Main

By its ability to phagocytize and then kill microorganisms, the neutrophil granulocyte plays an important role in the defense against infections, especially bacterial and fungal infections(1). To enable neutrophil phagocytosis of microorganisms, they have to be opsonized by complement components and/or antibodies, which bind to specific receptors on the cell surface of the neutrophil, the complement receptors, and the Fcγ-receptors.

Three different types of Fcγ-receptors are found on neutrophils and monocytes, FcγRI (CD64), FcγRII (CD32), and FcγRIII (CD16)(2–4). Interaction between Fcγ-receptors and the Fc-portion of immunoglobulin molecules triggers various important biological functions in the cell, such as phagocytosis, activation of the respiratory burst, degranulation, and antibody-dependent cell cytotoxicity(3,5,6). FcγRII and FcγRIII are normally expressed by neutrophils and are low affinity receptors that only bind polyvalent IgG-complexes. In contrast, FcγRI is constitutively expressed only by monocytes and to a very low extent by neutrophils. It is a high affinity receptor, which means that it binds monomeric IgG(3,7).

During bacterial infections the FcγRI expression on neutrophils has been shown to be increased in adult patients(8–10). This increased expression is probably caused by cytokines produced by host cells during the infection. IFN-γ induces FcγRI expression on neutrophils in vivo and in vitro as does G-CSF in vivo(11–16).

Newborn infants, especially preterm infants, have an increased susceptibility to serious and overwhelming bacterial infections(17). Neutrophils from preterm infants express FcγRII and FcγRIII to a significantly lesser extent than neutrophils from term infants and adults(18,19). Furthermore, the capacity for phagocytosis, as well as for other aspects of neutrophil function, is reduced in neutrophils from preterm infants(20–23).

The purpose was to investigate whether expression of FcγRI on neutrophils was increased during bacterial infections in preterm and term neonates, in older infants, and in children and to compare the results with those obtained in adults with bacterial infections.

MATERIALS AND METHODS

Patients. Three different groups of patients were studied. Group 1 consisted of 12 preterm and term newborn infants (gestational age 24-42 wk) with culture verified (n = 7) or suspected (n = 5) septicemia (Table 1). Suspected septicemia was defined as clinical symptoms in combination with negative blood culture, but with an elevated serum level of C-reactive protein and/or leucopenia. The infection started at the median age of 4 d (0 to 60 d), and the first analysis of receptor expression was performed at the median of 1 d (0 to 2 d) after start of treatment, and at the median of 1 d (0 to 3 d) after the onset of symptoms. From nine infants a second blood sample was obtained 1-6 wk (median 2 wk) after the first one.

Table 1 Newborn infants with diagnosed or suspected septicaemia
Full size table

Group 2 consisted of 14 infants and children, aged 10 mo to 6 yr, who were hospitalized because of an acute bacterial infection (Table 2). The blood sample was obtained at the median of 1 d (0 to 5 d) after the onset of treatment and 3 d (median) after start of symptoms. From seven children a second blood sample was available at a median time of 3 months (6 d to 9 mo) after the infection under study.

Table 2 Infants and children hospitalized because of an acute bacterial infection
Full size table

Group 3 consisted of six adult patients, 29 to 84 (mean 55) yr of age, who were hospitalized because of an acute bacterial infection (Table 3). The blood sample was obtained at the median of 1 d (0 to 2 d) after the onset of treatment.

Table 3 Adult patients hospitalized because of an acute bacterial infection
Full size table

The control groups consisted of 7 noninfected preterm neonates, 14 healthy term neonates, and 26 healthy adults, blood-donors as well as laboratory technicians.

The study was performed with permission from the Ethics Committee, Faculty of Medicine, Uppsala University. The blood samples were collected after informed consent from the patients and/or their parents.

Preparation of leukocytes. Leukocytes were prepared according to the procedure described by Hamblin et al.(24). Briefly, 0.5 or 1 mL heparinized blood was mixed with an equal volume of 0.4% (vol/vol) formaldehyde in PBS prewarmed to 37°C. The mixture was incubated for 4 min at 37°C and then 40 mL of prewarmed 0.85% (w/v) NH4Cl in Tris-HCl-buffer [Tris(hydroxymetyl)-aminomethan 0.01 mol/L, pH 7.4] was added and the mixture was incubated for another 15 min to lyse the erythrocytes. The cells were then centrifuged for 5 min at 350 × g after which the supernatant was removed. The cells were washed twice with 5 mL PBS containing sodium citrate (0.012 mol/L) and HSA (0.1%, w/v). Finally the cells were suspended in 0.5 mL PBS with sodium citrate and HSA and the concentration of granulocytes was determined. The granulocytes were then diluted to a concentration of 1.7-2.5 × 106/mL and kept on ice.

Mab. The following mouse Mab were used in the study: unconjugated anti-CD64 (clone 32.2, Medarex, Annandale, NJ), unconjugated anti-bromodeoxyuridine (anti-BrdU), and FITC-labeled rabbit-anti-mouse Ig purchased from Dakopatts (Dakopatts A/S, Glostrup, Denmark).

Labeling of leukocytes with antibodies to cell surface antigens. A total of 50 µL samples of the leukocyte suspension were mixed with 50 µL of optimally titrated mouse Mab against CD64 and incubated for 30 min at 4°C. After incubation the cells were washed twice with 3 mL PBS. The cells were then mixed with 50 µL FITC-conjugated rabbit-anti-mouse Ig and incubated for 30 min at 4°C. After incubation the cells were washed twice with 3 mL PBS and then diluted with 200 µL PBS containing sodium citrate and HSA. Leukocytes were also labeled by an identical procedure with negative isotype controls for mouse IgG1 (anti-BrdU). After labeling the cells were kept on ice until analysis.

Flow cytometry. Flow cytometric analysis was performed on an EPICS II Profile flow cytometer (Coulter Company Inc., Hialeah, FL). Granulocytes were identified based on their forward and side scatter dot-plot profiles. A gate was set around the granulocyte population and the FITC-fluorescence of the cells within the gate was measured. A minimum of 4000 events in the granulocyte gate was counted. Granulocyte expression of FcγRI (CD64) was measured as specific MFI of the whole granulocyte population and as the relative amount of FcγRI+ granulocytes. The specific MFI of neutrophil FcγRI expression was calculated by subtracting the background MFI obtained with the negative isotype control Mab from the value obtained with the anti-CD64 Mab. The relative amount of FcγRI+ granulocytes was calculated as the relative number of granulocytes showing a higher fluorescence intensity when stained with the anti-CD64 Mab than with the negative control Mab.

Statistical analysis. Statistical evaluations were made with Mann-Whitney U test and Wilcoxon's matched pairs test, both by the use of CSS Statistica (StatSoft Inc., Tulsa, OK).

RESULTS

Neutrophils from preterm and term neonates, older infants, children, and adults examined during the early phase of a bacterial infection showed a significantly higher expression of FcγRI compared with noninfected controls (p < 0.001) (Figs. 1 and 2 and Table 1). The markedly increased expression of FcγRI on neutrophils was evident both when measured as FcγRI positive cells, i.e. the proportion of cells showing a higher fluorescence intensity than the negative control (Fig. 1), and as the MFI of the whole neutrophil population (Fig. 2), which is related to the number of receptors per cell. The expression of FcγRI during bacterial infection was similar (p > 0.3) on neutrophils from preterm and term neonates, children and adults, respectively. In the preterm and term infants and children from whom a second blood sample was obtained after recovery from the infection, the neutrophil cell surface expression of FcγRI had decreased significantly (p < 0.01, respectively, p < 0.05) compared with the level during infection (Fig. 3).

Figure 1
figure 1

The relative amount (%) of FcγRI+ (CD64) granulocytes in preterm and term neonates (filled circles), infants and children (filled triangles), and adults (filled diamonds) with bacterial infection compared with healthy preterm (open circles) and term neonates (open squares) and healthy adults (open diamonds). The median values of each group are indicated.

Full size image
Figure 2
figure 2

Granulocyte cell surface expression, measured as specific MFI of FcγRI (CD64) on granulocytes from preterm and term neonates (filled circles), infants and children (filled triangles), and adults (filled diamonds) with bacterial infection compared with healthy preterm (open circles) and term neonates (open squares) and healthy adults (open diamonds). The median values of each group are indicated.

Full size image
Figure 3
figure 3

Comparison of the relative amount of FcγRI+ (CD64) granulocytes during and after a bacterial infection in preterm and term neonates and infants and children.

Full size image

No significant difference in neutrophil FcγRI expression during infection was found between patients with Gram-negative and Gram-positive infections in any group (results not shown). Neutrophils from healthy preterm and term neonates showed a significantly (p < 0.001) higher expression of FcγRI than neutrophils from healthy adults, when measured as relative amount of FcγRI-expressing neutrophils and MFI of the whole neutrophil population (Figs. 1 and 2).

DISCUSSION

During bacterial infections the expression of FcγRI on neutrophils was markedly increased in preterm and term infants, as well as in children and adults, compared with the expression on neutrophils from healthy individuals. The connection between bacterial infection and neutrophil FcγRI expression was further strengthened by the return to normal, or almost normal, values after recovery from the infection. Neutrophils from extremely preterm infants expressed FcγRI during bacterial infection to the same extent as neutrophils from term infants, children, and adults. Furthermore, although neutrophils from healthy preterm and term neonates demonstrated a somewhat higher expression of FcγRI than the almost insignificant expression by neutrophils from adults, the difference between infected and healthy individuals was equally marked. Measurement of receptor expression by means of mean fluorescence intensity is related to the number of receptor per cell and in that way shows that neutrophils are able to increase the expression of FcγRI markedly above the level where the neutrophils become positive for the receptor. These results are somewhat in contrast to the fact that newborn infants, and especially the preterm infants, are known to have an immature immune system in many aspects, as well as in the phagocyte system, contributing to a significant extent to their increased susceptibility to serious infections(18,19,21,25).

The only cytokines known to induce the expression of FcγRI on neutrophils in vivo are IFN-γ and G-CSF(11,12,15). In general, increased expression of FcγRI on neutrophils is considered as a result of the release of IFN-γ from activated T lymphocytes after antigenic challenge(26). The production of IFN-γ in vitro by lymphocytes from newborn infants is in general considered to be reduced compared with that of lymphocytes from adults, but results from different investigations vary(27,28). Studies on the stimulation of lymphocytes from neonates by Staphylococcus enterotoxin A and B demonstrate both normal and decreased responses, Staphylococcus aureus induced decreased responses, although stimulation with Listeria monocytogenes induced a normal production of IFN-γ(29–31). There are, however, until now no studies on the production of IFN-γ from lymphocytes obtained from newborn infants with bacterial infection. Adult patients, as well as newborn infants and even preterm infants, have been shown to produce G-CSF in high quantities in response to bacterial infections(32,33). However, also noninfected newborn infants often have high concentrations of G-CSF in serum(33), which makes it unlikely that G-CSF alone can explain the expression of FcγRI during bacterial infections. Further investigations are needed to elucidate the role of these two cytokines and possibly others in the up-regulation of neutrophil cell surface expression of FcγRI.

Stimulation of neutrophils through the high affinity receptor FcγRI is though to play an important role in the enhancement of neutrophil phagocytosis and intracellular killing of bacteria(5,34,35). This mechanism might be even more important in preterm infants, who have neutrophils with a function that is in many aspects inferior to the conditions in adults(21), and in whom the concentration of passively acquired IgG-antibodies from the mother is significantly lower compared with term infants(25). Recently Michon et al.(36) reported that neutrophils from children receiving recombinant human G-CSF for the collection of peripheral blood progenitor cells demonstrated a strongly increased expression of FcγRI and that antibody-dependent cell cytotoxicity exerted through this receptor was also increased. Based on these results the authors point to a possible therapeutic schedule using G-CSF in combination with Mab in adjuvant immunointervention.

Previous studies have indicated that the degree to which cell surface expression of FcγRI is up-regulated on neutrophils also depends on the type of bacteria causing the infection(8,10). In one study systemic infections caused by Gram-negative bacteria were found to induce a higher FcγRI receptor expression on neutrophils than systemic infections caused by Gram-positive bacteria(8). Another study showed that neutrophils from adult patients with pharyngitis caused by group A β-hemolytic streptococci expressed FcγRI to a higher extent than neutrophils from patients with pyelonephritis caused by Escherichia coli(10). In our study no such differences could be found, but the number of patients with culture verified infections in each group was small. One explanation for the variation in cell surface expression of FcγRI seen in our investigation could be the variation in the time interval between the time point of onset of symptoms and the performance of the test. In the group of infants and children, where the greatest variation is found, there is, however, no relationship between the cell surface expression of FcγRI and the duration between the start of treatment and performance of the test.

Early diagnosis and treatment is of crucial importance for the outcome of most bacterial infections in neonates. The access to a quick and reliable method for early detection of bacterial infections in this age group is therefore greatly desired. Because C-reactive protein, total and differential leukocyte count, the proportion of immature to mature neutrophils, and other tests currently used for this purpose have shown not to be reliable enough in the neonatal period, great efforts have been made to search for new and better diagnostic tools. The measurement of cytokines, such as interleukin-6 and tumor necrosis factor-α, and other substances, e.g. procalcitonin, as early indicators of bacterial infection has been evaluated in many studies. Although some of these markers do appear very early in the course of an infection, the results have so far not been convincing with respect to sensitivity and specificity(37–42). In addition, the concentration of cytokines in serum most often does not reflect their biological activity. Determination of the cellular response to cytokines, e.g. by measuring the cell surface expression of certain receptor molecules, may thus be a better way of revealing the early stage of the immune response to a bacterial infection. According to a recent study the expression of the complement and adhesion receptor CD11b on neutrophils might be a valuable diagnostic marker for early-onset neonatal infections(43). At least two studies also point to the possible usefulness of the expression of FcγRI on neutrophils as an early marker for severe bacterial infections, at least in adult patients(8,9). Measurement of the expression of FcγRI on neutrophils during bacterial infection could possibly be advantageous because this receptor is normally expressed to a very low extent.

In our study neutrophils from all children and adult patients with bacterial infection and all newborn infants, including the very preterm neonates, with verified or suspected bacterial infection expressed FcγRI. In addition, neutrophils from even extremely preterm infants expressed this receptor to the same extent as did neutrophils from adult patients. Further studies designed to evaluate the possible value of measuring the expression of FcγRI on neutrophils as an early indicator of bacterial infections, especially in the neonatal period, should be performed.