INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mammals and aves evolved distinct respiratory and gastrointestinal lymphoid compartments that participate in the defense of associated and distant mucosae. Respiratory lymphoid structures include the bronchus-associated lymphoid tissue (BALT) and lung parenchymal and alveolar cells in the respiratory tract (1). The gut-associated lymphoid tissues (GALT) comprise the Peyer's patches (PP), lamina propria cells (LP), and possibly solitary lymphoid nodules (SLN) in the intestine (2). In humans (3) and certain other species (4), the oropharyngeal lymphoid tissues (Waldeyer's ring), including the adenoid and the bilateral tubule, palatine, and lingual tonsils (7), might also participate in respiratory and gastrointestinal tract defense (8). Although similarities in the structure and function of the BALT and GALT have been explored in humans and animals (13), similar studies of Waldeyer's ring have been restricted largely to humans as structural and/or functional equivalents have not been defined in most animals.

Because the nasal mucosa is the first site of contact with inhaled antigens, it is possible that a nasal analogue of the BALT, GALT, or Waldeyer's ring exists in mammals that contains lymphocytes capable of participating in the defense of respiratory and other mucosal sites. For example, in rats (14), hamsters (15), and mice (16), paired lymphoid structures, termed the nasal-associated lymphoid tissue (NALT), are found at the entrance of the nasopharyngeal duct. Structural and functional evaluation of NALT has been challenging, largely because of difficulties in accessing its anatomic location, although the presence of M-cells in the lymphoepithelium overlying NALT in this site (14), suggests functional similarities to BALT and GALT. However, whether NALT is an analogue of human BALT, GALT, or Waldeyer's ring remains unclear.

We have developed a rapid and precise method for isolation of NALT to compare the composition and immune responsiveness of NALT and PP. Our results show that NALT contains cell populations expected in an immune-inductive site. However, whether the NALT is a BALT or a GALT equivalent remains uncertain; although NALT had the capacity to respond to nasally administered antigen, it contained distinct frequencies and ratios of immune cells when compared with PP.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Female BALB/c mice (Harlen Sprague Dawley Inc., Indianapolis, IN) entered experiments at 6 to 8 wk of age and were allowed food and water ad libitum.

Collection and Preparation of Sera and Cells

Individual blood samples were obtained via the retro-orbital plexus. For sera evaluation, insoluble material was removed by centrifugation, and sera were stored at 70° C until used. Pooled peripheral blood lymphocytes (PBL) were obtained by centrifuging fresh blood, containing 0.1 M ethylenediaminetetraacetic acid, over Ficoll-Paque (Pharmacia, Uppsala, Sweden) (20).

Pooled NALT cell suspensions were prepared using a dissection procedure. Mice were killed by cervical dislocation and decapitated, and the lower jaw and tongues were removed. After rinsing with ice-cold Hanks' balanced salt solution (HBSS) to remove blood, the heads were immobilized with pins on a wax dissection slab to reveal the upper palate. Using a dissection microscope (magnification, 160 diameters; Model M3B; Wild Leitz Canada, Willowdale, ON) and a fiberoptic WMB illuminator (Microlite Inc., Three Rivers, MA), palates were excised as illustrated in Figure 1 using a No. 15 scalpel blade (Lance Blades, Sheffield, UK). After the incisions, palates were gripped behind the incisor teeth with fine forceps and gently pulled toward the molar teeth while using the scalpel to gently free tissue between the palates, jawbones, and nasal septums. Palates were placed immediately into a 24-mm Petri dish containing 1.0 ml of ice-cold HBSS, and the NALT (visible under low angle fiberoptic illumination) was teased gently into the medium to release cells. NALT cell suspensions from individual animals were pooled and cells were washed twice with HBSS by centrifugation (200 × g for 10 min at 4° C) in HBSS and resuspended in RPMI-1640 medium supplemented with 5% heat-inactivated fetal bovine serum, 1% penicillin-streptomycin, 1% L-glutamine, and 1% N-n-hydroxyethylpiperizine-N'-2 ethanesulfonic acid (HEPES) at pH 7.4 (operationally termed RPMI-5).

View larger version (40K): [in this window] [in a new window]   Figure 1.   Anatomic location of nasal-associated lymphoid tissue (NALT) in Mus musculus. Illustrated view of ventral surface of upper palate. I = incisor tooth; L = ligament; M = molar tooth; N = NALT located on the dorsal surface of upper palate within the stippled area. Broken line describes the incision to remove upper palate.

Pooled spleens (SPL) and PP were isolated from decapitated carcasses and placed into ice-cold HBSS. Spleen cell suspensions were obtained by crushing organs between the frosted ends of two microscope slides. Cell suspensions of PP were obtained by stirring the tissues for 15 min in HBSS containing 1.5 mg/ml of Dispase II (Boehringer Mannheim, Mannhein, Germany) at 37° C. The supernatant overlying PP was collected and the remaining tissue was incubated for an additional 15 min in fresh collagenase solution. The supernatant was collected and the remaining tissues were gently teased apart using forceps. The dissociated cell suspension and supernatant were filtered through nylon mesh (BSH Thompson, Scarborough, ON) to remove adipose tissue and large cellular aggregates. Single cell SPL and PP suspensions were washed twice with HBSS by centrifugation. Erythrocytes and dead cells were removed using Ficoll-Paque and lymphocytes were resuspended in RPMI-5. Viability of cell preparations routinely exceeded 90% as judged by ethidium bromide/fluorescein diacetate staining (21, 22).

Electron Microscopy

Tissue specimens were fixed in 2.0% gluteraldehyde in 0.1 M sodium cacodylate for 1 h, rinsed in 0.2 M sodium cacodylate buffer, and postfixed for 1 h in buffered 1.0% OsO₄. Tissues were then dehydrated in a graded series of ethanol. For scanning electron microscopy (SEM), specimens were critical point-dried, sputter-coated with gold, and viewed in a Philips 501B scanning electron microscope. For transmission electron microscopy (TEM), dehydrated specimens were solvent-exchanged in propylene oxide, embedded in Spurr's resin. Sections were strained with uranyl acetate and viewed in a JEOL 1200 EX transmission electron microscope.

Flow Cytometry

Pooled cells from untreated animals were phenotypically characterized using a FACScan flow cytometer and Lysys-II software (Becton Dickinson, San Jose, CA). A minimum of 5,000 cells were analyzed for each phenotypic characteristic. Three-color staining for T-cell subsets was performed using fluorescienated (FITC) anti-CD4 monoclonal antibodies (mAb) (clone GK1.5, prepared in our laboratory (23), phycoerythrin-conjugated (PE-) anti-CD8 mAb (clone 53-5.8; Pharmingen, San Diego, CA) and biotinylated anti-CD8 mAb (clone 53.6-72; Pharmingen) followed by streptavidin-peridinin chlorophyll protein (Becton Dickinson, Sparks, MD). B-cells were detected with a FITC-anti-B220 mAb (clone 6B2, prepared in our laboratory) (24), FITC-goat antimouse IgA (Southern Biotechnology Assoc., Birmingham, AL) and PE-goat antimouse IgM (Southern Biotechnology). Macrophages were identified by two-color staining with PE-antimouse CD11b (clone MAC-1; Pharmingen) and FITC-antimouse IA/E (clone M5) (25).

B- and T-cells were analyzed within a lymphocyte gate defined by forward and side light scatter. An enlarged mononuclear gate was used to define the macrophage population. Equivalent gates were used in analyses of cells derived from different tissues for comparison. Dead cells were excluded from analyses on the basis of light scatter in reference to propridium iodide staining. Background staining was controlled by labeled isotype controls or biotinylated rat IgG (Pharmingen) and never exceeded 1.0% of cells. The results represented the percentage of positively stained cells in the total cell population exceeding the background staining signal.

Immunizations

Groups of six mice each were immunized intragastrically or intranasally on Days 0 and 10. Cholera toxin (CT) was chosen as it is a potent mucosal immunogen (11, 12). Animals received 10 µg of CT intragastrically in 500 µl of 0.2 M NaHCO₃ (to neutralize gastric acidity) or vehicle alone using PE50 tubing (Becton Dickinson) or intranasally in 10 µl of phosphate-buffered saline (PBS) at pH 7.2, distributed between the two nares.

Collection of Mucosal Washings

Mice were exsanguinated by cardiac puncture and CT-specific immunoglobulin IgA and IgG in nasal or intestinal washings were measured. Tracheas were exposed by dissection, ligated with a 3.0 silk suture (Ethicon, Somerville, NJ) and PE-50 polyethylene tubing (Becton Dickinson) was inserted, via the oropharynx, into the nasopharyngeal cavity. Contents of the nasal passages from individual mice were washed out of the nares with 0.5 ml of ice-cold enzyme inhibitor solution (26). Individual gut washings were obtained as previously described (26). Nasal and gut washings were stored at 70° C until used.

Enumeration of CT-specific Antibody-forming Cells

An enzyme-linked spot-forming assay (ELISPOT) was used to detect CT-specific antibody-forming cells (AFCs) in pooled NALT and PP preparations. Duplicate serial dilutions of single cell suspensions (beginning at 1 × 10⁶ cells/100 µl/well) were examined using nitrocellulose microtiter plates (Millititer HA; Millipore Corp., Bedford, MA) previously incubated for 2 h at 37° C with 100 µl/well of CT (100 µg/ ml in PBS), washed with PBS, and treated with 250 µl/well of 0.1% gelatin in PBS. After an 8-h incubation at 37° C, wells were washed three times each with PBS and PBS containing 0.05% Tween-20 (PBS/Tween). The wells were incubated overnight at 4° C with 100 µl of biotinylated goat antimouse IgA, IgG, or IgM (heavy chain-specific; Southern Biotechnology) diluted in PBS/Tween-20 containing 0.1% gelatin (dilution buffer). The wells were washed three times each with PBS and PBS/Tween and incubated for 1 h at room temperature with alkaline phosphatase-conjugated streptavidin (Southern Biotechnology) in dilution buffer. After washing the wells four times with PBS, spots were visualized by incubating with 100 mM NaHCO₃ and 1.0 mM MgCl₂ at pH 9.8 containing 0.15% wt/vol 5-bromo-4-chloro-3-indolylphosphate and 0.3% wt/vol nitroblue tetrazolium. Color development in the wells was halted by thoroughly rinsing the wells with tap water. After drying at room temperature, individual AFCs were enumerated with the aid of a dissecting microscope. The results represent the mean numbers of spots per 1 × 10⁶ cells in duplicate wells containing at least twofold more spots than wells incubated with cells from PBS-treated animals.

Measurement of CT-specific Antibody Responses

An enzyme-linked immunosorbant assay (ELISA) was used to measure CT-specific antibodies in individual nasal and gut washings and sera as previously described (26). After the addition of halving or one-third dilutions of sera or mucosal washings to CT-coated microtiter plates, anti-CT antibodies were quantified using alkaline phosphatase-conjugated goat antimouse IgG1, IgG2a, or IgA (heavy chain-specific; Southern Biotechnology) and 1.0 M diethanolamine buffer at pH 9.8 containing 50 mM MgCl₂ and 1.0 mg/ml p-nitrophenylphosphate. The results were expressed as reciprocal end-point titers representing the greatest sera dilutions giving optical density values exceeding two times normal mouse sera mean values.

Statistical Analyses

Statistical analyses were done using GraphPad Prizm^® Version 2.00 (GraphPad Software, San Diego, CA). One-way ANOVA, Tukey's test, pair-wise multiple comparisons, and Student's unpaired t tests were used to detect and compare mean differences between treatment groups. A level of significance of 95% was chosen for all tests.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolation of NALT

It can be seen in Figure 2 that after microdissection, paired NALT aggregates were clearly visible on the palate, near the entrance of the nasopharyngeal duct. NALT cells were readily isolated, yielding approximately 1 × 10⁶ viable cells per palate. Because most of the overlying epithelium remained attached to the nasopharynx, few ciliated cells were noted either by SEM (Figure 2), in cell suspensions as assessed by light microscopy, or by TEM of cell pellets (data not shown). Further, TEM analyses of NALT cells did not reveal the presence of any apoptotic lymphocyte nuclei (data not shown). Contamination of NALT cell preparations by erythrocytes was judged to be negligible. These results indicated that an adequate number of NALT cells could be attained for study.

View larger version (166K): [in this window] [in a new window]   Figure 2.   Scanning electron micrographs (SEM) of nasal-associated lymphoid tissue (NALT) in Mus musculus. (A) View of the dorsal surface of upper palate revealing NALT in correspondence to stippled areas in Figure 1. (B) Magnified left NALT structure in (A) rotated and viewed at a 60-degree angle. (C ) Magnification of (B) near center of NALT structure.

Flow Cytometry

T-cell composition of NALT. The similarity of NALT to other mucosal and systemic lymphoid tissues was assessed (Table 1). NALT isolates contained greater proportion of T-cells than PP (50 versus 30%, respectively), of which a greater percentage were CD4+ as compared with CD8+ T-cells (p < 0.0005). The CD4:CD8 cell ratio in the NALT (3.7:1) was closer to that noted in the spleen (2.7:1), yet it was less than that observed in PP and PBL isolates (10:1 and 30:1, respectively). Similar numbers of CD8+ T-cells, a unique subset common to intestinal epithelia (27), were detected in PP and PBL. Interestingly, no significant numbers of CD8+ T-cells were found in NALT cell isolates, in contrast to PP (p < 0.001) and similar to that seen in SPL cell isolates. Thus, if viewed on the basis of overall T-cell frequency and phenotype, NALT T-cell populations most closely resembled SPL isolates rather than those found in PP preparations.

B-cell composition of NALT. Although NALT cell preparations contained greater numbers of T-cells than PP, it can be seen in Table 1 that the total B-cell frequency of NALT was 15% lower than that found in PP (p < 0.05). However, NALT and PP-cell populations contained similar numbers of mature, IgM+IgA B-cells, which were significantly greater than that observed in SPL cell isolates (p < 0.01 and p < 0.001 compared with NALT and PP, respectively). Cells coexpressing surface IgA and IgM were detected at low but similar frequencies in NALT, PP, and SPL cell preparations. These cells likely represented activated B-cell types containing both IgM and IgA mRNA, which have yet to undergo gene rearrangement preceding the exclusive expression of surface IgA. Low frequencies of IgMIgA+ B-cells (committed entirely to IgA production) were detected in PP and PBL, but neither NALT nor SPL isolates had detectable levels of these cells. These results indicated that although NALT contained both normal mature B-cells and cells undergoing expression of surface IgA, unlike PP, NALT did not contain IgA switched cells. Thus, the normal murine NALT might not be as active as PP if viewed in terms of the differentiation and expansion of IgA-committed B-cell populations.

Macrophage composition of NALT. Macrophage frequencies in NALT and other tissues were identified via cell surface staining with CD11b (Mac-1)-specific and Ia (MHC class-II)- specific antibodies. As shown in Table 1, small but detectable frequencies of macrophages were observed in NALT and SPL cell isolates that were significantly higher than those in PP or PBL cell isolates (p < 0.05). The close resemblance of NALT macrophage and T-cell content to that of SPL might have a functional bearing on development of cell-mediated immune responses in NALT.

Detection of Antigen-specific Antibody-forming Cells in NALT and PP after Intranasal or Intragastric Immunization

Because Table 1 indicated that NALT and PP contained disparate frequencies of various cell populations, a direct comparison of the functional characteristics of these mucosae-associated tissues was conducted. Groups of mice were immunized intranasally or intragastrically with CT, and NALT- and PP- derived cell populations were subsequently examined for the presence of CT-specific AFCs. It can be seen in Table 2 that after two intranasal or intragastric immunizations with CT, IgA producing CT-specific AFCs were detected in the local lymphoid compartments. Intranasal immunization with CT resulted in appreciable number of CT-specific, IgA-producing AFCs in the NALT. However, intragastric immunization using the same dose of CT resulted in almost 3-fold more CT-specific IgA-producing AFCs in isolated PP. During this short period of time, there was no evidence that anti-CT AFCs appeared at distant mucosal sites, as intragastric or intranasal administration of CT did not result in anti-CT AFCs in NALT or PP tissue, respectively. Additionally, whereas intragastric immunization with CT elicited solely an IgA plasmacyte response in PP, antigen-specific plasma cells producing IgG and IgM, in addition to IgA, were detected in the NALT after intranasal immunization with the same dose of CT (Table 2). These results demonstrated that stimulation of antigen-specific nasally derived B-cells is best accomplished by intranasal immunization.

Antigen-specific Antibody Responses after Intranasal or Intragastric Immunization

Sera and mucosal secretions from groups of animals immunized intranasally or intragastrically with CT were examined for the presence of CT-specific antibodies. It can be seen in Table 3 that 4 d after secondary intranasal or intragastric immunization with CT, anti-CT IgG1 sera titres were comparable between both groups of animals. Sera anti-CT IgG2a responses were, however, 6-fold greater in animals receiving CT intranasally. Additionally, whereas CT-specific IgA was not detected in the sera of intranasally or intragastrically immunized animals, CT-specific IgA was detected in mucosal washes of animals immunized intranasally or intragastrically (Figure 3). Specifically, intragastric CT administration elicited a robust CT-specific mucosal IgA response in intestinal, but not nasal secretions (p < 0.05), whereas intranasal immunization stimulated nasal- but not intestinal-specific IgA production (p = 0.09). Thus, similar to the detection of CT-specific AFCs in NALT and PP (Table 2), administering CT at one mucosal site stimulated local mucosal IgA production but failed to provoke detectable levels of CT-specific IgA in mucosal secretions at distant sites. The lack of detectable levels of CT-specific sera IgA after intranasal or intragastric CT administration, together with the presence of antigen-specific mucosal IgA and CT-specific IgA AFCs detected in either mucosal lymphoid compartment (Table 2) suggested local production of CT-specific mucosal antibodies.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Cholera toxin (CT)-specific IgA in mucosal secretions after intranasal or intragastric immunization with CT. Groups of six mice were immunized intranasally (solid bars) or intragastrically (hatched bars) on Days 0 and 10 with 10 µg of CT in phosphate-buffered saline (PBS) or 0.2 M NaHCO3, respectively. On Day 14 animals were killed and nasal (A) and intestinal (B) washes were evaluated for the presence of CT-specific IgA using an enzyme immunoassay. Results are expressed as mean reciprocal end-point titers (± SEM) representing the greatest sera dilutions giving optical density values exceeding two times normal mouse control sera mean values.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The anatomic location of the NALT (Figure 1) has precluded its precise removal and analysis. Inexact methods have been employed to isolate NALT cells (18, 19), allowing for potential contamination from the adjacent nasal mucosa, lamina propria, and vasculature. We overcame this difficulty by developing an accurate dissection method to removal NALT tissue. NALT cell preparations were judged to be free of unwanted cell contamination as a result of careful dissection. This was corroborated by disparate frequencies of various cell populations observed in NALT and PBL cell preparations (Table 1).

We showed that, when compared with systemic (SPL, PBL) and mucosal (PP) cell preparations, NALT cell isolates from unimmunized mice contained disparate frequencies and phenotypes of immune cell populations. Although NALT and PP cell preparations contained similar numbers of IgM+IgA and IgM+IgA+ B-cells, when compared with PP cell isolates, NALT-derived B-cells comprised a smaller proportion of total lymphocytes (p < 0.05). Additionally, unlike PP preparations, NALT isolates contained no IgA-switched (IgMIgA+) B-cells (p < 0.05). In unpublished studies, we were also unable to detect IgA-committed B-cells in NALT cells isolated from various ages of naive mice (6 to 20 wk of age). This suggested that, even in the presence of environmental antigens, murine NALT might not be as active as PP, possibly reflecting exposure to a lower antigenic load.

NALT and PP cell preparations also contained distinct T-cell and macrophage populations. Compared with PP, NALT cell isolates contained higher frequencies of T-cells and macrophages and displayed a lower CD4:CD8 T-cell ratio (10:1 versus 3.7:1, respectively). Indeed, the frequencies and ratios of CD4+ and CD8+ cells in NALT preparations most closely resembled spleen cell isolates. Also, similar to SPL cell preparations, NALT cell isolates contained no significant numbers of CD8+ T-cells. This is in contrast to PP, which contained small, but significantly higher numbers (p < 0.001) of CD8+ T-cells, which highlights another dissimilarity between these mucosal-associated lymphoid structures. The relatively high frequency of CD8+ T-cells also observed in unstimulated NALT cell isolates, compared with PP (p < 0.05), might reflect a greater contribution of cytotoxic T-lymphocytes or CD8+ regulatory cells to NALT tissue. Cells with a macrophage phenotype (CD11b+Ia+) in NALT and SPL cell isolates were significantly higher in frequency than those in PP (p < 0.05 and p < 0.001, respectively), possibly reflecting a greater contribution of cell-mediated immune responses in NALT tissue. Indeed, after intranasal influenza virus immunization, Tamura and colleagues (28) observed cellular immune responses in crude NALT preparations that correlated with intranasal viral clearance. Thus, if viewed solely on the basis of immune cell content and ratio, NALT bears a greater resemblance to SPL than to PP.

Although normal NALT cell isolates contained few IgA-secreting B-cells (data not shown), after intranasal immunization with CT, CT-specific IgA AFCs were readily detected in NALT tissues. This suggested that B-cell isotype switching and differentiation occurs in the NALT after direct antigenic stimulation. Also, PP cells displayed increased numbers of CT-specific IgA AFCs after intragastric CT administration, albeit at higher levels than observed in NALT cell isolates. After intragastric or intranasal CT immunization, there was no evidence that CT-specific AFCs dispersed to the NALT or PP, respectively. These findings question whether stimulated GALT-derived lymphocytes migrate to the nasal mucosa and suggest that stimulation of NALT is best achieved by nasal antigen administration.

Despite detecting CT-specific IgA AFCs in NALT isolates after intranasal immunization, antigen-specific IgA was not observed in the sera of these animals. However, toxin-specific IgA was detected in nasal, but not in gut, washes after intranasal CT administration (p = 0.09). These results suggest that intranasal but not intragastric, immunization stimulates local antigen-specific mucosal IgA production via the selective stimulation of IgA-committed NALT-derived B-cells. Similarly, intragastric CT administration elicited a robust CT-specific mucosal IgA response in intestinal but not in nasal secretions (p < 0.05), suggesting regionalized activation of the mucosal immune system. Indeed, Husband and Gowans (29) observed the selective migration of IgA plasma cell precursors to the intestinal lamina propria, but not PP, after the adoptive transfer of mesenteric lymph node cells. This homing highlights the potential compartmentalization of the different tissues comprising the common mucosal immune system (13) and suggests that local immunization at one mucosal site might best stimulate the production of local mucosal immunity.

Although intranasally immunized animals failed to generate detectable CT-specific sera IgA responses, intranasal CT administration stimulated robust antigen-specific sera IgG1 and IgG2a responses. In contrast, intragastrically immunized mice elicited chiefly an IgG1 antigen-specific sera response. These results suggest that different types of helper T (Th) cell activity may be dependent upon the route of inoculation. Th cells are classified into two subsets depending on their cytokine profile. Th1 cell clones exclusively produce IL-2, IFN-, and lymphotoxin and help to generate IgG2a responses, whereas Th2 cells synthesize IL-4, IL-5, IL-6, and IL-10 and provide help in mounting IgA, IgE, and IgG1 responses (30, 31). The present work suggests that both Th1 and Th2 cells in the NALT were likely responsible for driving the sera antibody responses observed after intranasal CT administration. Increased frequencies of CD4+ and/or CD8+ T cells in NALT isolates, compared with PP, might be a source of Th1 cytokines capable of promoting antigen-specific IgG2a responses as Tamura and colleagues (28) detected IFN- production from crude preparations of nasally derived CD4+ T-cells.

NALT has been studied primarily in rat and mouse models (15). Immunohistochemical characterization of rat NALT has shown that B- and T-cells are distributed in distinct areas with a high CD4:CD8 T-cell ratio and a predominance of B- over T-cells (32). Although murine NALT has not yet been well described immunohistologically, our initial studies suggest that in mice, NALT is distinct from that found in rats and, if examined solely on immune cell content and subset ratios, more closely resembles spleen and not PP. Nevertheless, the capability of NALT to elicit specific IgA responses locally suggest that this structure might represent a unique mucosal lymphoid tissue that is capable of expressing both mucosal and systemic immune responses.

Despite growing interest in using the intranasal route for mucosal vaccine delivery, little attention has addressed the local site(s) and mechanism(s) that might be responsible for the induction of such responses. In humans, the participation of oropharyngeal lymphoid tissues (Waldeyer's ring) in intranasally immune responsiveness is unclear. However, Ogra (33) reported that combined tonsillectomy and adenoidectomy in children resulted in diminished polio-virus-specific antibody levels in nasopharyngeal secretions. In agreement with these studies, preliminary studies in our laboratory show that, in young BALB/c mice (less than 9 wk of age), the phenotypes and frequencies of NALT cells are comparable to those seen in adult animals (11 to 20 wk of age) demonstrating the maturity and possible immune responsiveness of NALT even in young mice (data not shown). Despite a paucity of studies, it is clear that rodent NALT also has clear potential as an immunocompetant inductive site and might represent a functional equivalent of Waldeyer's ring in humans. Investigations of the types of precursor cell populations that can be stimulated in the murine NALT structure are now needed. Studies in our laboratory are currently underway to investigate the interaction of cells within this tissue with a model protein antigen.