Prolonged deprivation of dietary nucleotides appears to supress the immune response. Mice fed a nucleotide-free diet for more than two weeks demonstrate decreased cellular and humoral immunity (Jyonouchi 1994, Rudolph et al. 1990, Van Buren et al. 1994). These mice easily succumb to death caused by bacterial and fungal pathogens (Adjei et al. 1992, Fanslow et al. 1990, Kulkarni et al. 1986). Nevertheless, the mechanisms of nucleotide action on the immune system are not well understood, which makes it difficult to understand the role of dietary nucleotides.

The immune system protects an individual by developing antigen (Ag)³-specific immune responses. Previously we analyzed nucleotide action on Ag-mediated immune responses. We have shown in vivo and in vitro that nucleotides augment antibody (Ab) production against T-dependent Ag in mice (Jyonouchi et al. 1992, 1993a, 1994). This enhancement has been demonstrated in several normal strains of mice, and the degree of enhancement was independent of major histocompatibility complex (MHC) restriction (Jyonouchi et al. 1993b) and was largely attributed to polynucleotides (Jyonouchi 1994, Jyonouchi et al. 1992, 1995b). Our results indicate that polynucleotides augment T-dependent Ab production by modulating processes of Ag-mediated T-helper (Th) cell activation (Jyonouchi, et al. 1993a, 1995b). The actions of polynucleotides likely occur at the local site of inflammation when injured cells release polynucleotides. Polynucleotides may augment activation of naive Th cells and amplify local immune responses.

When Ag are presented by Ag-presenting cells (APC), naive Th cells differentiate into at least three types of effector-stage Th cells: type 0, type 1 and type 2 Th cells (Th0, Th1 and Th2, respectively), differentiated on the basis of cytokine production pattern (Romagnani 1994, Seder and Paul 1994, Swain et al. 1991). The effector-stage Th cells are more easily activated than resting Th cells (Croft 1994) and are much more active in cytokine production and other Th cell activities. If polynucleotides enhance effector-stage Th cells nonspecifically, this could lead to potentially hazardous, hyperactive immune responses. However, we found that nucleotides do not augment polyclonal B cell activation or T cell responses potentiated by polyclonal stimulants (Jyonouchi et al. 1992, 1993b, 1994a). We thus hypothesize that polynucleotide effects differ depending on the stage of Th cell activation. In this study we examined polynucleotide action on effector-stage Th1 and Th2 cell clones.

The use of cloned Th cell lines enables us to examine the effects of polynucleotides in each Th cell subset in a quantitative manner. Fewer than 0.3% of Th cells isolated from regional draining lymph node cells following in vivo Ag challenge are Ag-specific, and those consist of mixed populations of Th0, Th1, and Th2 cells (Miller 1991, Seder and Paul 1994, Swain 1991). In contrast, the two cloned Th cell lines used in this study were each developed from a single primed Th cell, are Ag-specific, and each produces a large amount of cytokines characteristic a Th subset (Paul and Seder 1994 and Swain et al. 1991). The use of cloned Th1 and Th2 cells permits us to greatly increase the proportion of Ag-specific Th cells in the assay as compared with primary cultures of Ag-primed cells. This system could be used a model for elucidating polynucleotide action at the molecular level.

MATERIALS AND METHODS

5

ELISPOT assay for IgM and IgG antibody-secreting cells. Numbers of Ag-secreting cells were measured by enzyme-linked immunospot (ELISPOT) assay (Sedgwick et al. 1983). Cloned Th1 cells were incubated with CD4 syngeneic spleen cells; A.E7 cells (2 × 10⁸ cells/L) with CD4 B10.A spleen cells (1.25 × 10⁹ cells/L), or CDC35 cells (2.5 × 10⁷ cells/L) with CD4 DBA/2 spleen cells (2 × 10⁹ cells/L). This ratio of APC to Th clones produced the optimal results in this assay. Cells were incubated for 5 d in the same medium used for maintaining Th clones in a 24-well tissue culture plate (Costar, Cambridge, MA). The cells were harvested, counted, and resuspended in Iscove's Modified Dulbecco's Medium supplemented per liter with 50 mL HI-FCS (HyClone), 10⁵ U penicillin G, 100 mg streptomycin and 2 mmol glutamine. The harvested cell mixture (100 µL/well) was incubated overnight at 37°C in a 5% CO₂ incubator in a 96-well microtiter plate (Costar) coated with Ag (pigeon cytochrome C; 5 mmol/L) and preincubated with a blocking buffer (PBS, pH 7.4 with 10 g bovine serum albumin/L) for 30 min at 37°C. Cells were removed the next day by vigorous washing, and goat anti-mouse IgG or IgM antibody-alkaline phosphatase conjugate (1:1000 in dilution buffer used for ELISA; 100 µL/well, Sigma) was added to each well. The plate was incubated overnight at 4°C and washed again. Ab secreted by cells were detected by adding gel substrate solution [5-bromo-4-chloro-3-indolyl phosphate (75 mg/L), and nitroblue tetrazolium chloride (150 mg/L) in 50 mmol/L NaHCO₃, pH 9.8 with MgCl₂ (5 mmol/L) and agarose (5 g/L, agarose Type 1-A; low EEO, Sigma), 100 µL/well] and incubating the plate at 37°C for 4-5 h. Triplicates were tested for each sample, and results are expressed as mean number per 10⁶ viable cells based on the cell count after 5 days' culture. This culture period provided the optimal responses and cell viability declined when cells were cultured more than 6 days. When CD4 spleen cells were not reconstituted by Th clones, no ELISPOT was detected (this served as a negative control). The cell mixture cultured without Ag stimuli also served as control for ELISPOT assay.

ELISA for IFN and IL-5.

RESULTS

Experiment 2: Polynucleotide action on IFN and IL-5 production by the cloned Th cells.

IL-5 production by CDC35 cells was not affected by polynucleotides when unprimed spleen cells were used (data not shown). With the use of in vivo Ag-primed spleen cells, less IL-5 was produced with polynucleotide supplementation than in controls at rabbit -globulin concentrations of 2.5 and 5 mg/L (Fig. 7). This suppressive action of polynucleotides appeared to be concentration-dependent at rabbit -globulin concentrations of 2.5 mg/L (Fig. 8) and 5 mg/L (data not shown). This action of polynucleotides was observed only when they were present from day 0 of the culture (data not shown).

DISCUSSION

Nucleotides also exert more specific actions on the immune system. We have shown that nucleotides preferentially affect T cell-mediated humoral immunity both in vivo and in vitro (Jyonouchi 1994, Jyonouchi, et al. 1992, 1993a, 1994). This enhancing action in vitro is attributed to polynucleotides but not to mononucleotides or nucleotide metabolites, while oral supplementation of both polynucleotides and a mononucleotide mixture prevented decline of humoral immunity and Ag-mediated cytokine production in mice fed a nucleotide-free diet (Jyonouchi et al. 1992, 1994, 1995b, 1996). Polynucleotides must be present from day 0 of the culture in order to fully exert their actions (Jyonouchi et al. 1993a). They may modify Ag-mediated T cell activation in the initial stages of Ag presentation. We hypothesized that such polynucleotide action can take place in vivo, most likely at the site of inflammation or tissue injury. That is, the injured tissue could release polynucleotides which may temporarily reach a high local concentration and modulate local immune responses. In mice, deprivation of dietary nucleotides leads to the preservation of DNA at the expense of cellular RNA and proteins, resulting in depletion of the tissue nucleotide pool (Leleiko et al. 1987). Under such conditions, injured tissues release less nucleotides, which may result in attenuated Th cell activation. Dietary nucleotides may be important in maintaining tissue nucleotide pools that may initiate effective local immune responses in certain circumstances.

Naive Th cells generally require more potent signals for activation (Croft 1994), and thus the enhancing action of polynucleotides may promote primary immune responses. Once activated, naive Th cells differentiate into either Th0, Th1, or Th2 effector-stage Th cells (Paul and Seder 1994, Swain et al 1991). Th0 cells produce a variety of cytokines, but Th1 and Th2 cells have a limited pattern of cytokine production (Th1 cells; IL-2 and IFN, Th2 cells; IL-4, IL-5, IL-10, and IL-13). Th1 cells are vital in phagocytic cell-mediated, cellular and proinflammatory immune responses, while Th2 cells promote B cell-mediated humoral immune responses and down-regulate inflammatory responses (Garside and Mowat 1995, Paul and Seder 1994). The effector-stage Th1 and Th2 cells produce 100-fold more cytokines than do naive Th cells and are activated more easily and swiftly upon the exposure to the same Ag (recall Ag) than naive Th cells (Croft 1994, Swain et al. 1991). If polynucleotides augment Th cell activation independent of differentiation stages of Th cells, polynucleotides could be hazardous due to induction of hyperactive immune responses. This study was designed to evaluate the actions of polynucleotides on activated, effector stage Th cells using cloned Th cells. The use of cloned Th cells also permitted us to study nucleotide action in Th1 and Th2 cell subsets.

Our results demonstrated that the actions of the polynucleotides differ between effector Th cells and naive Th cells and even in each Th cell subset. We found that polynucleotides suppress Th1 clone-mediated Ab production when Ab production was assessed by the number of Ab-secreting cells. This suppressive action was shown with the use of in vivo Ag-primed spleen cells and appeared to be dose-dependent. Polynucleotides did not augment Th2 clone-mediated Ab production. With suboptimal Ag challenges, polynucleotides augmented IFN production by the Th1 clone but suppressed IL-5 production by the Th2 clone in a dose-dependent manner. With optimal Ag challenges, no significant effects of polynucleotides were observed.

Activation of Th cells generally requires at least two signals, one through T cell receptor mediated by Ag and the other from costimulatory molecules (Matzinger and Fuchs 1996). The activation signals for Th cells differ among naive and effector-stage Th cells (Croft 1994). Effector-stage Th cells require fewer stimulatory signals, and when strong signals are provided through T cell Ag receptors by a large dose of Ag, effector-stage Th cells may not require many additional signals through costimulatory molecules. Polynucleotide immunomodulation was not observed with optimal Ag challenges in Th1 and Th2 clones. Thus we speculate that polynucleotides may affect Th cell action by modulating signals of costimulatory molecules.

The effects of polynucleotides supplemented to the cultures were evident with the use of in vivo Ag-primed spleen cells but not with unprimed spleen cells in the assays of Th clone-mediated Ab production and cytokine production. In primed spleen cells, the numbers of activated APC and Ag-specific B cells were higher than in unprimed spleen cells (Bradley et al. 1993). Consequently, primed spleen cells likely stimulate Th clones more effectively due to the presence of larger numbers of activated APC than unprimed spleen cells. They also likely produce a higher number of Ab-secreting cells in the ELISPOT assay secondary to elevated numbers of Ag-specific B cells. As a consequence the effects of polynucleotides may have been more dramatic with primed spleen cells than with unprimed spleen cells. In preliminary studies, we examined numbers of macrophages, B cells, and CD4,⁺ and CD8⁺ T cells following in vivo Ag challenges. The frequency of B cells and CD4⁺ T cells increased by 5-10% in this system but the differences were not statistically significant (unpublished observations). We now plan to investigate changes in activation marker expression for B cells and macrophages in response to in vivo Ag challenge (cytochrome C and rabbit -globulin).

In contrast to the suppressive action of polynucleotides on Th1 clone-mediated Ab production and IL-5 production by Th2 cells, polynucloetides augmented IFN production by Th1 cells. This enhancing action of polynucleotides may be attributed to IFN production induced by double stranded (ds) RNA contained in the yeast RNA preparations used as the source of polynucleotides (Hubbell et al. 1991, Mémet et al. 1991, Tiwari et al. 1987). However, polynucleotides enhanced IFN production with suboptimal but not optimal Ag challenges and did not augment IFN production in the absence of Ag. The amount of dsRNA used by others ranged 0.05 to 0.2 g/L (Bourgeade et al. 1993, Decker 1992, Hubbell et al. 1991, Mémet et al. 1991, Tiwari et al. 1987). Such large amounts of dsRNA were unlikely to be provided in the RNA preparations used here, since the highest dose of RNA employed in this study was 0.1 g/L, and the RNA was already degraded to  1,000 nucleotide lengths of predominantly single stranded polynucleotides (Jyonouchi et al. 1992). Therefore, enhancing actions of polynucleotides on IFN production by Th1 cells are unlikely to be attributed to the actions of dsRNA. It may be that polynucleotides enhance IFN production by Th1 cells by modulating the processes of Ag-mediated Th cell activation; further studies are needed to prove this assumption. IFN suppresses humoral immune responses (Paul and Seder 1994). Enhancing actions of polynucleotides on IFN production may partly account for the decrease in Th1 clone-mediated Ab production with suboptimal Ag challenges.

In general, most pathogens induce mixed Th1 and Th2 responses. When Ag concentrations are suboptimal, polynucleotides released from the injured tissue may shift local immune responses to Th1 dominant cellular immune responses by enhancing IFN production and suppressing IL-5 production. Th1 responses are generally more effective in removing intracellular pathogens such as bacteria and viruses (Biron 1994, Garside and Mowat 1995). It will be interesting to examine the local immune responses in animals challenged with Ag or pathogens.

In summary, polynucleotides neither augmented nor suppressed functions of effector-stage Th cells nonspecifically. Our results demonstrate the complex actions of polynucleotides on Th cell functions, depending on the stage of Th cell activation. Thus they are unlikely to induce hyperactive, uncontrolled immune responses.