.
Our data supports the view that IL-2 is a multifunctional cytokine which influences a variety of immune cells and activities probably depending on the hosts' current immune state and the local tissue milieu.