Main

Neonates are vulnerable to infections because of their immature immunity. They are deficient in the magnitude and speed of T- and B-cell responses, including T- and B-cell proliferation, impaired cytokine production, and poor antibody responses (1–3). The impaired IFN-γ production of neonatal T cells is a major factor in their impaired defenses against intracellular pathogens. The decreased IL-4 production is associated with the poor antibody response of B cells, which can be overcome in vitro using mutant thymoma cells and T-cell supernatant (1–4).

Compared with adults, neonates have relative deficiencies in several arms of the immune responses. These form a complex network of interaction, with T cells playing a pivotal role. The immune deficiency of neonatal T cells may be related to their immature phenotype, as the majority of human cord blood T cells has the CD45RA+ ‘naive' phenotype, whereas both the CD45RA+ ‘naive' and the CD45RO+ ‘memory/mature' phenotypes are in equal numbers in adult peripheral blood MNC (5, 6).

IGF-I, as a lymphohemopoietic cytokine, has been reported to have profound positive effects on immune function (7). It has been found to enhance phagocytosis of human polymorphonuclear leukocytes and natural killer cell activity (8–10). IGF-I also stimulates human B-cell proliferation and antibody secretion (11–13). Moreover, it has the potential to augment lectin or anti-CD3-stimulated T-cell proliferation (14–16). However, relatively little is known about the effects of IGF-I on cytokine expression. Recently, several pieces of evidence indicated that the age-related decline of immunocompetence may be caused, at least in part, by the decline of GH and IGF-I production (17–19). Interestingly, serum concentration of IGF-I increased slowly from 76 ng/mL in newborns through early childhood, then, with a steep increase during puberty, to approximately 500 ng/mL. After puberty, a continuous fall in serum IGF-I concentration was apparent throughout adulthood, to a mean of 100 ng/mL at the age of 80 y (20). The developmental profile of IGF-I seems to reflect that of the immune function. Thus, we hypothesize IGF-I may modulate immune function and perhaps even augment or hasten the neonatal immune response.

In this study, we investigated the effects of IGF-I on cytokine mRNA expression and protein production in neonates using cord blood MNC as models. This, to the best of our knowledge, is the first report on the effects of IGF-I on cytokine mRNA expression and protein production in neonates.

Methods

Isolation of MNC.

Human umbilical cord blood was obtained from the placentas of normal, full-term infants, after the placentas were delivered and separated from the infants, with prior written informed consent of their parents. The protocol was approved by the Ethics Committee of the University of Hong Kong. Adult peripheral blood was obtained from healthy adult volunteers aged 25 to 45 y old. All samples were collected in heparinized flasks. Neonatal and adult MNC were isolated from whole blood by centrifugation, using Ficoll-Hypaque gradients (Pharmacia Biotech, Uppsala, Sweden). The MNC at the interface were collected, washed three times, and resuspended at a density of 1 × 106 cells/mL in a serum- and hormone-free medium, Dulbecco's Modified Eagle's Medium Nutrient Mixture F-12 Ham (DME/F-12; Sigma Chemical Co., St. Louis, MO), which did not contain insulin, IGF-I, or other hormones, supplemented with 50 IU/mL penicillin and 50 μg/mL streptomycin. Cell viability, as measured by trypan blue exclusion, was more than 99%.

Culture and activation of MNC in vitro.

A total of 1 × 106 MNC were cultured in the presence or absence of PHA (1 or 10 μg/mL; Sigma), with and without IGF-I (100 ng/mL; R&D System, Minneapolis, MN), and incubated for varying periods at 37°C in a humidified atmosphere containing 5% CO2. For cytokine and LAG-3 mRNA analysis, cells were collected after 12 h of culture. For cytokine protein assay, cells were incubated for 3 d. After incubation, supernatants and cells were collected and stored at −80°C until further processing.

Proliferation assay.

Cell proliferation was assayed by [3H]thymidine (TdR) incorporation. Cells (2 × 105) were cultured in triplicate samples in U-bottomed 96-well microtiter plates for 3 d at 37°C in a 5% CO2 atmosphere. [3H]-TdR (Amersham, Buckinghamshire, England) was added to a final concentration of 18.5 kBq/well during the last 6 h of culture. Cells were harvested on a glass-fiber filter, and counted by liquid scintillation counting.

RNA isolation.

Cell total RNA was extracted by using RNeasy Total RNA kits (QIAGEN Inc., Valencia, CA). The RNA content of the solution was quantified, using the OD at 260 nm measured on a Genequant spectrophotometer supplied by Pharmacia-Biotech (Cambridge, England). The RNA aliquots were stored at −80°C until analysis. The ratio 260/280 nm was always more than 1.8.

Semiquantitative RT-PCR.

Semiquantitation of cytokine and LAG-3 mRNA expressions was assayed by comparative RT-PCR analysis as described (21, 22). Briefly, RNA was treated with deoxyribonuclease I (amplification grade; GIBCO-BRL, Gaithersburg, MD) for digesting single- and double-stranded DNA to oligonucleotides before cDNA synthesis. cDNA was synthesized from oligo-dT-primer RNA by RT with moloney murine leukemia virus superscript reverse transcriptase from GIBCO-BRL. One microgram of total RNA and 1 μL of oligo(dT)12–18 (500 μg/mL), purchased from GIBCO-BRL, were mixed with diethylpyrocarbonate H2O in a final volume of 12 μL, and the mixture was heated to 70°C for 10 min and quickly chilled on ice. Four microliters of 5× first-strand buffer, 2 μL of 0.1 M DTT, 1 μL of 10 mM deoxynucleotides (dNTPs), and 1 μL (200 units) of Superscript, obtained from GIBCO-BRL, were added to the mixture and incubated for 50 min at 42°C. The reaction was inactivated by heating for 15 min at 70°C. The final cDNA product was stored at −20°C for subsequent cDNA amplification by PCR.

PCR-conditions, allowing reliable comparison of IL-2, IL-4, IL-6, IFN-γ, or LAG-3 mRNA expression with β-actin or GAPDH mRNA expression in different samples, were established by making serial dilutions of template cDNA at constant cycle numbers for each primer pair to verify the linearity of PCR amplification. Reaction mixtures for PCR of β-actin, GAPDH, cytokines, and LAG-3 contained 1 μL of cDNA, 5 μL of 10× PCR buffer (200 mM Tris-HCl pH 8.4, 500 mM KCl), 1 μL of 10 mM dNTPs, 1.5 μL of 50 mM MgCl2, 1 μL of 10 μM each primer mix (Table 1), 0.25 μL of Taq DNA polymerase (5 U/mL), obtained from GIBCO-BRL, and 37.25 μL of distilled water in a final volume of 50 μL. PCR was performed on a thermal cycler purchased from Perkin-Elmer (Foster City, CA). The optimal cycling profile for β-actin, IL-2, and IL-6 was 94°C (1 min), 60°C (1 min), and 72°C (2 min) for 30 cycles; for IL-4 and IFN-γ, 94°C (1 min), 60°C (1.5 min), and 72°C (2 min) for 35 cycles; and for GAPDH and LAG-3, 94°C (1 min), 58°C (1 min), and 72°C (2 min) for 30 cycles; with a final extension at 72°C for 5 min.

Table 1 Primers and PCR product sizes
Full size table

To semiquantify the amounts of each cytokine and LAG-3 mRNA expression, the signal intensity of each cytokine and LAG-3 production was compared with that of β-actin or GAPDH product amplified from that same cDNA sample in separate reactions. The products were separated on 2% agarose gel and stained with ethidium bromide, and then quantitative analysis by Fluorimager (Molecular Dynamics, Sunnyvale, CA) was performed for each of the fragments. The density of β-actin or GAPDH of each sample was set at 100%. The value for each cytokine and LAG-3 mRNA was expressed as a percentage of the density of β-actin or GAPDH in the same RNA sample.

Cytokine assays.

Concentrations of IL-2, IL-4, IL-6, and IFN-γ in culture supernatants were measured by specific ELISA assays with commercially available kits (Genzyme Diagnostics, Cambridge, MA) according to the manufacturer's instructions. Each sample was determined in duplicate. Minimal detectable concentration of IL-2, IL-4, IL-6, and IFN-γ was 4, 6, 18, and 3 pg/mL, respectively.

Statistical methods.

To determine difference between paired groups the Wilcoxon signed-rank sum test was used. The Mann-Whitney U-statistic test was used to determine difference between unpaired groups.

RESULTS

Decreased proliferative response and cytokine protein production inneonates.

To evaluate neonatal immune function, we compared the proliferative response and cytokine protein production between neonates and adults. The proliferative response of neonatal MNC to PHA (Fig. 1A) and the production of IFN-γ from PHA-stimulated neonatal MNC (Fig. 2) were reduced, as compared with that of adults, when using 1 μg/mL of PHA. IFN-γ production in neonates was only about 29% of that of adults. The production of IL-2 and IL-4 from both neonatal and adult MNC, when stimulated with 1 μg/mL of PHA, was undetectable. Using 10 μg/mL of PHA to stimulate MNC resulted in decreased production of both IL-2 and IL-4 in neonatal MNC when compared with that of adults (IL-2, 523.31 ± 277.70 pg/mL versus 1401.40 ± 50.62 pg/mL, p< 0.05; IL-4, 12.54 ± 3.06 pg/mL versus 64.87 ± 12.67 pg/mL, p< 0.05). The production of IL-2 in PHA-stimulated neonatal MNC was 37% of that of adults, and IL-4 was only 13% of that of adults. In contrast, the concentration of IL-6 protein production in PHA-stimulated neonatal MNC was higher than that of adults (Fig. 3).

Figure 1
figure 1

A, Proliferative responses of neonatal and adult MNC to IGF-I. Neonatal (n= 20) and adult (n= 10) MNC (1 × 106/mL) were stimulated with IGF-I (100 ng/mL) and/or PHA (1 μg/mL) for 3 d. The differences of proliferative responses between PHA and IGF-I+PHA were significant for neonates (p< 0.001) and adults (p< 0.01). Results are expressed as mean ± SEM. B, Proliferative responses of neonatal MNC to different concentrations of IGF-I. Neonatal MNC (1 × 106/mL) were stimulated with IGF-I at 1, 5, 25, 100, and 200 ng/mL and/or PHA (1 μg/mL) for 3 d. Results shown are representative of five different experiments. Proliferative responses were measured by [3H]-TdR incorporation as described in “Methods.”

Full size image
Figure 2
figure 2

IFN-γ production by neonatal and adult MNC in vitro. MNC (1 × 106/mL) were stimulated with IGF-I (100 ng/mL) and/or PHA (1 μg/mL) for 3 d. IFN-γ concentrations in supernatants were measured by specific ELISA kits. Results shown indicate IFN-γ production from nine neonates and 10 adults. The differences of IFN-γ production between neonates and adults induced by IGF-I (p> 0.05) and IGF-I+PHA (p> 0.05) were not significant. The difference of IFN-γ production between neonates and adults induced by PHA was significant (p< 0.05).

Full size image
Figure 3
figure 3

IL-6 production by neonatal adult MNC in vitro. MNC (1 × 106/mL) were stimulated with IGF-I (100 ng/mL) and/or PHA (1 μg/mL) for 3 d. IL-6 concentrations in supernatants were measured by specific ELISA kits. Results shown indicate IL-6 production from 11 neonates and 10 adults. The differences of IL-6 production between neonates and adults induced by IGF-I (p< 0.05) and PHA (p< 0.01) were significant. The difference of IL-6 production between neonates and adults induced by IGF-I+PHA was not significant (p> 0.05).

Full size image

Augmentation of proliferative response to PHA by IGF-I.

There was no proliferative response to IGF-I from 1 ng/mL to 200 ng/mL (Fig. 1B). However, IGF-I increased the proliferative response of neonatal MNC to PHA in a dose-response fashion (Fig. 1B). IGF-I enhanced both neonatal and adult MNC proliferative responses to PHA significantly. Moreover, IGF-I could increase proliferative response of neonatal MNC to PHA back to a level similar to that of adults (Fig. 1A).

Increase of IFN-γ and IL-6 protein production by IGF-I.

IGF-I alone did not induce IFN-γ, IL-2, and IL-4 production in most neonates and adults, except for two neonates whose IFN-γ production could be induced by IGF-I alone (Fig. 2). However, IGF-I alone induced IL-6 production in both neonatal and adult MNC (Fig. 3). The production of IL-6 induced by IGF-I alone in neonates was lower than that of adults (Fig. 3). In the presence of 1 μg/mL of PHA, IGF-I significantly increased IFN-γ and IL-6 production in both neonatal and adult MNC (Figs. 2 and 3). For PHA-stimulated neonatal MNC, IGF-I increased IFN-γ production by 8-fold, whereas for adult MNC, by only 1.5-fold. As a result, IGF-I increased neonatal IFN-γ production in PHA-stimulated MNC back to a level similar to that of adults (Fig. 2). The production of IL-2 and IL-4 of both neonatal MNC (IL-2, 523.31 ± 277.70 pg/mL versus 388.77 ± 196.06 pg/mL, p> 0.05; IL-4, 12.54 ± 3.06 pg/mL versus 15.38 ± 3.70 pg/mL, p> 0.05) and adult MNC (IL-2, 1404.40 ± 50.62 pg/mL versus 1275.20 ± 52.38 pg/mL, p> 0.05; IL-4, 64.87 ± 12.67 pg/mL versus 67.87 ± 12.15 pg/mL, p> 0.05), when stimulated by 10 μg/mL of PHA, was not affected by IGF-I.

Increase of IFN-γ and IL-6 mRNA expression by IGF-I.

IGF-I alone did not induce neonatal IL-2, IL-4, and IFN-γ mRNA expression, but could induce IL-6 mRNA expression in neonatal MNC after 12 h of culture (Fig. 4). When neonatal MNC were stimulated with PHA (1 μg/mL), IGF-I further increased IL-6 and IFN-γ mRNA expression but had no influence on IL-2 and IL-4 mRNA expression (Fig. 4).

Figure 4
figure 4

Cytokine mRNA expression in neonatal MNC. MNC (1 × 106/mL) were stimulated with IGF-I (100 ng/mL) and/or PHA (1 μg/mL) for 12 h. RNA isolation and semiquantitative RT-PCR for cytokine mRNA expression were as described in “Methods.” Amplification products of cytokines and β-actin cDNA were separated on 2% agarose gel (A) and evaluated by Fluorimager (B). Results shown in (A) are representative of, and in (B) are the mean of, 10 different neonates for IL-2, 6 different neonates for IL-4 and IFN-γ, and 14 different neonates for IL-6.

Full size image

Enhancement of LAG-3 mRNA expression by IGF-I.

Neonatal MNC, when unstimulated, expressed a low level of LAG-3 mRNA. IGF-I alone did not induce LAG-3 mRNA expression in neonatal MNC. When stimulated with PHA, significant up-regulation of LAG-3 mRNA expression in neonatal MNC was detected after 12 h of culture. IGF-I further enhanced LAG-3 mRNA expression in PHA-stimulated neonatal MNC (Fig. 5).

Figure 5
figure 5

LAG-3 mRNA expression in neonatal MNC. MNC (1 × 106/mL) were stimulated with IGF-I (100 ng/mL) and/or PHA (1 μg/mL) for 12 h. RNA isolation and semiquantitative RT-PCR for LAG-3 mRNA expression were as described in “Methods.” Amplification products of LAG-3 and GAPDH cDNA were separated on 2% agarose gel (A) and evaluated by Fluorimager (B). Results shown in (A) are representative of, and in (B) are the mean of, nine different neonates for LAG-3.

Full size image

DISCUSSION

Neonates are susceptible to infection with viruses and other pathogens, such as Toxoplasma, Listeria, Salmonella, and Mycobacterium tuberculosis, that survive and replicate within cells. Cellular immunity is the major mechanism of host defense against these intracellular pathogens. Deficiencies in T cells and their secreting cytokines appear to be a major factor in neonates' susceptibility to these infections (3, 23). Our results showed that the proliferation of and IFN-γ and IL-4 production in neonatal MNC in response to PHA were much lower than those of adult MNC (Figs. 1 and 2). These results are consistent with those obtained by ELISA or bioassay after stimulation of neonatal MNC in various studies (24–28). The relative defects of both IFN-γ and IL-4 production by neonatal T cells seem to correlate with their lack of CD45RO+ memory T cells (1, 29). Neonatal IL-6 production in response to PHA, on the contrary, was higher than that of adults (Fig. 3), the reason for which remains unclear. Almost all studies demonstrated that neonatal IL-2 production in response to PHA was similar to that of adults after 48 h of culture, but Sautois et al. (27) reported that IL-2 production in PHA-stimulated neonatal MNC was significantly lower after 72 and 96 h of culture compared with that of adults. Our results also indicate that IL-2 production in neonates was significantly less than that of adults after 72 h of culture.

The deficiencies of IL-2, IL-4, and IFN-γ production, as stated above, play a significant role in the vulnerability of neonates. Therefore, it may be of clinical importance to identify agents, such as IGF-I, that may enhance production of these cytokines in neonates.

IGF-I, as a lymphohemopoietic cytokine, has been found to have profound positive effects on immune function in both animals and humans (7–16). Kooijman et al. (14) and Roldan et al. (15) have demonstrated that IGF-I is an important potentiating factor in human and bovine serum for mitogen-induced T-cell DNA synthesis. In serum-free culture, IGF-I alone did not induce proliferation, but it potentiated lectin- or anti-CD3-induced T-cell DNA synthesis (16), and its effects on DNA synthesis are possibly mediated by its potentiating effects on IL-2 production (30).

In this study, we found that IGF-I alone could not induce proliferation in either neonatal or adult MNC in serum-free culture (Fig. 1). IGF-I alone also could not induce IFN-γ, IL-2, and IL-4 protein production and mRNA expression in neonates (Figs. 2 and 4). IGF-I alone could not induce IFN-γ, IL-2, and IL-4 protein production in adults either (Fig. 2). However, IGF-I alone induced a high level of IL-6 protein production and mRNA expression in neonatal MNC as well as IL-6 protein production in adult MNC in our serum-free culture system (Figs. 3 and 4). IGF-I could further increase significantly IL-6 protein production and/or mRNA expression in both PHA-stimulated neonatal and adult MNC (Figs. 3 and 4).

Although IL-6 was initially thought to be a proinflammatory cytokine, recent findings suggest that IL-6 has many anti-inflammatory effects. Administration of IL-6 in humans resulted in the induction of circulating IL-1 receptor antagonist and TNF receptor p55, but not IL-1β and TNF (31). IL-1 receptor antagonist can reduce pathologic processes such as septic shock, and TNF receptors can block lipopolysaccharide-mediated lethality in animal models (32). IGF-I could enhance proliferation and IL-6 production in concanavalin A- or lipopolysaccharide-stimulated peripheral blood MNC from rats immunosuppressed by dexamethasone (33). This may, in part, explain why IGF-I could improve survival in a murine model of Gram-negative sepsis (32). Based on our observation that IGF-I alone could induce significant neonatal IL-6 production, we believe it is of interest to investigate whether IGF-I could also improve survival in neonatal sepsis.

In NSE/hIL-6 transgenic mice, which expressed high concentrations of circulating IL-6, circulating IGF-I concentrations were significantly lower than those of nontransgenic littermates. Treatment of nontransgenic mice of the same strain with IL-6 resulted in a significant decrease in IGF-I concentration (34). It was also demonstrated that circulating IL-6 concentrations were negatively correlated with IGF-I concentrations in patients with systemic juvenile rheumatoid arthritis (34). This negative correlation between IGF-I and IL-6 could be, at least in part, related to the increase of IGF binding protein-1 production of liver cells by IL-6 (35). Taken together with our observation that IGF-I could increase IL-6 production by MNC, it is conceivable that a feedback control may exist between IGF-I and IL-6, with IGF-I having a positive effect on IL-6, and IL-6 a negative effect on IGF-I.

CD4+ T cells (Th) have been classified into two populations (Th1 and Th2) on the basis of their profiles of cytokine secretion. Th1 cells produce IFN-γ, IL-2, and TNF-β, which mediate macrophage activation and delayed-type hypersensitivity reactions. Th2 cells produce IL-4, IL-5, IL-6, and IL-10, which act as growth and/or differentiation factors for B cells, and provide optimal help for humoral immune responses. CD4+ cell subsets with a less-differentiated cytokine profile than Th1 or Th2 cells are designated Th0, which probably represent a heterogeneous population of T cells.

The cytokine response of the single Th0 cell can remain mixed or further differentiate into the polarized Th1 or Th2 pathway in subjects with a particular genetic background or under the influence of microenvironmental signals (36–39). Through their secretion of IFN-γ, Th1 cells can protect against intracellular pathogens, whereas Th2 responses have been associated with allergic diseases and asthma (36–39). The selection of Th cell phenotype occurs at an early stage of the immune response, and is determined by a number of factors. Cytokines appear to play the most important role, and it is likely that the other factors affect T-cell differentiation by acting on the production of endogenous cytokines at priming (38, 39).

For PHA-stimulated neonatal MNC, our results demonstrated that IGF-I could significantly increase IFN-γ protein production and mRNA expression (Figs. 2 and 4), but had no influence on IL-2 and IL-4 protein production and mRNA expression (Fig. 4). Similar results were also found in PHA-stimulated adult MNC, although adult MNC were less responsive to IGF-I stimulation with respect to IFN-γ production when compared with neonatal MNC (1.5-fold versus 8-fold;Fig. 2). These results suggest that IGF-I could promote the maturation of neonatal T cells and regulate T-cell development toward Th1 in both neonates and adults, when MNC were stimulated with PHA.

It has been demonstrated recently that IL-12 could induce IFN-γ secretion by T cells, promote growth of activated T and natural killer cells, and modulate IgE synthesis. The rapid induction of IFN-γ production by IL-12 both in vitro and in vivo indicates that it may be involved in the differentiation of Th1 cells during the normal immune response (40, 41). D'Andrea et al. (42) have reported that PHA could not induce expression of IL-12 p40 mRNA and secretion of IL-12 p40 or p70 in adult MNC. We also found that PHA and IGF-I could not induce IL-12 p40 mRNA expression in neonatal MNC (data not shown). However, McKenzie et al. (43) demonstrated that the inability of neonatal MNC to produce IFN-γ might partially reside in the neonatal mononuclear phagocyte. The ability of the mixture of neonatal lymphocytes and monocytes in subsequent endotoxin-stimulated culture to produce IFN-γ could be enhanced by specific treatments in vitro of the neonatal monocytes. The treatments included 1) in vitro preincubation of the monocytes for 2 wk;2) exposure of the monocytes to conditioned medium from healthy adults; or 3) exposure to macrophage colony-stimulating factor, granulocyte-macrophage colony-stimulating factor, or IFN-γ (43). These treatments probably promoted the maturation of neonatal monocytes and increased their ability to secrete the necessary T lymphocyte-stimulating cytokines, such as IL-12. Our observations of an increase of IFN-γ production in PHA-stimulated neonatal MNC may have resulted from promoting the maturity of neonatal monocytes by IGF-I.

LAG-3 is a member of the immunoglobulin superfamily that is selectively transcribed in human activated T and natural killer cells (44). LAG-3 and CD4 proteins share the same ligand for MHC class II molecules, which suggests that LAG-3 is closely related to CD4 (45). Annunziato et al. (46) demonstrated that the expression and release of LAG-3-encoded protein by human CD4+ T cells were associated with IFN-γ production and that LAG-3 expression in activated CD4+ human T cells appeared to be preferentially associated with the differentiation of Th1. Our results indicated that IGF-I not only increased IFN-γ transcript and protein production (Figs. 2 and 4), but also enhanced LAG-3 mRNA expression in PHA-stimulated neonatal MNC (Fig. 5). These results suggest that, in neonatal MNC, LAG-3 expression was associated with the increase of IFN-γ production by IGF-I. The increase of LAG-3 expression may be involved in the regulation of human T-cell differentiation toward the Th1 profile by IGF-I.

However, IGF-I could also further increase IL-6 protein production and/or mRNA expression in both PHA-stimulated neonatal and adult MNC (Figs. 3 and 4). These results paradoxically indicated that IGF-I would also promote T-cell differentiation toward the Th2 profile. It has, however, been shown that besides T cells, monocytes and macrophages, on stimulation, can also secrete IL-6. Therefore, it needs to be clarified whether secretion of IL-6 in MNC stimulated by IGF-I is from cells other than T cells. It is not known whether IGF-I can promote T-cell differentiation toward the Th2 profile.

In conclusion, IGF-I alone can induce significant IL-6 production by neonatal MNC. It can also increase neonatal IFN-γ production in PHA-stimulated MNC to adult concentrations, which is associated with an increase in LAG-3 expression.