Isotype-specific immunoregulation. IgA-binding factors produced by Fc alpha receptor-positive T cell hybridomas regulate IgA responses.

T-T hybridomas, produced by fusions between R1.1 T lymphoma and cloned T helper cells that promote IgA responses (Th A cells) were characterized in this study. A total of 85 cloned cell lines were produced, and their supernatants were assessed for support of antigen-dependent IgA (and IgM and IgG) responses. 16 of 85 culture fractions supported IgA anti-sheep red blood cell, -horse red blood cell, or -trinitrophenyl responses in either lipopolysaccharide-triggered splenic B cell, or normal Peyer's patch B cell cultures, and the responses were specific for the antigen used for in vitro immunization. None of the supernatants from the cell lines induced significant polyclonal responses in these B cell cultures. Interestingly, the 16 hybridomas that produced supernatants with IgA-promoting properties had Fc receptors for IgA (Fc alpha R), but did not express Fc mu R or Fc gamma R. When supernatants from Fc alpha R+ T cell lines were subjected to IgA affinity chromatography, the IgA-promoting activity bound to IgA (IBF alpha) and was recovered in the eluate. No binding of active fractions occurred when supernates were passed through IgM or IgG immunoadsorbent columns. High concentrations of purified IBF alpha suppressed T-dependent IgA responses, while an optimal level was required for enhancement of this isotype response. These results suggest that Fc alpha R+ hybridomas derived from Th A cells release IBF alpha into the culture medium, and that these molecules regulate IgA responses to various T-dependent antigens.

Separation of two populations of cells with gamma-glutamyl transpeptidase from carcinogen-treated rat liver.

Noninbred Sprague-Dawley rats were maintained on a choline-deficient diet containing 0.05% ethionine. After 10-13 weeks, livers were dispersed with collagenase, lysozyme, collagenase and hyaluronidase. Pronase, or a selected batch of trypsin. The highest yield of cells with histochemically demonstrable gamma-glutamyl transpeptidase (GGT) was obtained with trypsin. After velocity sedimentation in an isokinetic gradient of Ficoll in tissue culture medium, two modal populations of cells with histochemically demonstrable GGT were observed. The first mode contained cells that were morphologically different from hepatocytes and that may be oval cells. The second, more rapidly sedimenting modal population of cells with GGT was morphologically similar to hepatocytes as assessed with Wright's stain; the location of this population in the gradient was the same as the location of cells with the appearance of hepatocytes that lacked iron and that had decreased glucose 6-phosphatase. In multiple experiments, the purest fractions contained 71.7 +/- 3.5% cells (mean +/- SD) with the appearance of hepatocytes with histochemically demonstrable GGT.

Eosinophil infiltration of human colonic carcinomas as a prognostic indicator.

In experiments with suspensions of cells from colonic carcinomas, we noted that some colonic carcinomas contain large numbers of eosinophils. We therefore carried out a prospective study with 67 almost consecutive colonic carcinomas in our medical center after optimal fixation and staining for the demonstration of eosinophils. Infiltration of the primary tumor by eosinophils was found to have marked prognostic significance. The proportion (4 of 17 or 23.5%) of carcinomas with more than 30 eosinophils/sq mm that had metastases was significantly less (p = 0.01) than the proportion (31 of 50 or 62.0%) of carcinomas with less than 30 eosinophils/sq mm that had metastases. At 18 months, following the resection of tumor in patients without metastases, all of the patients (9 of 9) with greater than 30 eosinophils/sq mm were alive in contrast to 73.7% (11 of 15) of the patients with less than 30 eosinophils/sq mm. The number of survivors at 18 months for the total population without regard to metastases was significantly greater (p = 0.028) for those with greater than 30 eosinophils/sq mm than for those with less than 30 eosinophils/sq mm. We conclude that the quantitative assessment of eosinophils is one of the most important aspects of the microscopic evaluation of this common human tumor.

Heterogeneity and prognostic significance of macrophages in human colonic carcinomas.

Previously, we reported that high concentrations of eosinophils in human colonic carcinomas are associated with better prognoses, that sections taken 1 cm remote from (deep to) the margin of tumor (SRM) and sections contiguous to the margin (SCM) of tumor and adjacent uninvolved colon contain significantly different concentrations of eosinophils, and that concentrations of eosinophils in SCM and SRM are both useful and complementary for the prediction of prognosis. As a first step towards studying the ecology of the eosinophil in colonic carcinoma and with the goal of identifying other kinds of cells that might be useful for the prediction of prognosis, we counted cells in SCM and SRM that expressed histochemically demonstrable acid phosphatase, alpha-naphthylbutyrate esterase, and peroxidase. The tumors of patients with and without metastases at the time of resection of the primary tumor contained different (P = 0.0314) concentrations of cells with histochemically demonstrable alpha-naphthylbutyrate esterase in SCM but not in SRM. In contiguous 1- to 2-micron sections, morphologically macrophage-like cells with histochemically demonstrable acid phosphatase and cells with histochemically demonstrable alpha-naphthylbutyrate esterase were found to be present in different concentrations both in SCM (P less than 0.01) and in SRM (P less than 0.01); i.e., these phenotypic markers appear to identify different subpopulations of macrophages in tumors. In contrast to our previous study of human pulmonary alveolar macrophages, examination of sections stained sequentially for these phenotypic markers that are commonly used for the identification of macrophages in tumors revealed numerous cells in the same sections that expressed histochemically demonstrable acid phosphatase (red) but not alpha-naphthyl butyrate esterase (brown) and vice versa. Several of these markers promise to be useful and complementary for the prediction of prognosis.

Separation and characterization of the neoplastic and stromal elements of the R3230AC mammary adenocarcinoma.

The purification of the various types of cells in solid tumors is necessary for study of their biochemical and immunological functions. The R3230AC adenocarcinoma contains both a variety of stromal cells and well-differentiated neoplastic cells that possess many of the biosynthetic capabilities of normal rat mammary cells. Suspensions of cells from tumors weighing less than 1.2 g consisted of 70.4 +/- 7.2% (S.D.) malignant cells by morphology, 6.8 +/- 3.0% lymphocytes, 6.3 +/- 3.2% macrophages, 15.2 +/- 5.2% red blood cells, 0.8 +/- 0.5% granulocytes, and 0.5 +/- 0.6% unidentified cells. Sedimentation of the cells from the R3230AC tumor in a previously described isokinetic gradient resulted in a 5- to 6-fold purification of lymphocytes and macrophages. The modal fraction of malignant cells contained 95.3 +/- 2.9% malignant cells. Detection of alpha-lactalbumin by the direct peroxidase conjugate technique gave vacuolar staining of malignant cells, in contrast to the indirect and peroxidase-antiperoxidase techniques which stained ducts from normal lactating mammary gland and a wide variety of cells without vacuoles in the tumor. The best fixative for frozen sections, paraffin-embedded tissue, and cell suspensions was 50% ethanol-50% acetone. The suspensions of tumor cells contained 14.4 +/- 9.4% cells with histochemically demonstrable alpha-lactalbumin. Squamous metaplasia was commonly observed in tumors than lactating rats.

Large-capacity separation of malignant cells and lymphocytes from the Furth mast cell tumor in a reorienting zonal rotor.

The Furth murine mastocytoma was adapted to the ascitic form and separated into fractions enriched with respect to lymphocytes and malignant cells by velocity sedimentation in the SZ-14 reorienting zonal rotor or in the isokinetic gradient. Lymphocytes were more highly purified (p less than 0.01) in the isokinetic gradient than in the zonal rotor, i.e., lymphocytes comprised 99.1% of the nucleated cells in the purest fraction from the isokinetic gradient and 80.1% of the nucleated cells in the purest fraction from the zonal rotor. Neoplastic mast cells were similarly purified by the two methods; they comprised 67.7 and 78.5% of the nucleated cells in the purest fractions from the isokinetic gradient and zonal rotor, respectively. Up to 160 million tumor cells can be purified in a single step with the reorienting zonal rotor, whereas 30 to 40 million cells per gradient approach the limit of the isokinetic gradient. After centrifugation in the zonal rotor, recovery was 85.6 +/- 12% (S.D.) of the cells layered over the gradient; and the separated tumor cells retained their ability to form tumors when transplanted into mice. The separation of large numbers of lymphocytes and malignant cells from the same tumor in the SZ-14 rotor should aid in the biochemical and immunological characterization of cancer.

Cytologic characterization of pulmonary alveolar macrophages by enzyme histochemistry in plastic.

Quantitative morphometric analyses of macrophages should facilitate a more precise definition of the role of macrophages in neoplastic diseases and inflammatory diseases of lymphoid and hemopoietic tissues. The usefulness of the available phenotypic markers is affected by the extent to which they are expressed by macrophages in tissue sections. For our study of phenotypic markers for macrophages in human tissue sections, we selected lung, since pulmonary alveolar macrophages are more readily identified morphologically in tissue sections than macrophages in most other tissues. In 1-2 micron sections of freshly fixed lung embedded in methacrylate, 89.7% of pulmonary alveolar macrophages had histochemically demonstrable acid phosphatase; 86.2%, nonspecific esterase with naphthol ASD chloroacetate as the substrate; 86.2%, nonspecific esterase with alpha-naphthyl butyrate as the substrate; 80.5%, peroxidase; and 75.8%, iron. With respect to their expression of markers, the observed heterogeneity among pulmonary alveolar macrophages is interesting; this heterogeneity may reflect the degree to which they have been activated, the different periods of time since they arrived in the lung, and differences in their local environments. Except for peroxidase, all examined markers were as well demonstrated when tissues were fixed after storage over liquid nitrogen as when fixation was carried out with fresh tissue. Acid phosphatase and nonspecific esterase with the chloroactetate substrate gave bright colors that would facilitate morphometric analyses. The storage of tissue over liquid nitrogen will be equally satisfactory for the characterization of macrophages with histochemical markers and monoclonal antibodies.

Separation of proximal tubule cells from suspensions of rat kidney cells in density gradients of Ficoll in tissue culture medium.

Rat kidneys were disaggregated with 0.25% trypsin. Cell were separated by velocity sedimentation in a previously described isokinetic gradient, by isopycnic sedimentation, and by velocity sedimentation followed by isopycnic sedimetation. In some fractions from the isokinetic gradient, 71.8+/-2.4+ of the nucleated cells contained histochemically demonstrable alkaline phosphatase (HDAP); in semithin sections, 62.7+/-2.3% of these cells had brush borders. The correspondence between fractions enriched for cells with HDAP and fractions enriched for brush border suggested that HDAP might be a suitable marker for rat proximal tubule cells. These cell constituted 46.5+/-2.6% of the nucleated cells in the starting sample suspension of kidney cells, and 81.9+/-2.7% of nucleated cells in the purified fractions from the gradients. More than 98% of nucleated cells in these fractions excluded typan blue. Following isopycnin centrifugation, the purest fractions contained 87.3+/-1.5% nucleated cells with HDAP, 9.6+/-2.5% nucleated cells iwithout HDAP, and 3.1+/-2.5% red blood cells. These proximal tubule cells had densities of 1.036 to 1.052 g/ml. With rate-zonal separation followed by isopycnic separation, the purest gradient fraction contained 93.0+/-1.9% nucleated cell with HDAP, 6.0+/-2.3% nucleated cells with HDAP, and 1.0+/-0.9% red blood cells. These proximal tubule cells sedimented a density of 1.041 g/ml. More than 98% of these cells excluded trypan blue.

A soft-agar procedure measuring growth of human colonic carcinomas.

Cell suspensions from 5 human colonic carcinomas were fractionated by velocity sedimentation and plated in soft agar. Cluster formation was restricted to the purest fraction of epithelial cells, as had been determined by immuno- and histochemical criteria. Plating efficiencies for the 5 specimens were 1.0-4.5%. The effects of varying the incubation period and inoculum size upon growth were studied using unseparated cell suspensions from 6 specimens. Clusters were apparent after 3 weeks in culture, and maximum cluster formation was typically seen by 5 weeks. Cluster formation appeared concentration-dependent, and individual specimens varied with respect to the inoculum most conducive to growth. The maximum plating efficiencies for unseparated cells were unseparated cells were 0.4-1.7%.

A characterization of human tonsillar lymphocytes after separation from other tonsillar cells in an isokinetic gradient of ficoll in tissue culture medium.

Purified human tonsillar lymphocytes responded less to mitogentic stimulation than did unseparated tonsillar cells. Their response to mitogens was restored when they were combined with cells from all other gradient fractions. We interpret our data as evidence that the majority of tonsillar lymphocytes require the presence of more rapidly sedimenting cells for a maximum response to the tested mitogens. The purified tonsillar lymphoyctes were 47.0% lymphocytes that have detectable surface immunoglobulin and 29.9% lumphocytes that form rosettes with sheep red blood cells. The predominant cell surface immunoglobulin was IgM. Digestion of the tonsil with trypsin yielded tenfold more plasma cells, more vialbe cells, and a larger proportion of blasts, histiocytes, and binucleated cells than were obtained by mechanical dissociation of the tissue.

Partial purification of human colonic carcinoma cells by sedimentation.

We have purified epithelial cells from human colonic tumours by velocity sedimentation in an isokinetic density gradient of Ficoll in tissue culture medium. In frozen sections of colonic carcinoma, histochemically demonstrable N-acetyl-beta-D-glucosaminidase (HDAG) was observed primarily in epithelial cells. We used this enzyme as a histochemical marker of epithelial cells. Initial suspensions of cells from colonic tumours suspended with 0-25% trypsin contained an average of 24% of the nucleated cells with HDAG. In the purest fraction obtained from gradient centrifugations, an average of 74% of the nucleated cells contained HDAG. After centrifugation, the quarter of the density gradient which contained the most rapidly sedimenting cells was purified 2-4-fold over that in the initial suspension. Cells in this zone of the gradient also gave rise to colonies in soft agar. Cells from initial suspension resulted in 15-25% as many colonies of 7 or more cells in cultures inoculated with the same number of nucleated cells. For the most part, cells obtained from the other zones of the gradient did not give rise to colonies in soft agar.

Heterogeneity and subcompartmentalization in the distribution of eosinophils in human colonic carcinomas.

Previously the authors reported that high concentrations of eosinophils in human colonic carcinomas are associated with better prognoses. They wished to know whether there were stromal subcompartments in these tumors with respect to the distribution of eosinophils. They evaluated concentrations of eosinophils at the margin and deep to the margin. For tumors with sufficient eosinophils to indicate good prognoses, the concentrations of eosinophils at the margins were strikingly lower than the concentrations in the samples remote from the margins. This is the first report that any kind of inflammatory cell is less concentrated in the periphery of a tumor than more centrally. When eosinophil concentrations were subjected to a square-root transformation, the transformed data for patients with and without metastases were different both for tumor from the margin (P = 0.001) and for tumor remote from the margin (P = 0.021).

Effects of the lpr/lpr mutation on T and B cell populations in the lamina propria of the small intestine, a mucosal effector site.

MRL mice, which develop a lymphoproliferative disease characterized by increased numbers of alpha/beta T cell receptor+ (TCR+) B220/6B2+CD4-CD8- T cells [lymphoproliferation (lpr) T cells], were studied for the effect of the lpr/lpr mutation on the mucosal immune system in the gastrointestinal (GI) tract. We analyzed the effect of the lpr gene mutation on T and B cell populations in the Peyer's patches (PP) and the lamina propria lymphocytes (LPLs), as examples of major IgA inductive and effector tissues in the GI tract respectively. Normal mouse PP contain B cells committed to IgA (surface IgA+) but only low numbers of B cells producing IgA. However, enhanced spontaneous IgA and IgG synthesis occurs in the PP of MRL mice. Further, we have now shown that PP of MRL mice are populated by lpr T cells. Interestingly, lpr T cells were not present in significant numbers in LPLs of MRL mice, even in older animals. Of interest was the finding that the ratio of CD4+ to CD8+ T cells in the lamina propria was lower in MRL when compared with control mice, and the CD8+ T cell subset actually predominates in LPLs of autoimmune mice. In addition, the number of gamma/delta TCR+ T cells in LPL of MRL lpr/lpr mice was significantly increased, especially in MRL lpr/lpr mice at 6 and 12 weeks of age. When the isotype distribution of B cells in LPLs was analyzed, no changes were noted in MRL lpr/lpr mice in comparison with MRL +/+ or normal control mice, and the pattern was IgA much greater than IgM greater than IgG. These results show that although increased numbers of CD8+ T cells and gamma/delta TCR+ cells occur in the LPLs of MRL mice, a normal distribution of plasma cell isotypes (IgA much greater than IgM greater than IgG) is found in this mucosal compartment. Further, Ipr T cells do not develop in the lamina propria compartment of the GI tract.

Cellular analysis of functional mononuclear cells from chronically inflamed gingival tissue.

Functional mononuclear cells from chronically inflamed gingiva of different stages of adult periodontitis (AP) were isolated by using a novel enzymatic dissociation method and extensively characterized at the single cell level. Fresh tissues obtained following surgery were cut into 1-2 mm3 pieces, washed free of residual blood cells and dissociated with the neutral protease enzyme Dispase. Mononuclear cells were then obtained by separation on a mini-Ficoll-Hypaque gradient. This procedure yields an average of 2 x 10(6) cells/400 mg of tissue. The viability of unfractionated cells immediately after enzymatic dissociation was greater than 94%, and after Ficoll-Hypaque gradient separation this was greater than 99%. The relative percentages of lymphocytes, plasma cells, monocytes (MN)/macrophages (M phi) and granulocytes were determined on cytospin preparations by Wright's-Giemsa staining. Cell differential analysis showed that lymphocytes and MN/M phi predominated at all stages of disease, while the frequency of plasma cells increased with the severity of disease. Numbers of plasma cells specific for IgG, IgA, or IgM were determined by immunofluorescence using fluoresceinated anti-heavy chain specific antibodies. The majority of Ig-containing cells were of the IgG isotype, followed by significant numbers of IgA and few IgM-positive cells. IgG, IgA, and IgM secreting cells were also determined by the ELISPOT assay. Total numbers of spot forming cells (SFC) were measured in both the moderate and advanced stages of disease, and the major SFC isotype was IgG followed by IgA; essentially no IgM SFC were detected. Furthermore, higher numbers of IgG and IgA SFC were noted in the advanced stage when compared with the moderate stage of disease.(ABSTRACT TRUNCATED AT 250 WORDS)

Regulation of mucosal responses by CD4+ T lymphocytes: effects of anti-L3T4 treatment on the gastrointestinal immune system.

The role of CD4+ T cells in the gastrointestinal (GI) immune system in vivo was studied in mice selectively depleted of this subset by treatment with monoclonal anti-L3T4 (GK1.5) mAb. Treatment of young BALB/c mice with weekly injections of anti-L3T4 mAb resulted in a selective depletion of CD4+ T cells in both IgA effector (lamina propria regions of the intestine; LP) and inductive (Peyer's patch; PP) sites. However, levels of CD3+CD4-CD8+ and CD4-CD8- (double negative) T cells remained constant or increased. When sections of small intestine were assessed for the isotype of Ig-containing cells, normal mice contained predominantly IgA plasma cells with small numbers of IgM and IgG plasma cells while anti-L3T4 treatment dramatically reduced the numbers of IgA plasma cells. When numbers of IgA-producing cells were assessed by the isotype-specific ELISPOT assay, the LPL of anti-L3T4 mAb-treated mice showed an 80% reduction in the number of IgA spot-forming cells. The effect of anti-L3T4 mAb treatment on IgA inductive sites was also studied and this treatment reduced the overall size of PP as well as the germinal centers in this tissue. Although anti-L3T4 treatment depleted CD3+CD4+ T cells in PP, the relative frequency of surface IgA-positive (slgA+) B cells in this tissue did not change. These results show that repeated injection of anti-L3T4 mAb results in a CD4+ T cell deficiency in both IgA inductive (PP) and effector (LP) sites. The depletion of CD4+ T cells resulted in reductions in the numbers of mature IgA plasma cells present in the LP of gut-associated tissues, and reduced the overall size of PP including germinal centers, but did not affect the frequency of sIgA+ B cells in this IgA inductive site.