Real-time observation of micrometastasis formation in the living mouse liver using a green fluorescent protein gene-tagged rat tongue carcinoma cell line
S. Ito et al., Real-time observation of micrometastasis formation in the living mouse liver using a green fluorescent protein gene-tagged rat tongue carcinoma cell line, INT J CANC, 93(2), 2001, pp. 212-217
Initial arrest, attachment, extravasation and subsequent extravascular grow
th of tumor cells in the secondary organs are believed to be crucial events
for hematogenous metastasis, but the actual processes in living animals re
main unclear. For the present study, we established green fluorescent prote
in (GFP)-expressing rat tongue carcinoma cell lines (RSC3) that permit real
-time analysis of micrometastasis formation in combination with intravital
video microscopy (IVVM). With this system, OFF-expressing metastatic (LM-EG
FP) and non-metastatic (E2-EGFP) cell lines could be visualized at the cell
ular level in live mice for more than month. Real-time IVVM analysis of liv
er metastases after intraportal injection of cells via a mesenteric vein re
vealed that both LM-EGFP and E2-EGFP tumor cells arrest similarly in sinuso
idal vessels near terminal portal venules within 0.4 sec, during which time
no evidence of a "rolling"-like movement along endothelial cell surfaces w
as observed, Quantitative analysis of GFP-positive foci showed that E2-EGFP
cells were completely sheared from the liver sinusoid within 3 days,with n
o solitary dormant cells, whereas a substantial number of LM-EGFP cells rem
ained in the liver, probably due to stable attachment to the sinusoidal wal
l. Confocal laser scanning microscopic study in combination with laminin im
munohistochemistry revealed that only LM-EGFP cells started growth at 3 to
4 days after inoculation and that most of the growing foci were surrounded
by subsinusoidal basement membrane. Our results suggest that micrometastasi
s formation by LM-EGFP cells consists of initial tumor cell arrest due to s
ize constraints of the vessel, stable attachment to subsinusoidal basement
membrane and subsequent intravascular growth before extravasation, The diff
erence in metastatic potential between the 2 lines may reside in their capa
city to attach stably to the vessel wall rather than their potential for in
itial cell arrest or subsequent growth. The system used in the present stud
y may be a powerful tool for analyzing targets for various anti-metastatic
agents in the sequential process of metastasis. (C) 2001 Wiley-Liss, Inc.