Search

Morphologic patterns in regional lymph nodes draining resected tumors correlate with in vitro immunologic data and have been reported to predict survival, stage of disease, and presence or absence of metastases. The strength of the relationship between immunomorphologic findings and survival rates varies with the type of tumor, perhaps because of relative immunogenicity, tumor size and growth rate, or even the stimulation of the lymph nodes by infection. Nevertheless, there is remarkable agreement among investigators on key observations, which yield more prognostic information than can be obtained by standard methods.

Protein antigens covalently conjugated with lipid groups (dodecanoic acid) have previously been shown to stimulate strong delayed-type hypersensitivity (DTH) without the aid of adjuvants. The present experiments show that lipid-conjugated bovine serum albumin (L-BSA) is taken up in vitro by macrophages (Mpsi) 25- to 50-fold more than unconjugated BSA or aminidated BSA, neither of which induces DTH. Macrophages that take up 125I-labeled L-BSA in vitro stimulate DTH even more efficiently, when injected into syngeneic guinea pigs, than does soluble L-BSA. Tracer studies on the fate of radiolabeled BSA and L-BSA showed that much more L-BSA than BSA was retained by draining lymph nodes. Autoradiography demonstrated that 125I-L-BSA is rapidly taken up by Mpsi in the medullary sinuses of the lymph nodes. Some of this antigen is then transported into the paracortex, a region in which T lymphocytes predominate. The capacity of lipophilic antigens to stimulate cell-mediated immune responses may be caused by their increased uptake by Mpsi, resulting in more efficient presentation to immunocompetent T lymphocytes. The anatomical site of this Mpsi-T cell interaction may be within the sinusoids or paracortex of the draining lymph nodes.

The covalent conjugation of fatty acid to a tumor cell membrane preparation transformed it from an antigen that enhanced tumor growth to one that suppressed it. A crude cell membrane preparation was made by sequential hypertonic and hypotonic salt extraction of tumor cells from a fibrosarcoma induced in hamsters by simian virus 40. The membranes were chemically conjugated with dodecanoic anhydride in 0.5 M carbonate buffer (pH 9.0). Injection of unmodified membranes 10 days before transplantation of live tumor cells produced clear-cut enhancement of the tumor growth rate. In contrast, injection of lipid-conjugated membranes in a similar dose and protocol suppressed tumor growth. The lymphoid proliferative reactions to the tumor cells as demonstrated by the histology of both the tumor and regional lymph nodes were consistent with the hypothesis that unmodified membranes stimulated the production of antibody which participated in the enhancement of tumor growth, and that lipid-conjugated membranes stimulated the production of cell-mediated immunity which suppressed this growth.

We evaluated the molecular and physicochemical properties of surfaces formed by defined layers of block copolymers which were especially effective as adjuvants or in the induction of granulomas. The copolymers which were adjuvants formed hydrophilic surfaces with a large area. They bound protein in a way which left it particularly accessible to antibody and induced the activation of complement. Copolymers which induced granulomas, in contrast, formed hydrophobic crystalline surfaces. They bound less protein and did not activate complement, but were toxic for macrophages. Their surfaces were found to be similar to those of the mycobacterial glycolipid trehalose-6,6'-dimycolate or quartz, in that they consisted of regular geometric arrays of hydrophilic and hydrophobic adsorptive domains. These studies demonstrated that changes in the size and arrangement of hydrophilic and hydrophobic blocks in copolymers produce a diversity of surface physicochemical properties which correlate with biologic activity.

A battery of immunologic tests were performed on 42 patients with recurrent cancer of the head and neck whose median survival time was 6.8 months. Two tests correlated with survival: total white blood cell (WBC) count and the percentage of lymphoid cells (%LG) in the Ficoll-Hypaque gradient separated leukocyte fraction of blood. Patients with abnormally low values of %LB had a poor prognosis, as did patients who had greater than 8000 WBC count. Of the variation in %LG, 43% was due to the differences in differential WBC counts among the patients, with the remaining variation attributed to changes in the buoyant densities of the leukocytes. These data suggest that further study of the factors effecting buoyant density of cells may be clinically useful.

Transferrin levels were measured immunochemically and the levels of several other plasma proteins were estimated by high resolution agarose gel electrophoresis in serial serum samples from nine patients undergoing bone marrow transplantation for aplastic anemia or leukemia. Three of five patients with low pretreatment transferrin levels died during initial hospitalization, and the other two suffered severe complications prior to recovery. All four patients with pretreatment transferrin levels within the normal range were discharged, three of them with minimal complications. Serial studies demonstrated that transferrin levels fell with the onset of infections or graft-versus-host disease and rose with recovery. Similar changes were not seen with any of the other plasma proteins measured. These results suggest that measurement of serum transferrin may reveal important prognostic information about patients undergoing bone marrow transplantation.

A series of block co-polymers composed of polyoxyethylene and polyoxypropylene were investigated for their ability to induce in vitro activation of mouse mast cells. We found that six of these co-polymers could cause histamine release from mouse mast cells in vitro. At low concentrations, the most efficacious co-polymer, T130R2, caused rapid and extensive concentration-dependent release of histamine from mouse mast cells. The release process was not cytotoxic; it required metabolic energy and was not accompanied by release of lactate dehydrogenase. Optimal release of histamine was dependent on both calcium and sodium ions in the extracellular medium. The degree of in vitro histamine release correlated with in vivo inflammation and in vitro ionophore activity. We believe that this represents the first report of the activation of mediator-containing cells by an ionophore selective for monovalent cations. These copolymers may therefore represent new reagents for investigations of cellular excitation.

The adjuvant activities of four chemically similar, but physicochemically different nonionic surface-active agents called pluronic polyols F 68, L 31, L 101 and L 121 were studied. These four agents were tested in mice using an experimental model developed for studying the adjuvant activity of the cationic surface-active agent dimethyl dioctadecyl ammonium bromide (DDA). L 121 and DDA enhanced the primary antibody response to sheep red blood cells (SRBC) while F 68, L 31 and L 101 suppressed this response. The secondary humoral response to SRBC was enhanced by the polyol L 121 while the secondary response to dinitrophenylated bovine serum albumin (DNP22-BSA) was enhanced by both L 121 and L 101. DDA and the polyol L 101 were very effective adjuvants for induction of delayed-type hypersensitivity (DTH) to SRBC and DNP22-BSA after intracutaneous immunization of mice with a mixture of antigen and adjuvant. Since the four pluronic polyols were composed of identical chemical constituents, we proposed that difference in their activities as adjuvants were due to variation in their physicochemical properties. A correlation was found between a physicochemical parameter, the hydrophilelipophile balance (HLB), and the adjuvant activities of the pluronic polyols and several other types of nonionic surface-active agents. The agents which were strong adjuvants all had HLB values within a narrow range which classified them as spreading agents.

Co-polymers composed of polyoxyethylene and polyoxypropylene have been shown previously to trigger histamine release from mouse peritoneal mast cells; this property quantitatively is directly related to the ionophorous ability of these compounds to cause a functional exchange of intracellular K+ for extracellular Na+ across the cell membrane. We investigated the effect of an inflammatory copolymer, T130R2, on human basophils. The data demonstrate that T130R2 can cause calcium-dependent histamine release from human basophils in vitro. Further, at concentrations that do not cause histamine release, this co-polymer markedly augments release by suboptimal concentrations of the lectin Con A or anti-IgE antibody and the phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate but not the calcium ionophore A23187. Thus, these co-polymers induce mediator release from cells of both rodents and humans. In both instances it is likely that calcium-dependent cell triggering is the result of an influx of sodium ions with concomitant depolarization of the transmembrane potential. In common with the calcium ionophore A23187, the co-polymer T130R2 has the ability to synergize with stimuli which trigger the IgE receptor as well as those which directly activate the cellular calcium- and phospholipid-dependent protein kinase.

Patients with advanced cancer have a depressed immunological function. We performed a battery of tests on peripheral blood samples from 42 patients with gynecological cancer to determine the extent to which this depression was due to abnormal lymphocyte function, as compared to changes in the number of lymphoid cells in the peripheral blood or in the efficiency of purification of cells in Ficoll:Hypaque gradients in preparation for testing. The percentage of lymphocytes in the gradient-derived cell suspension (% LG) and the absolute lymphocyte count were more informative than mitogen stimulation, mixed leukocyte culture, and T- and B-cell measurements. Both values decreased significantly with the advancing stage of cancer, and we were able to predict survival of patients with uniform stage of disease. The % LG correlated with survival better than did any other test when multivariate analyses of all test combinations were performed. Low values of % LG reflected both the depressed lymphocyte counts and the altered buoyant density of the leukocytes of many patients with advanced cancer. A large portion of the depression in other immune function tests was statistically attributed to changes in % LG and the lymphocyte counts. We concluded that these two simple measurements provide valuable information about patients with gynecological cancer.

Pulmonary hyalinizing granuloma (PHG) is a disease of slowly enlarging pulmonary nodules made up of dense bundles of collagen accompanied by an infiltrate of chronic inflammatory cells. The etiology is unknown. Although it has been suggested that the lesions represent an exaggerated immune response to unidentified agents, results of a detailed immunologic work-up of these patients have not been published. This report presents the laboratory findings of two patients with biopsy-proven PHG who have been followed four and eighteen years. Autoantibodies were detected (antinuclear antibody, rheumatoid factor, and positive antiglobulin tests), although clinically there was no evidence of a specific collagen-vascular disorder. Both patients had elevated levels of circulating immune complexes. These data suggest that immune complex mechanisms may be important in the pathogenesis of PHG.

Patients with leukemia were found to have a high percentage of saturation of their serum transferrin with iron to an extent only rarely observed with other malignancies. This was associated with a reduced ability of their serum to inhibit the growth of a test strain of Pseudomonas aeruginosa. Serum iron, transferrin, and related parameters were measured serially in patients undergoing bone marrow transplantation for leukemia or aplastic anemia. It was found that a high proportion of these patients also have a high saturation of their transferrin with iron. This was related to three distinct physiologic deficits: a low level of serum transferrin; a high level of iron; and an inability to reduce the level of serum iron during infection. Three of six patients who were unable to reduce their serum during fever and infection subsequently died of sepsis. These data support the hypothesis that derangements in nonspecific serologic defense mechanisms involving iron contribute to susceptibility to infection in patients with leukemia undergoing bone marrow transplantation.

Beagle serum proteins were separated by 2-dimensional electrophoresis and identified by histochemical and autoradiographic techniques. Haptoglobin, hemoglobin, ceruloplasmin, transferrin, and lipoprotein and glycoprotein fractions were identified. Use of a continuous gradient polyacrylamide gel revealed serum protein fractions not previously described.

Axillary lymph nodes obtained from 37 patients with cancer of the breast were used in a microcytotoxicity assay against a human mammary cancer cell line, ALAB. Cell suspensions made from 60 individual lymph nodes and 51 lymph node pools were tested. Each lymph node was graded for the extent of the histologic changes of sinus histiocytosis (SH), paracortical hyperplasia (PCH), and germinal center hyperplasia (GCH). High levels of cytotoxicity correlated significantly with the presence of SH and PCH. When multiple lymph nodes of individual patients were pooled and tested, a high intensity of GCH correlated with a low degree of cytototoxicity. The cytotoxic activity of pooled lymph nodes from different axillary regions was studied in 20 patients. In 7 of these patients, low axillary lymph node cells were more cytotoxic than high axillary node cells. The reverse was found for 7 patients, and there was no difference in cytotoxicity between axillary regions in 6. Cytotoxic lymph node pools had high SH and low GCH whether they were obtained from the low or high region. Noncytotoxic lymph node pools from the high axillary region had a low intensity of all three histologic reactions. In contrast, noncytotoxic lymph node pools from the low axillary regions had high intensities of SH, PCH, and GCH. These results suggest that germinal center hyperplasia in the lymph nodes nearest an advancing tumor is associated with a local suppression of cytotoxic cell activity.

We have observed an electrophoretically abnormal, nonfunctional species of alpha 1-antitrypsin in serum from patients who were receiving bone-marrow transplants for treatment of leukemia or aplastic anemia. Three of four patients in whose serum this protein appeared died soon after; the fourth recovered, and the disappearance of the abnormal alpha 1-antitrypsin paralleled his recovery. This suggests that the inability to maintain functional activity of alpha 1-antitrypsin predisposes patients to life-threatening complications during recovery from bone-marrow transplants.

We describe an enzymic assay for the functional activity of alpha 1-antitrypsin, in which tosyl-Gly-Pro-Lys-p-nitroanilide acetate (Chromozym PL) is the trypsin substrate. For the indicating reaction, the concentration of substrate is about double the Km of its reaction with trypsin. This assay is an improvement over reactions involving N-alpha-benzoyl-DL-arginine p-nitroanilide as substrate, the solubility of which in water is less than the Km of its reaction with trypsin. The assay is standardized in terms of active sites of trypsin inhibited per liter of serum, with p-nitrophenyl-p'-guanidinobenzoate as the active site titrant. The results agreed well with those by an immunonephelometric assay for alpha 1-antitrypsin (r = 0.953). The within-run and run-to-run precision was approximately 3%. The results of the assay varied linearly with alpha 1-antitrypsin concentration from 0 to 60 mumol/L of serum. The minimum sample size was 50 microL, and analysis of 24 samples took less than 20 min.

We tested the ability of 17 surface-active agents to enhance antibody formation and inflammation. The surfactants were all block copolymers of hydrophilic polyoxyethylene (POE) and hydrophobic polyoxypropylene (POP), which differed in m.w. and mode of linkage of POP to POE. Mice were injected in each rear footpad with 1.25 mg of each surfactant with 25 micrograms of bovine serum albumin in an oil-in-water emulsion. Each agent produced a distinct pattern of immune response and inflammation. Preparations that are large and insoluble with the POE chains flanking the POP chains were very effective adjuvants for increasing antibody formation. They also activated complement and induced the release of chemotactic factors from serum. Increasing the percent of POE decreased adjuvant activity and inflammation. Decreasing the m.w. of the molecules while maintaining the proportions of POP and POE decreased the adjuvant activity and increased inflammation. Preparations synthesized in the reverse order, with the POP flanking POE, tended to induce granulomas instead of antibody. These data demonstrate that block copolymer surfactants have a spectrum of biologic activities that depend on the size and arrangement of their constituent parts. We suggest that much of the activity of these agents derives from their ability to form adsorptive surfaces similar to those of a mycobacterial glycolipid, quartz, and monosodium urate.

The magnitude of the skin test response in a group of normal volunteers to intradermal Candida antigen correlated closely with the level of cytolytic activity stimulated by Candida antigen in vitro in the peripheral blood lymphocytes from those same individuals. The cytolytic activity stimulated by Candida antigen was cell mediated, and cold target inhibition studies demonstrated that Candida antigen-stimulated effector cells were capable of cross-species (i.e., nonspecific) killing. Analysis of the Candida antigen-induced effector cell population for various cell markers did not enable absolute identification of the cell responsible for the killing. The findings in this study indicate that the stimulation of cell-mediated cytolysis by Candida antigen may be related to the delayed-type hypersensitivity response produced by the same antigen.