Main

Infection remains one of the most significant causes of morbidity and mortality in the newborn period(1). The increased susceptibility of the newborn to pathogens including bacteria and viruses appears to be due mainly to developmental deficiencies of the neonatal host defense system(2–5). With respect to viral infection, disease dissemination has been attributed to immaturity of the immune system including a defect in NKC and cytokine production in neonates(6–10). It has been shown that IL-12, IL-2, and IFN are the major cytokines involved in the up-regulation of NKC(11–14).

IFN-γ is an important pleiotropic molecule produced by activated T cells and NK cells. In contrast to IFN-α and IFN-β, the major biologic functions of IFN-γ are related to its immunomodulatory properties instead of antiviral activity(15). IFN-γ increases the expression of major histocompatibility complex I and II antigens, regulates the differentiation and function of macrophages, modulates T helper cell differentiation, and enhances NKC. Previous reports showed reduced expression of IFN-γ by mitogen-stimulated neonatal T cells compared with adult T cells(6–8). This deficiency in IFN-γ expression may explain the partly defective NKC observed in neonates.

Although it is well described that neonatal cells express lower levels of IFN-γ in response to mitogen stimulation, mechanisms underlying this defect remain largely unknown. This may include reduced transcription of IFN-γ in response to stimulation, inactivation, or methylation of enhancers of the IFN-γ gene which renders the promoter hyporesponsive to stimulation, increase in degradation of IFN-γ transcripts, inhibition of translation, or post-translational modification that prevents the release of functional proteins. Because it is known that IFN-γ expression in T cells is regulated by IL-12(16–18), it is also possible that a defect in IL-12 responsiveness by neonatal cells may contribute to lower IFN-γ expression and NKC. We therefore postulate that the relative defect in NKC in newborns may be due to decreased expression of IFN-γ in response to IL-12 or decreased NKC response to IL-12 directly. Here, we demonstrate that cord BMC responded to IL-12 stimulation resulting in induction of IFN-γ transcription, synthesis of IFN-γ, and activation of NKC, comparable to that seen using adult cells.

METHODS

Reagents. Recombinant human IFN-γ and rabbit polyclonal anti-IFN-γ antibodies were purchased from R&D Systems (Minneapolis, MN) and Endogen (Cambridge, MA), respectively. Human recombinant IL-12 was kindly provided by Dr. Stan Wolf, Genetics Institute (Boston, MA). Ionomycin and phorbol myristate acetate were purchased from CalBiochemicals (La Jolla, CA) and Sigma (St Louis, MO), respectively. Ficoll-hypaque and RNAguard were purchased from Pharmacia Biotech Inc. (Piscataway, NJ), whereas human A-B serum for NK assays was obtained from Bio-Whittaker (Walkersville, MD). In addition, random primers and enzymes for molecular biology experiments including RQ1 RNase-free DNase, Moloney murine leukemia virus reverse transcriptase, and Taq DNA polymerase were purchased from Promega (Madison, WI). ELISA kits for IFN-γ assay were purchased from Endogen.

Preparation of blood mononuclear cells. BMC were obtained from healthy HIV-seronegative adult volunteers or from the umbilical cords of healthy infants born by vaginal delivery and not at risk of HIV infection as previously described(19). Whole blood (cord or adult) was anticoagulated with 20 U/mL heparin (Elkins-Sinn, Cherry Hill, NJ); and BMC were isolated by density gradient centrifugation with Ficoll-hypaque. Cells were washed four times in Hanks' buffered saline solution before use. When umbilical cord samples were compared with adult BMC, the adult volunteer was phlebotomized within 1 h of collection of the umbilical cord samples.

After cell separation, cord or adult BMC were resuspended in RPMI medium supplemented with 10% human A-B serum, 100 U of penicillin, and 100 μg of streptomycin per mL (hereafter referred to as complete medium). BMC were cultured at a concentration of 1 × 106 cells/mL, and 5-10 mL of cell suspension were added to 10-cm plastic Petri dishes (Falcon, Oxnard, CA). Cells were incubated for indicated periods of time at 37°C in 5% CO2-enriched air, before use in the various assays described below.

RNA extraction and RT reaction. Total cytoplasmic RNA was extracted from 1 to 5 × 106 cells according to the method of Chomczynski and Sacchi(20). In brief, cells were lysed in 0.5 mL of denaturing solution containing guanidinium thiocyanate. Total RNA was obtained by extraction with phenol and chloroform-isoamyl alcohol. The OD260/280 ratio of extracted total RNA was routinely at or greater than 1.8. The extracted total RNA appeared to be undegraded when electrophoresed in a 1% agarose gel in TAE buffer (0.04 M Tris-acetate, 0.001 M EDTA). The total RNA extracted was then used for RT or Northern blot analysis.

To perform RT-PCR analysis, total RNA (2 μg) was treated initially with RQ1 RNase-free DNase for 30 min at 37°C to digest residual genomic DNA in the RNA preparation. After inactivation of the DNase by incubation at 95°C for 10 min, first-strand cDNA was reverse-transcribed from total RNA(21). In a 20-μl reaction, 2 μg of total RNA and 100 ng of random primers were heated at 95°C for 1 min and cooled briefly on ice before adding 5× RT buffer (250 mM Tris-HCl, pH 8.3, 375 mM KCl, 15 mM MgCl2, 50 mM DTT), dNTP (1.5 mM, final concentration), 8 U of RNAguard, and 200 U of Moloney murine leukemia virus reverse transcriptase. After 1 h at 37°C, the reaction was terminated by heating at 95°C for 5 min before cooling on ice.

Specific primers and cDNA probes for IFN -γ. The sequences of oligonucleotide primers specific for IFN-γ used in the experiments are as follows: upstream primer, 5′-AATGCAGGTCATTCAGATGTAGCGC-3′; downstream primer, 5′-GGATGAGTTCATGTATTGCTTTGCG-3′. This pair of primers generate a fragment of the IFN-γ gene consisting of 298 bp. To semiquantitate the steady-state levels of mRNA specific for IFN-γ, we used an internal reference gene, GAPDH as controls. GAPDH is an important enzyme for normal cellular function, and its steady-state mRNA levels are known to remain constant in most cell types under diverse conditions. RT-PCR on GAPDH mRNA was used as an internal control to compensate for differences in the starting amount and integrity of the RNA preparation used and in the efficiency of cDNA synthesis. The primer set (upstream: 5′-CAAAAGGGTCATCATCTCTG-3′; downstream: 5′-CCTGCTTCACCACCTTCTTG-3′) for generating a 446-bp PCR product specific for the GAPDH message was included in each PCR reaction.

PCR amplification of IFN -γ messages. Amplification of the reverse-transcribed cDNA was accomplished by subjecting up to 7 μ1 of the cDNA to a 100-μl PCR reaction with oligonucleotide primers specific for IFN-γ and GAPDH. The reaction mixture contained 50 mM KCl, 10 mM Tris-HCl (pH 9.0 at 25°C), 0.1% Triton X-100, 2 mM MgCl2, 0.2 mM dNTP, 50 pmol of each primer, and 2.5 U of Taq DNA polymerase. Mineral oil (100 μL) was overlaid on top of the mixtures to prevent evaporation during PCR. The reaction was performed for 33-35 cycles in a DNA thermal cycler (Coy Corporation, Grass Lake, MI). The PCR reaction was allowed to proceed for 33-35 cycles (95°C for 1 min, 55°C for 2 min, and 72°C for 3 min) before arresting the reaction in its logarithmic phase by soaking at 4°C. For each sample, a 30-μL aliquot of reaction mixtures was analyzed by ethidium bromide staining of the agarose gel after electrophoresis(21).

Northern blot analysis for IFN -γ. For Northern blotting, total RNA (10 μg/lane) was electrophoretically fractionated on 0.8% agarose gels and transferred by capillary action to a nitrocellulose membrane (GeneScreen, DuPont, Boston, MA). Hybridization and washing of the membrane was performed according to Church and Gilbert(22) using 32P-random primed probes specific for IFN-γ or GAPDH. Briefly, the hybridization solution for the oligonucleotide probes contained 3 × SSPE. The blot membrane was hybridized with randomly primed 32P-labeled specific IFN-γ or GAPDH cDNA probes at 68°C overnight. Excess probe was removed by two washes with 2 × SSPE and 0.1% SDS followed by two additional washes with 0.2 × SSPE and 0.1% sodium dodecyl sulfate for 15 min each at 68°C. The membrane was dried and autoradiographed with a XAR-5 film (Eastman Kodak, Rochester, NY) at -70°C in the presence of an intensifying screen for 5 d. As controls to ensure even loading of the gel with the same amount of total RNA, the membrane was probed with the GAPDH-specific cDNA.

Quantification of IFN -γ in culture supernatant. Quantification of the IFN-γ protein in the supernatant of BMC cultures after IL-12 treatments was done by means of the ELISA kit for IFN-γ from Endogen. According to the manufacturer's instructions, the standards in this ELISA have been calibrated to the NIH reference standard lot no. Gg23-901-530. One NIH unit equals 115 pg of the standard in this ELISA. The interassay variability of the assay in our laboratory was less than 10%.

NKC Assay. NK cytotoxicity against HIV-infected cells were measured as previously reported by us(19, 23). Approximately 106 HIV3B (HXB2)-infected target cells were labeled with chromium-51 and then washed. These target cells (2 × 103) resuspended in 50 μL of complete medium were added to each well of round-bottom microtiter trays (Corning, Corning, NY). BMC were then added at an effector-to-target ratio of 50:1 in 100 μL of complete medium, plates were centrifuged at 300 × g for 3 min before incubation. Triplicate wells were used for each condition. After incubation for 4 h, 100μL of supernatant were harvested from each well (sample A), the remaining 100 μL and cell pellet were collected separately after decontamination and solubilization with 50 μL of 50% sodium hypochlorite (sample B).

Percent of radioactivity released was calculated as:Equation NKC was calculated as follows:Equation

In 4-h assays, spontaneous release by target cells averaged 7.5% (SD 2.4%). Target cells infected with HXB-2 maintained high expression of viral antigen, with 88% of live cells expressing surface antigens by immunofluorescence assay after several months of continuous passage without addition of either virus or feeder cells(19).

Subjects. Informed consent was obtained from all patients or their guardians, and adult volunteers. Human experimentation guidelines of the U.S. Department of Health and Human Services and the UCSF-Committee on Human Research were followed in all cases.

RESULTS

IL-12 induction of IFN -γ transcription in cord and adult BMC. To investigate whether cord BMC were capable of synthesizing IFN-γ, cord BMC were treated with IL-12 at indicated concentrations, ranging from 10 to 1000 U/mL. At various time points (4, 8, 18, and 24 h) after IL-12 treatment, the cells were extracted for total RNA and the steady state levels of IFN-γ specific mRNA were examined. In the event that only very small amounts of cytokines are produced and locally consumed, low levels of cytokine messages in the limited number of cord BMC may not be detectable by Northern analysis. To circumvent this, we performed RT-PCR studies to examine the level of expression of IFN-γ. Here, we showed that IL-12, at concentrations as low as 10 U/mL, induced significant amounts of IFN-γ messages in cord cells, similar to the effects of IL-12 on adult BMC (Fig. 1). From the kinetics experiments, the induction of IFN-γ messages by IL-12 was detectable as early as 4 h after IL-12 treatment in both cord and adult cells (results not shown), whereas the optimal time for induction of IFN-γ mRNA in both cord and adult cells was determined to be 18 h.

Figure 1
figure 1

IL-12 induction of IFN-γ transcription in cord and adult blood mononuclear cells. Cord and adult BMC were treated with indicated concentrations of IL-12 ranging from 10 to 1000 U/mL for 18 h. Total cytoplasmic RNA was extracted, and RT-PCR assays were performed on the samples using oligonucleotide primers specific for IFN-γ and GAPDH genes. The PCR products were analyzed by ethidium bromide staining of the agarose gel after electrophoresis. The gel shown is representative of six experiments performed. Upper panel, IFN-γ PCR products; lower panel, GAPDH PCR products as controls.

Full size image

It is interesting to note that there was spontaneous synthesis of IFN-γ-specific mRNA in adult cells upon in vitro incubation in the absence of IL-12 incubation (Fig. 1, lane 1). Reasons underlying this spontaneous production of IFN-γ by adult cells are not well understood. This may be due to inherent immunocompetence of the adult cells and their activation by FCS in the culture medium, or due to in vitro incubation with culture plates. In contrast, IFN-γ messages were not detected in cord cells without IL-12 treatment. It therefore appears that IL-12 was required to render the cord BMC competent in synthesizing IFN-γ mRNA.

Effect of IL-12 treatment on IFN -γ protein synthesis. Having determined that IL-12 up-regulates the transcription of IFN-γ mRNA in both cord and adult BMC, we next investigated the levels of IFN-γ protein in the culture supernatants of the cells. With regard to adult BMC, there was spontaneous production of IFN-γ upon in vitro incubation for 18 h without any cytokine treatment, compared with adult cells at 4 h of in vitro incubation[53 ± 8.44(mean ± SE, n = 13) pg/mLversus 11.3 ± 5.1(mean ± SE, n = 4) pg/mL, respectively; p < 0.03 by nonpaired t test]. In contrast, cord BMC did not produce significant amounts of IFN-γ in the absence of IL-12 stimulation.

Similar to mRNA studies, our kinetics experiments demonstrated that 18 h was the optimal time for production of IFN-γ after IL-12 treatment (not shown). With 100 U/mL of IL-12 treatment, both cord and adult BMC had significant induction of IFN-γ (Fig. 2). In cord BMC after 18 h of IL-12 incubation, the level of IFN-γ increased from 8.43 ± 2.26(mean ± SE) pg/mL to 296 ± 97 pg/mL(n = 7; p < 0.05 by paired t test). Similarly, the levels of IFN-γ in adult cell cultures increased from 45.3 ± 1.8(mean ± SE) pg/mL to 290 ± 104 pg/mL(n = 9, p < 0.05 by paired t test) after 18 h of IL-12 treatment. The levels of IFN-γ induced by IL-12 were thus similar in cord and adult BMC. In several experiments, low levels of IFN-γ were detected 4 h after IL-12 stimulation in both adult and cord cells (Fig. 2); however, the increases were statistically not significant.

Figure 2
figure 2

Effect of IL-12 treatment on IFN-γ protein synthesis. Cord and adult BMC (1 × 106/mL) were treated with IL-12 (100 U/mL) in the complete RPMI medium as described. Cells without IL-12 treatment were used as controls. Culture supernatants were sampled at 4 and 18 h after IL-12 treatment and assayed for IFN-γ protein levels by ELISA.

Full size image

Regulation of IFN -γ induction by ionomycin and phorbol ester. Although IL-12 is capable of inducing IFN-γ mRNA in both cord and adult cells, we were interested in examining the effects of ionomycin (a calcium ionophore) and phorbol ester on IFN-γ mRNA induction. Ionomycin and phorbol esters, because of their intrinsic properties in mobilizing calcium, are potent inducers of cytokine gene expression. Cells were treated with IL-12 (100 U/mL) or ionomycin/phorbol myristate acetate (0.5 μM/50 ng/mL) for 18 h, and cells were harvested for Northern analysis. As shown in Figure 3, the combination of ionomycin and phorbol ester was a more potent inducer of IFN-γ transcription than IL-12 alone in both cord and adult cells, after taking into account the effects of uneven loading of the RNA samples as reflected by the GAPDH controls. Previous studies have documented that GAPDH is an important enzyme for normal cellular function, and its steady-state mRNA levels are known to remain constant in most cell types under diverse conditions. In addition, as previously described(24), there appeared to be higher levels of IFN-γ messages inducible in adult BMC, compared with cord BMC, after ionomycin/phorbol ester treatment. This increase in the synthesis of IFN-γ was further confirmed by measuring IFN-γ protein levels in the culture supernatants by ELISA. IFN-γ protein levels in adult cell cultures increased from 41 ± 10(SE) pg/mL to 12,840 ± 3,203(SE) pg/mL (n = 5) whereas the cord cell levels increased from 11.8 ± 1.9(SE) pg/mL to 2,200± 470(SE) pg/mL (n = 5; adult versus cord cells after ionomycin/phorbol ester stimulation, p < 0.012 by paired t test). In contrast, as demonstrated with RT-PCR and ELISA, cord BMC had similar levels of IFN-γ production as adult BMC in response to IL-12 stimulation.

Figure 3
figure 3

Regulation of IFN-γ induction by ionomycin and phorbol ester. Cord and adult BMC (5 × 106) were treated with IL-12 (100 U/mL) or with ionomycin/phorbol myristate acetate (0.5 μM/50 ng/mL) for 18 h. Total cytoplasmic RNA was extracted for agarose gel electrophoresis (10 μg/lane). Northern hybridization was performed on the samples using cDNA probes specific for IFN-γ and GAPDH genes as described in “Methods.” Upper panel, IFN-γ mRNA;lower panel, GAPDH mRNA as controls.

Full size image

IL-12 stimulation of NK cytotoxicity to HIV-infected cells. To examine the cellular responsiveness of cord BMC to IL-12 stimulation, NK cytotoxicity to HIV-infected cells by these cells was examined. Adult BMC were used as controls for comparison with cord cells. BMC were cultured in media containing various concentrations of IL-12, ranging from 5 to 1000 U/mL, for 18 h. BMC were then washed and used in the cytotoxicity assay with51 Cr-labeled HIV-1-infected H9 cells as targets (see“Methods”). As shown in Figure 4, there was a dose-dependent increase in NK activity upon IL-12 treatment of BMC for both cord and adult cells. Statistically significant increases in NK cytotoxicity of BMC from neonates and adults were observed with IL-12 concentration of 100 and 1000 U/mL (all with p < 0.05 by paired t test compared with controls, Fig. 4). For instance, after 100 U/mL IL-12 treatment, the NK activity of cord cells increased from 33.21± 8.22(SE)% to 45.81 ± 7.13% (p < 0.01 by paired t test). At the same concentration of IL-12, NK activity of adult BMC increased from 14.98 ± 22.17(SE)%, to 22.17 ± 3.21% (p < 0.05 by paired t test). In a larger subset of cells from adults, a significant increase was also seen utilizing 10 U/mL of IL-12 (n = 9, p < 0.05 by paired t test). Although NKC of cord BMC was higher than that of adults in control incubations without IL-12, this was not statistically significant. However, NKC of BMC from neonates was significantly higher than that of adults after incubation in 10, 100, or 1000 U/mL of IL-12. In separate experiments, addition of polyclonal anti-IFN-γ antibodies (2-10 g/mL; Endogen) to adult BMC treated with IL-12 (100 U/mL) did not reduce the stimulatory effects of IL-12 on NKC. Thus, it appears that IL-12 induction of NKC in BMC is independent of IFN-γ expression.

Figure 4
figure 4

IL-12 stimulation of NK cytotoxicity to HIV-infected cells. Cord and adult BMC were treated with indicated concentrations of IL-12 ranging from 5 to 1000 U/mL for 18 h. NKC of the BMC to HIV3B-infected H9 cells as targets was performed as described in “Methods.” The effector-to-target ratio was 50:1. Open circle, cord cells;closed circle, adult cells; n represents the number of paired assays at each point.

Full size image

DISCUSSION

The increased susceptibility of the newborn to infection has been attributed to developmental deficiencies of the neonatal host defense system. Among the relative immune defects in neonates are decreases in production of granulocyte macrophage-colony stimulating factor and granulocyte-colony stimulating factor by BMC(25), a relative decrease in TNF-α expression by monocytes(5, 9), diminished synthesis of IgG by cells in response to Staphylococcus aureus stimulation(26), decreases in IFN-γ production by T cells(6–8), and deficiency in antibody-dependent cellular cytotoxicity and NKC(10). NK cells are a subpopulation of lymphocytes that kill virus-infected cells and tumor cells without previous sensitization. It is an important component of the innate immune defense against pathogens. NK cells mediate cytotoxicity against susceptible virus-infected cells by either direct cell-mediated cytotoxicity (NKC) or by antibody-dependent cellular cytotoxicity(27).

IL-12, formerly known as NK stimulatory factor, is a heterodimer consisting of two subunit proteins, p35 and p40(16, 28). Because its original isolation from culture fluids of EBV-transformed human lymphoblastoid cell lines, IL-12 has been shown to be produced by monocytes/macrophages, B cells, and other accessory cells(16, 28, 29). Its level of synthesis is greatly augmented in response to pathogen invasion. IL-12 can up-regulate two antiviral mechanisms, IFN-γ production and NKC mediated by cells from adult humans(16, 17, 24, 30). The relative decrease in IFN-γ production by cord BMC cells in response to mitogens (compared with adult cells) may be due to deficient IFN-γ synthesis because of cellular immaturity with an inability to respond to IFN-γ stimulators such as endogenous IL-12 or due to deficiency of IL-12 production. Here, we showed by RT-PCR assay that cord cells were responsive to IL-12 stimulation at concentrations as low as 10 U/mL (Fig. 1). A previous study demonstrated that IL-12 induced IFN-γ synthesis in cord BMC and CD4 T cells after IL-12 treatment of the neonatal cells for 1 wk. These studies showed that it took IL-12 6 d to induce IFN-γ messages in pooled cord BMC in culture as assayed by Northern blot analysis(24). Because it required more than 24 h for IL-12 to induce IFN-γ in the cord BMC, it is possible that other intermediary cytokines induced by IL-12 may have been involved. In contrast, in our present report, we demonstrated by RT-PCR that IL-12 induced IFN-γ transcription in cord cells within 18 h. RT-PCR is a far more sensitive assay than Northern blot analysis, thus allowing us to detect IFN-γ earlier than that previously reported. Although cord cells required treatment with IL-12 to render them competent in IFN-γ transcription (Fig. 1), adults cells synthesized small amounts of IFN-γ mRNA spontaneously upon in vitro incubation. We have also confirmed these findings with respect to IFN-γ protein synthesis (Fig. 2). The ability of cord cells to respond in a mature fashion to IL-12 resulting in IFN-γ production is in contrast to their response when treated with ionomycin/phorbol ester (Fig. 3) as previously reported(24). Although a potentially important endogenous natural regulator of IFN-γ expression, IL-12 is a weak inducer of IFN-γ when compared with ionomycin/phorbol ester.

Several cytokines can regulate the activation and proliferation of NK cells. Cytokines including IL-1, IL-4, IL-6, and IL-7, although able to activate lymphokine-activated killer cells in long-term culture of BMC, have minimal activity to up-regulate NK cytotoxic activity in short-term cultures of less than 24 h. However, IL-2, IL-12, and IFN are able to directly enhance NKC within a few hours of stimulation(11). IL-12, in particular, has been demonstrated to be a powerful immunomodulator in activating NKC in short-term cultures of adult cells at molar concentrations 100-fold lower than that of the concentration required for IL-2 or IFN-γ(31–33). We have shown that IL-12 induced up-regulation of NK activity in both adult and cord BMC, once again demonstrating that cord cells are competent in eliciting cellular response to IL-12 treatment. As we have recently reported, NKC against HIV-infected cells by cord BMC was comparable to that of the cells from adults(19). This finding contrasts with previous studies which showed deficiencies in neonatal NKC of tumor cell lines and herpes simplex virus-infected cells(10, 34–37). Reasons underlying higher NK activity of neonatal cells against retrovirus-infected cells, compared with herpes simplex virus-infected cells, are not known. Here, we demonstrated that cord BMC showed higher levels of IL-12-stimulated NKC against HIV-infected cells in comparison with adult cells(Fig. 4). Because IL-12 induces IFN-γ production in BMC, it is possible that this activation of NKC is via the induction of IFN-γ. With the addition of anti-IFN-γ antibodies to the BMC culture medium simultaneously with IL-12, there was no changes in the level of NK activity. Thus, the effects of IL-12 on NK cytotoxicity against HIV-infected H9 cells appear to be independent of IFN-γ. This observation is in agreement with recent reports that IL-12 augmentation of adult human NKC and cytolytic T cell activity in a murine model were independent of IFN-γ production(11, 12, 38).

In summary, our findings demonstrate that cord BMC are capable of responding to IL-12 stimulation, competent in synthesizing IFN-γ, and able to mount NKC. Thus, it appears that the deficiency in IFN-γ production or NKC in cord cells is not due to an inherent defect in IL-12 response of the cord cells. We are currently investigating whether cord cells have lower levels of IL-12 transcription or protein synthesis. Understanding the mechanisms underlying the relative immune defects of neonates may provide a novel approach to augment immune response of neonates against life-threatening infections.