Search

Background: We investigated the possibility that lymph nodes might increase metastatic efficiency of tumor cells lodged there by measuring changes in tumor cell invasiveness after physical contact with an in vitro approximation of a lymph node environment., Study Design: The experimental model involved growing Lewis lung carcinoma (LL) or B16 melanoma cells on microcarrier beads, rolling them on a "lymph node endothelial surface," which was created by growing endothelial cells on a differentiating acid extract of lymph node biomatrix, and testing the ability of those tumor cells to invade across matrigel-coated filters at rest (buffer) and in response to a chemotactic stimulus (3T3 conditioned media)., Results: Compared with contact with plastic, LL invasiveness was increased fivefold (buffer or conditioned media) and B16 invasiveness fourfold (conditioned media). Tumor cell invasiveness was not increased by exposure to the acid extract of biomatrix alone. Invasiveness to buffer or conditioned media after exposure to endothelial cells alone was 70 and 54 percent (LL) and 42 and 80 percent (B16), respectively, of the invasiveness induced by exposure to both. Compared with invasiveness induced by exposure to lymph node (100 percent), exposure to a "lung endothelial surface" induced invasiveness of 63 and 85 percent (LL) and 40 and 52 percent (B16) to buffer and conditioned media, respectively. Exposure to a hepatic endothelial surface induced invasiveness similar to that induced by lymph node; 90 and 82 percent (LL) and 110 and 86 percent (B16) of lymph node-induced invasiveness., Conclusions: A lymph node environment may modulate the metastatic potential of tumor cells.

Background: Adhesion to lymph nodes, rather than growth stimulation, accounted for preferential colonization of lymph nodes by a metastatic B16 melanoma. We investigated these adhesive interactions., Methods: Four classes of molecules were tested for inhibition of melanoma adhesion to cryostat sections of lymph node., Results: Calcium chelators ethylenediaminetetraacetic acid and ethyleneglycol-bis-(beta-aminoethylether)-N,N,N',N'-tetra ace tic acid completely inhibited adhesion (50% adhesion, half-maximal inhibition, at 1 to 3 mmol/L). Cytochalasin B, which impairs contractile microfilaments, inhibited adhesion (60% adhesion at .001 mmol/L, 28% at .01 mmol/L). Colchicine, which disaggregates microtubules, had a similar effect (20% at .01 mmol/L, lowest dose tested). Trypsin slightly increased adhesion (125% adhesion at 10 micrograms/ml). Neuraminidase, which removed sialic acid residues, inhibited it (50% adhesion at 5 micrograms/ml). Gly-arg-gly-asp-ser, a peptide with a cell binding sequence of fibronectin, did not consistently inhibit adhesion (69% adhesion at 0.1 mg/ml, 83% adhesion at 1 mg/ml) or substantially differ from gly-arg-gly-glu-ser-pro (59% adhesion at 0.1 mg/ml, 90% adhesion at 1 mg/ml). In contrast, a peptide with a cell binding region of laminin (tyr-ile-gly-ser-arg) inhibited adhesion (50% adhesion at .05 mg/ml)., Conclusions: Tumor cell-lymph node adhesion is a calcium-dependent process, requiring a functional cytoskeleton, that is mediated by both sialic acid moieties and trypsin-resistant, laminin-related, adhesion molecules.

Metastases from solid tumors to lymph nodes do not portend as poor a prognosis as metastases to other sites. The authors wished to determine whether specific subpopulations of cells metastasized to lymph nodes and whether they have different properties than cells metastatic to visceral sites. Repetitive selection for "spontaneous" metastases of a B16 melanoma to either lung or lymph node increased the incidence of lymph node metastases. Cells derived from pulmonary and lymph node metastases were assayed for their ability to adhere to cryostat sections of lung and lymph node and respond to target organ-conditioned media in serum-free conditions. Both cell types were four times more adherent to lymph node than lung, and consistently attached to the hilar and subcapsular sinuses. Attachment of cells derived from pulmonary metastases to either tissue was threefold greater than that of cells derived from nodal metastases. Lung-conditioned media stimulated proliferation of both cell types, and transiently induced differentiated morphology in cells derived from lymph node metastases, but not in cells from pulmonary metastases. Neither effect was found in lymph-node-conditioned medium. These results suggest that cells metastasize to lymph nodes preferentially not because of a specific predilection for lymph node, but because it is an easy site to colonize. Adhesive interactions in the lymph node rather than trophic ones appear to account for this effect. Cells metastatic to lymph node may be less "malignant" than cells metastatic to visceral sites because less has been required for them to succeed as a metastatic focus.

The ability to metastasize requires that tumor cells be able to degrade matrix. Nontoxic compounds that inhibit matrix digestion might be useful as anti-metastatic agents. We have investigated whether phenytoin, a drug commonly used in clinical practice that inhibits the production of collagenase by some cells, inhibits metastases in a standard animal model of metastasis: In vitro, phenytoin inhibited the proliferative response of B16 F10 melanoma cells to serum-containing media (75% inhibition at 25 micrograms/ml) but had no effect on their ability to degrade a type I collagen gel (1-100 micrograms/ml). Treatment of these cells with phenytoin prior to inoculation in vivo did not inhibit tumor growth, implantation in a surgical wound, or incidence of spontaneous metastases from a primary tumor growing in the foot. Pretreatment of mice with phenytoin (15, 40, and 75 mg/kg/day) diminished pulmonary metastases following tail vein injection in a minimal but dose dependent fashion; mean number of pulmonary colonies 4.6 +/- 3.1 (75/mg/kg/day) vs. 10.2 +/- 9.9 (control). However, tumor growth, implantation, and spontaneous metastases were not inhibited by pretreating the mice with the same doses of phenytoin. It is concluded that phenytoin has an insignificant inhibitory effect on tumor growth and metastasis.