Invasiveness in hepatocyte and fibroblast monolayers and metastatic potential of T-cell hybridomas in mice.

Fusion of noninvasive, nonmetastatic BW5147 T-lymphoma cells with normal T-lymphocytes usually resulted in highly invasive and metastatic T-cell hybridomas, apparently due to properties derived from the normal T-cell. Occasionally hybrids arose that were non- or low invasive, probably by loss of relevant genes upon chromosome segregation, since these cells contained much less DNA than highly invasive hybrids. The metastatic potential of 20 representative T-cell hybridomas was tested by tail vein injection in syngeneic mice and cells were found to be either nonmetastatic (NM), low metastatic (LM), or high metastatic (HM). NM hybrids were tumorigenic but did not form metastases and HM hybridomas caused wide-spread metastasis. LM cells formed metastases in a limited number of mice and predominantly in lymphoid tissues. In hepatocyte cultures, NM cell lines were found to be the least invasive, HM cells the most, whereas LM hybrids exhibited intermediate levels. Invasiveness was not only measured in rat hepatocyte cultures but also in rat embryo fibroblast monolayers, and the relative invasive capacity in both model systems correlated well. Pertussis toxin inhibited invasion in both systems to 20-30% of control values. This suggests that the mechanisms of invasion into hepatocyte and fibroblast cultures are at least partially similar and that the fibroblast invasion assay is a relevant model to study aspects of lymphoma metastasis. We conclude that invasive potential is a prerequisite for T-cell hybridomas to colonize tissues from the bloodstream and that a minimum level of invasiveness is necessary for extensive and wide-spread metastasis formation.

Cell surface sialic acid and the invasive and metastatic potential of T-cell hybridomas.

T-cell hybridomas prepared by fusion of non-invasive non-metastatic BW5147 T-lymphoma cells and activated normal T-cells were found to be highly invasive in vitro and highly metastatic in vivo upon tail vein injection. By prolonged culturing and subcloning, non-invasive, non-metastatic hybrids were selected with modal DNA/cell contents close to the diploid value of both fusion partners. Since normal activated T-cells were invasive in vitro in hepatocyte cultures, these data suggest that invasiveness of the hybrids is derived from the parental normal T-cells and is one of the properties responsible for the metastatic potential of these cells. Analysis of a large panel of T-cell hybrids with fluorescein isothiocyanate conjugated lectins, specific for terminal galactose and/or N-acetylgalactosamine sugar residues, showed an inverse correlation between expression of lectin receptor sites and invasive and metastatic potential of the hybrids. Soybean agglutinin, as well as peanut agglutinin and Ricinus communis agglutinin, reacted strongly with non- or low-invasive hybrids but only weakly with invasive hybrids. The difference in lectin binding between both types of hybrids appeared to be due to masking of receptor sites by sialic acid. Removal of cell surface sialic acid by neuraminidase treatment unmasked the lectin receptor sites of invasive hybrids to the level of the corresponding sites of non- or low-invasive cells. This increase in active lectin binding sites was simultaneously accompanied by a striking decrease of invasiveness to the level of the low-invasive hybrids. Conversely, the blocking of R. communis agglutinin receptors by sialic acid allowed selection of invasive hybrids from segregating cell populations with the toxic lectin R. communis agglutinin. The results taken together indicate that sialylation of particular cell surface carbohydrate residues on the T-cell hybridomas is associated with the invasive and metastatic potential of these hybrids. The reduction of invasive potential after removal of cell surface sialic acid provides further evidence for a functional role of this sugar residue in invasiveness of the T-cell hybrids.

Invasiveness of T-cell hybridomas in vitro and their metastatic potential in vivo.

BW5147 lymphoma cells, which are noninvasive and nonmetastatic, were fused with normal T-lymphocytes. The invasiveness of the generated T-cell hybridomas was tested in hepatocyte cultures, and their metastatic potential was tested by tail vein injection. A total of 29 hybridomas generated from alloantigen-activated T-cells were all found to be invasive. One of these cell lines rapidly lost invasiveness in culture. Most hybridomas generated from nonstimulated spleen T-cells were also invasive, but 5 of 27 were not. Six invasive and four noninvasive hybridomas were injected into the tail vein of syngeneic mice. All invasive cell lines caused extensive and widespread tumor growth, particularly in the liver, which was usually severalfold enlarged; the spleen; kidneys; and ovaries. In contrast the noninvasive hybrids, which were tumorigenic upon s.c. injection, did not form any metastases. We conclude that properties derived from normal T-cells, when introduced into noninvasive T-lymphoma cells, cause them to become invasive as well as metastatic. Furthermore for this tumor cell type invasiveness as measured in hepatocyte cultures appears to be closely associated with the ability to colonize organs from the bloodstream.

In vitro invasiveness of CTL clones and in vivo dissemination of CTL hybridomas.

Activated spleen T cells are invasive in hepatocyte and fibroblast cultures, and this property is dominantly expressed in T cell hybridomas. The invasive potential of the hybrids correlates with their capacity to disseminate in vivo. We have used this model to study the invasive and migratory properties of cytotoxic T lymphocytes (CTLs). Two murine CTL clones were highly invasive, independent of their state of activation. CTL hybridomas, derived from one of the clones, were similarly invasive. In vivo, CTL hybridoma cells disseminated to extravascular sites in the liver, kidneys, lungs, ovaria, tubae, uterus, and lymphoid, mesenchymal, and fat tissues. Within 7 to 14 days, 10(6) cells were lethal in 100% of mice. The adhesion molecules CD2, CD8, CD54, L-selectin, and CD49d (VLA-4 and LPAM-1 alpha-chain) were not expressed by all CTL hybridomas and therefore not indispensable for invasion in vitro and dissemination in vivo. In contrast, LFA-1 (CD11a/CD18), CD44, and VLA-6 (CD49f/CD29) were expressed on all hybrids. LFA-1 antibodies inhibited CTL hybridoma invasion in vitro, but antibodies inhibiting CD44-hyaluronate and VLA-6-laminin interaction had no effect. These results suggest that migration of cytotoxic T cells into noninflamed tissues is independent of their activation state and does not require L-selectin, LPAM-1, CD2, and VLA-4.

Inhibition and stimulation of LFA-1 and Mac-1 functions by antibodies against murine CD18. Evidence that the LFA-1 binding sites for ICAM-1, -2, and -3 are distinct.

The murine CD18 monoclonal antibody (mAb) M18/2 was reported to inhibit lymphoma metastasis [Zahalka, M. A. et al. (1993) J. Immunol. 150, 4466]. To identify the pathways potentially blocked, we studied the effects of M18/2 compared with two new mAb against murine CD18, GAME-46, and -245. Whereas the GAME mAb blocked most Mac-1-mediated interactions, M18/2 had no effect, or even stimulated. The same was true for adhesion of LFA-1 to ICAM-1. To test effects on interactions with different ICAMs, we used L cells transfected with human ICAM-1, -2, and -3. As previously described, mouse LFA-1 does not bind to human ICAM-1 but we show here that mouse LFA-1 does bind to human ICAM-2 and -3. Again, the GAME mAb blocked completely, but M18/2 did not. These results indicate that the LFA-1 binding sites for ICAM-1 and ICAM-2 and -3, although in close vicinity, are distinct. Furthermore, effects of M18/2 on metastasis cannot be ascribed to blocking of any known beta2-integrin activity.

Unilateral renal vein thrombosis and nephrotic syndrome. Report of a case with protein selectivity and antithrombin III clearance studies.

A case of the nephrotic syndrome with unilateral renal vein thrombosis is reported. The patient, an 18 year old man, presented with a six month history of edema and the recent development of a left-sided varicocele. An enlarged left kidney and a thrombus in the left renal vein were demonstrated roentgenographically. A biopsy specimen of the right kidney was interpreted as membranous glomerulonephritis. Selective renal function studies showed nearly identical creatinine excretion, and similar total protein excretion and protein selectivity from each kidney. Thus, the thrombus in the left renal vein did not influence glomerular filtration rate or quantitative or qualitative protein excretion. A high urinary output and a decreased serum level of antithrombin III were measured. These findings suggest a mechanism to explain the increased thrombotic tendency seen in this and other patients with the nephrotic syndrome.

Presence of circulating macromolecular IgA in patients with hematuria due to primary IgA nephropathy.

The relation between renal histologic features and the presence of circulating immune complexes was studied in 50 patients with hematuria. Primary IgA nephropathy was found in 25 patients, and various other forms of glomerulopathy were seen in the remaining 25 patients. Circulating immune complexes were detected with the 125I-C1q-binding assay, the conglutinin-binding assay, and the anti-IgA inhibition binding assay, the latter detecting specifically IgA-containing immune complex-like material. The 125I-C1q-binding assay gave negative findings for all patients except one. With the conglutinin-binding assay, immune complexes were found in a similar frequency for patients with and without IgA nephropathy. However, the anti-IgA inhibition binding assay gave positive results only in patients with primary IgA nephropathy (68 percent) and in none of the other patients. Sucrose density ultracentrifugation, as well as experiments in which the anti-IgA inhibition binding assay was performed with and without pretreatment of serum with polyethylene glycol, showed the presumed IgA immune complexes to have intermediate sedimentation coefficients (11 to 21S). The presence and level of this macromolecular IgA in the circulation correlated significantly (p less than 0.001) with the presence of hematuria in patients who had this clinical manifestation intermittently. Furthermore, a significant correlation (r = 0.69, p less than 0.0001) was found between the degree of hematuria and the degree of positive findings of the anti-IgA inhibition binding assay. This study shows that in patients presenting with hematuria, a positive finding on the anti-IgA inhibition binding assay is restricted to patients with primary IgA nephropathy and therefore could be of diagnostic value.

Antibodies directed against antigens on the endothelium of peritubular capillaries in patients with rejecting renal allografts.

This study was undertaken to examine the humoral immune response against endothelial antigens of the donor kidney in human renal allograft recipients. Sera from 61 transplant recipients who received 62 grafts were studied for the presence of circulating endothelial antibodies (CEAb) using an indirect immunofluorescence technique with a pretransplant biopsy of the graft as a substrate. IgG antibodies directed against the endothelium of peritubular capillaries were found in the sera of 6 of the 10 patients with graft rejection within 7 weeks after transplantation, whereas these antibodies were not found in the absence of rejection (P less than 0.001). Immunofluorescence studies of post-transplant biopsies showed IgG along the endothelium of peritubular capillaries only in the grafts of patients with CEAb. Eluates from these grafts contained IgG antibodies that bound to the endothelium of the donor as shown by the indirect immunofluorescence technique. Absorption of endothelial antibody (EAb)-positive sera with human platelets or Wistar strain rat erythrocytes showed that the EAb were not directed against serologically defined HLA antigens or against heterophile antigens on rat erythrocytes. We conclude from this study that the presence of antibodies directed against endothelial antigens is associated with poor graft prognosis and that these antibodies may be responsible for the rejection process.

Growth of an aorta-coronary anastomosis. An experimental study in pigs.

The influence of growth of an aorta-coronary anastomosis, comparable to the coronary translocation anastomosis during the arterial switch operation, was studied in pigs. The anastomosis between the right coronary artery and the aorta did not grow, and this lack of growth may result in stenosis. With another technique, by which the coronary artery was excised with a cuff of aortic wall, the effects caused by absence of growth were circumvented and a normal-sized coronary ostium was present after growth. However, when no cuff was used, stenosis occurred at the suture line and caused growth retardation of the animal as well as histologic damage to the right ventricle.

Effect of the adjuvant dimethyl dioctadecyl ammonium bromide on the humoral and cellular immune responses to encephalomyocarditis virus.

The effects of the adjuvant dimethyl dioctadecyl ammonium bromide (DDA) on the immune responses to encephalomyocarditis (EMC) virus were studied in mice. The humoral response, as measured by appearance of neutralizing antibodies, was slightly enhanced in mice immunized by the intraperitoneal route. Intracutaneously, DDA almost did not affect the humoral response but resulted in distinct enhancement of delayed type hypersensitivity (DH), as measured by the footpad swelling test. DH to EMC virus was found to be antigen-specific and could be passively transferred to normal mice with peritoneal exudate cells from immunized mice. Dose-response curves for DH and humoral antibody responses to EMC virus were not concordant. Low doses induced DH on day 6 without measurable circulating antibodies; high doses gave good antibody responses but suboptimal DH reactions. Immunization conferred a state of resistance to infection with virulent EMC virus. Protection seemed more related to DH than to the prevalence of specific antibodies at the time of infection.

Medullary differentiation of anaplastic thyroid carcinoma.

Fourteen cases of anaplastic carcinoma of the thyroid (ACT) were studied with silver staining and immunohistochemical technics for signs of medullary differentiation. For comparison a series of 11 cases of medullary carcinoma of the thyroid (MCT) was studied. In nine out of 14 cases of ACT both argyrophilia and immunoreactive calcitonin could be detected as evidence of an apparent C-cel origin of these tumors. Most of the patients with these anaplastic medullary carcinomas died within 6 months after the histological diagnosis was made. It is concluded that the histological diagnosis of anaplastic variants of MCT can only be made on the basis of argyrophilic cytoplasmic granules and calcitonin immunoreactivity. In contrast to well differentiated MCT the anaplastic medullary carcinoma is characterized by a poor prognosis.

Optimization of a classical aromatase activity assay and application in normal, adenomatous and malignant breast parenchyma.

The tritium water release assay, originally described for the analysis of aromatase activity in placental tissue, was used to estimate aromatase activity in breast tissue samples. The lower activity in this tissue necessitates longer incubation times and thus optimization of the assay conditions. To prevent oxidative and proteolytic inactivation of aromatase, dithiothreitol and albumin were added to the incubation mixture. Extra NADPH, cofactor in the aromatase reaction, also improved reaction rate in placental incubations, but after approximately 120 min activity rapidly decreased. Inhibitors gradually produced during the incubation could explain this phenomenon. Quantitative gas chromatography-mass spectrometry (GC-MS) analyses of testosterone, oestradiol, oestrone and androstenedione after incubation with non-labelled androstenedione proved that a substantial amount of the substrate is converted into testosterone. Qualitative GC-MS steroid profiling of the incubation mixture demonstrated the presence of hydroxylated oestradiol and hydroxylated testosterone, produced during incubation, which could have caused partial aromatase inhibition. The adjusted assay was used to analyse 84 breast tissue samples, histologically classified as normal, adenoma or carcinoma. Aromatase activity was found in 56% of all samples and ranged from 0.6 to 26 pmol oestrogen/g protein per hour. Aromatase positivity was found in 80% of the normal samples, 56% of the adenoma samples and 48% of the carcinoma samples. Although carcinoma samples were less often aromatase positive than normal tissue samples (chi2 = 4.80; P < 0.050) there was no difference in absolute aromatase activity. Because no less than approximately 50% of the carcinomas contained aromatase activity and because of the non-routine character of the assay we conclude that it is justified to start aromatase inhibition therapy without previous knowledge of the aromatase status.

Donor pretreatment-mediated and cold storage-dependent proximal tubular necrosis--a unique form of acute renal insufficiency.

After a 5-hour period of donor pretreatment with cyclophosphamide (CY) and methylprednisolone (P) (100 mg/kg each), cold storage of pretreated canine renal allografts may cause early and severe postoperative renal insufficiency. This renal insufficiency is mediated by CY metabolites and depends on the number of hours of cold storage, for severe renal insufficiency is not observed after 6 hours of cold storage but is invariably present after cold storage beyond 18 hours. The renal insufficiency is associated with coagulation necrosis of the proximal tubules, particularly the pars recta. Since the repair of ischemia-medicated proximal tubular lesions requires mitotic activity, results suggest that the proximal tubules of donor pretreated kidneys are subjected to a concentration of CY metabolites sufficient to cause an extent of DNA damage that, in the absence of a sufficient time for nuclear repair, inevitably leads to cell death and renal insufficiency when the tubular cells are driven to mitosis by cold storage-mediated ischemia.

Chromosomal changes in renal oncocytomas. Evidence that t(5;11)(q35;q13) may characterize a second subgroup of oncocytomas.

Many of the reported oncocytomas have different chromosome abnormalities, indicating that they comprise a cytogenetically heterogenous group of tumors consisting of potentially cytogenetic subgroups. We have performed cytogenetic studies on nine renal oncocytomas. Clonal abnormalities were present in eight tumors. The findings most observed were the loss of the Y chromosome, and abnormalities of chromosomes 1 and 22. We also observed telomeric associations (tas) in two tumors and structural aberrations of chromosomes 9p and 19q, as well as monosomy 10. In two cases we found a similar reciprocal t(5;11)(q35;q13) in two cases. Review of the literature disclosed one other oncocytoma with a t(5;11) (q35;q13). This suggests that t(5;11)(q35;q13) defines a (second) subset of oncocytomas apart from the subgroup specifically associated with the loss of chromosomes 1 and Y.

Negative sural nerve biopsy in neurolymphomatosis.

Patients with non-Hodgkin's lymphoma occasionally develop widespread invasion of peripheral nerves by tumor cells or neurolymphomatosis (NL). Clinically this usually results in asymmetrical, progressive, and painful polyneuropathy. Diagnosis rests on the identification of tumor cells in peripheral nerves. To avoid false-negative biopsy findings in patients with malignant lymphomatous infiltration of peripheral nerves it has been recommended to biopsy clinically involved nerves. We present two patients with histologically confirmed NL in whom sural the nerve biopsy finding was negative despite clinical and neurophysiological evidence of involvement of the sural nerve a. The clinical features of NL are reviewed. Some patients with neurolyphomatosis have only focal or proximal involvement of nerves, requiring the biopsy of an affected part of these nerves. Magnetic resonance imaging may be useful in identifying affected nerves.

Selective alterations in guanine nucleotide regulation of opiate receptor binding and coupling with adenylate cyclase.

Pretreatment of rat brain membranes at pH 4.5 increased GTP and Gpp(NH)p regulation of [3H]-D-ala2-met5-enkephalinamide (D-ala-enk) binding with no change in absolute binding itself. Pretreatment at pH 4.5 did not alter basal adenylate cyclase activity but did cause a loss (50-90%) in NaF- and Gpp(NH)p-stimulated activity, indicating a functional loss of GTP-coupling proteins. The addition of cis-vaccenic acid partially restored NaF- and Gpp(NH)p-stimulated cyclase after low pH pretreatment but, in the same membranes, did not reverse the increase in GTP regulation of agonist binding. These results suggest that GTP regulation of binding and stimulation of adenylate cyclase occur by fundamentally different mechanisms.

Preexisting glomerulonephritis in allografted kidneys. Occurrence in man.

Biopsy specimens were taken from 43 renal allografts after completion of the anastomoses. In five kidneys from four donors, mesangial lesions compatible with preexisting glomerulonephritis were found. Immunofluorescence study of three kidneys indicated immunecomplex pathogenesis. In one donor, congenital heart disease was present: the others were free of manifest (renal) disease. Clinical follow-up data and repeated renal biopsy specimens indicated an uneventful course of these allografts. Awareness of this kind of lesions is useful for interpretation of glomerular lesions in biopsy specimens from renal transplants.