[CANCERRESEARCH54, 3620—3624,July 1, 19941

reported a quantitative colorimetric method, MTT3 assay (22). In1990, Schlechte et aL (23) applied the MTT assay to the invasionassay using a large size Transwell chamber. The uniformity of thecoated Matrigel, however, was not examined well. It is difficult tocoat an even and uniform Matrigel layer on the filter of the assaychamber with the use of a small amount of Matrigel. We have oftenobserved that cells passed easily through the Matrigel-coated filterwithout invasion because of the uneven thickness or absence ofMatrigel. Therefore, we devised a simple, easily reproducible in vitroassay for quantitating tumor cell invasion.

MATERIALS AND METhODS

Agents. Matrigel (Collaborative Research Co., Bedford, MA), a solubilizedbasement membrane preparation extracted from the Engelbreth-Holm-Swarmmouse sarcoma, was prepared. Its major components are laminin, collagen typeIV, and heparan sulfate proteoglycan. In this study, Matrigel concentrations of1.0 and 0.2 mg/ml, diluted 10- and 50-fold with cold distilled water, were used.A conditioned medium, obtained by incubatingNIH3T3 cells for 24 h inserum-free medium (Azinomoto Co., Tokyo, Japan), was used as a chemoattractant. MU (Sigma Chemical Co., St. Louis, MO) was used as a colonmetric assay (22, 24, 25) and dimethyl sulfoxide (Nacalai Tesque, Inc., Kyoto,Japan) was used to dissolve formation crystals of MTF.

Cells. Seven cell lines, NFSaY83, BALB/3T3 clone A31, NIH3T3/14-1,MRC-5, MRC-5 SV1 TG1, Vero, and HeLa, and a primary culture of normalhuman umbilical endothelial cells were used. The NFSa Y83 cell line, provided by Dr. K. Ando, was derived from a fibrosarcoma that had highlymetastatic potential to the lung (26). The BALB/3T3 clone A31 cell linederived from a BALB/c mouse embryo contained fibroblast-like cells withcontact inhibition. The NIH3T3/14-1 cell line was a spontaneous transformantof NIH3T3 without contact inhibition. Both cell lines were provided by RikenCell Bank (Tokyo, Japan) and were maintained in minimum essential mediumEagle's. The MRC-5 cell line, composed of fibroblast-like cells, was derivedfrom normal lung tissue (27); and the MRC-5 SV1 TG1 cell line, composed ofa SV4Otransformant of MRC-5, was provided by Riken Cell Bank. These cellswere maintained in a mixed medium of equal parts of Dulbecco's modifiedEagle's medium and Ham's F-12 nutrient mixture. Vero cells derived from therenal tissue of a normal African monkey (28) were a normal epithelium-likecell line with contact inhibition and were also provided by Riken Cell Bank.HeLa and Vero cells were maintained in minimum essential medium Eagle's.All of the culture media were supplemented with 10% heat-inactivated fetalbovine serum with penicillin (100 units/mi) and streptomycin (100 @ag/ml).

Water-repellent Treatment of Assay Chamber. A Chemotaxicell chamber (Kubota Co., Tokyo, Japan) with a filter 8 mm in diameter, similar to aTranswell chamber (Costar Co., Cambridge, MA) with a filter 6.5 mm indiameter, was used for this study (Fig. IA). The filter was a polyvinylpyrrolidone-free polycarbonate membrane with 8-@.tm pores (Nucleopore). The

water-repellent treatment of a Chemotaxicell chamber with paraffin was performed to prevent the meniscus phenomenon (Fig. 1B). A block of paraffinwas rubbed on the inside wall of this chamber, and its residue was removedusing a cotton swab. This chamber was used as the modified assay chamber inthis study.

Matrigel Coating. The 1.0-mg/mi Matngel was applied on the membranefilters of the Chemotaxicell chambers with the paraffin treatment and was alsoapplied on the membrane filters of the Chemotaxicell chambers without theparaffm treatment. This was repeated using the 0.2-mg/mi Matrigel. The filters

3 The abbreviation used is: MTF, 3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazo

hum bromide.

ABSTRACT

Reconstituted basementmembrane matrix (Matrigel) has been utilized form;@oamayofwmorcelI mvasaonin recentyeat@.In theconventionalchamber forthe invasion amay, however, a large number of celk passed ea@ly through the

center of the Matrigel-coated filter because the Matrigel layer could not becompletely unifonn b@ythe meniscus fonnation To prevent the men@ais p1w-

nomenon of the Matsigel layer, we devised a wateirepellent treatment of theInside wall ofthe @ychamber with paraffin. Consequently, very few erythro

cytespassed through the Matr@el-coated @IterOfthLSmodifiedchamberwith theel7throcyte amay, which wm used todemon.strate theevenness and unifonnlty of

the Matrigd layer on the filter. For quandtating a sinai number ofcells whichInvaded through the Matrigel-coated filter by the invasion @y,a tdrazolium.

based colodmetaic @(4$@methylthIazo1@yf)-2,5-diphenyltdrazo&Imbromide) @ywas used. The inva@ve abilities of the eight diff@rent caib were

determined by 11th 1V@On-3-(4,5-dimmbyIthIaZOIe.2-yI)-2,5.dIp1wnyItetroZohum bromide assay using the modified chamberwith a filtercoated with 70p1 of

the 0.2-nag/miMatrigeLAfter 72 h of incubation, the mnlignnntcell linessignif.Icantly exceeded the normal cell lines in the percentage of invasion (P < 0.01).Therefore, the modifiedIV8SIOU.344,5-duInethyIthIaZOIe-2.yI)-2,5*henyltetramilum bromide amay provides a simpIe@easily reprodudhie fri vifro assay forquantitating tumor cell lnvasion

INTRODUCTION

Tumor metastasis is a multistage process (1) in which the basementmembrane plays a critical role as a barrier against invasion (2, 3). Inthe study of tumor invasion and metastasis, it is thus important toinvestigate the relationship between tumor cells and the basementmembrane (4—7).

Four kinds of connective tissues, the bladder wall (8, 9), amnion(10—12), lens capsule (13), and chicken chorioallantonic membrane(14), have been used to assess the invasion of tumor cells in vitro.Such systems are difficult to standardize and somewhat complicated.To make the more quantitative and reproducible invasion assay, thereconstituted basement membrane has been developed in assay systems under various experimental conditions (15—21). In 1986,Terranova et aL (15) used a pressed disc composed of laminin andtype I and type IV collagens resembling the basement membrane toassaythe invasionof tumorcells in a Boydenchamber.KleinmanetaL (16) reported the development of a reconstituted basement membrane, Matrigel (17). The following year, Albini et aL (18) devised anin vitro procedure, assay tumor invasion using Matrigel in a Boydenchamber. In 1989, Repesh et al. (20) presented a new assay system,which used radiolabeling for quantitating tumor cell invasion, as asimpler way for evaluation of tumor cell invasion using a Transwellchamber with Matrigel. In 1983, on the other hand, Mossmann

Received 2/8/94; accepted 5/4/94.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

I This study was supported in part by a Grant-in-Aid for Scientific Research and for

Cancer Research from the Ministry of Education, Science and Culture of Japan.2 Present address: Department of Surgery, The Johns Hopkins University School of

Medicine, Ross Building Room 754, 720 Rutland Avenue, Baltimore, Maryland 21205.To whomrequestsfor reprintsshouldbe addressed,at First Departmentof Surgery,Kagoshima University School of Medicine, 8-35-1 Sakuragaoka, Kagoshima 890, Japan.

3620

A Modified Invasion-344,5..Dimethylthiazole-2-.yl)-2,5-diphenyltetrazolium Bromide

Assay for Quantitating Tumor Cell Invasion'

Hiroshi Imamura, Sonshin Takao,2 and Takashi Aikou

The First Department ofSurgery, Kagoshinia UniversitySchoolof Medicine, 8-35-1 Sakuragaoka,Kagoshima890, Japan

on May 25, 2021. © 1994 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSAY FOR QUANTITATINGTUMOR CELL INVASION

RESULTS

Uniformity of the Matrigel Layer. In a Chemotaxicell chamberwithout the water-repellent treatment as the conventional assay charnber, the Matrigel layer showed a meniscus-like form during thecoating process, thick at the periphery and thin at the center (Fig. 3,left). Following the water-repellent treatment, however, Matrigelformed an even and uniform layer on the filter (Fig. 3, right). In theerythrocyte assay using the conventional assay chamber, a largenumber of erythrocytes passed through the 1.0-mg/mi Matrigel filtersand the 0.2-mg/mI Matrigel filters. With the 0.2-mg/mi Matrigel,which created an adequate condition for tumor invasion, the meanpercentage of the erythrocytes that fell into the lower well was over5% in all of the different amounts of diluted Matrigel. In the modifiedassay chamber, however, no erythrocytes passed through the Matrigel-coated filter and fell into the lower well with the use of anyvolume of the 1.0-mg/mi Matrigel or 70 to 100 @lof the 0.2-mg/mi

Fig. 1. Conventional and modified chambers for the invasion assay. Chemotaxicellchambers without (A) or with (B) water-repellent treatment were used for the upper wellof the invasion assay. The chamber filter was 8 mm in diameter with 8-sampores and wascoated with Matrigel. A 24-well culture plate was used as the lower well.

were dried overnight under a laminar flow hood to produce the Matrigel-coatedfilters.

Erythrocyte Assay. The erythrocyte assay was used to demonstrate theevenness and uniformity of the Matrigel layer on the filter. Erythrocytes wereseparated from peripheral blood obtained from healthy volunteers. A total of1 X i07 erythrocytes in 200 @alof RPM! 1640 were seeded onto the Matrigelcoated filters of the conventional and modified assay chambers. After 24 h ofincubation, the number of erythrocytes that passed through the filter and fell

into the lower well were counted under the microscope.Invasion-MTT Assay. This assay was carried out using the methods

described by Repesh (20) and Schlechte et aL (23) but with modifications. Themodified Chemotaxicell chambers with water-repellent treatment were used as

the upper wells, and each filter was coated with 70 pi of a 50-fold dilution (0.2mg/mi) of Matrigel which was a required condition for tumor cell invasion.After the invasion assay, the MT1' assay was used for quantitating the invadedcells through the Matngel-coated filter (Fig. 2).

Percentage of Invasion. The percentage of invasion was calculated tocompare the invasive potential of the 8 different cells after the modifiedinvasion-MTf assay using the following formula. A standard curve wascalculated from the absorbance of 1 X 10―cells.

. . Absorbance of invaded cells

% of invasion = x 100Absorbance of seeded cells

Fig. 3. Matrigel layer on the filter. In the conventional chamber (left), the Matrigellayer showed a meniscus form. Consequently, after drying, the Matrigel coating in thecenter of the filter was very thin or had disappeared, leaving small bare spots. In themodified chamber (right), it showed a uniform and even Matrigel layer on the filterwithout the meniscus phenomenon.Statistical Analysis. Student's t test was used for statistical analysis.

A

Matrigel@

Porous

Filter

Fig. 2. Procedure for the invasion-MiT assay. In A, a modified chamber as the upper well was placed into a 24-well culture plate of a lower well. A filter membrane was coatedwith 70 @lof the 0.2-mg/mi Matrigel. As a chemoattractant, 750 @alof NH-13T3-conditioned medium were placed in the lower well. Matrigel was reconstituted with 100 @alof growthmedium, and then 2 x 10@cells were seeded to the upper well with 100 @alof growth medium. The chamber was incubated at 37°Cwith 5% CO2. In B, after 72 h of incubation, thecells with invasive ability invaded through the Matrigel-coated filter, appearing on the lower surface of the filter. In C, the medium in the upper well was removed, and then noninvasivecells on the upper surface of the filter were completely swept away using a couon applicator. Only invaded cells remained on the lower surface of the filter. In D, 20 jal of a 5-mg/mlMTF solution were added in the lower well. After 4 h of incubation, the invaded cells on the lower surface of the filter formed dark blue crystals of MTI'-formazan. In E, the upperwell with formazan crystals was removed from the lower well. In F, it was replaced into another 24-well culture plate as a lower well in which 200 @.dof dimethyl sulfoxide were storedto dissolve the formazan crystals. In G, after 5 ruin, the formazan-dissolved solution was poured into a 96-well microplate, and then the absorbance was measured by a multiple scanningspectrometer ELISA reader (SLT-Laboinstruments, Austria), using a test wave length of 570 nm and a reference wave length of 690 nm.

3621

@!@V!! Matrigel@ Paraffin-treated

on May 25, 2021. © 1994 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSAY FOR QUANTITA11NG11JMOR CELL INVASION

MRC-5 and human umbilical endothelial cells, was very low and

showed no increase in 72 h. The difference of absorbance betweenmalignant and normal cell lines showed the maximum levels after

Fig. 5. Macroscopic views of the erythrocytes (A, B, E, F) or NIH3T3/14-1 cells (C,D, G, H) assay on the uniformity of the Matrigel-coated filter after 24 h (A, B, C, D) and72 h (E, F, G, H) of incubation. The conventional (A, C, E, G) and modified (B, D, F, H)chambers were used. The filter of each chamber was coated with 70 ,.,Jof the 0.2-mg/mIMatrigel. The part of the Matrigel-coated filter where the erythrocytes or NIH3T3/14-1cells passed through appeared as a white area or as dark spots by MTF reaction,respectively. In the conventional chambers, both the erythrocytes and Nll13T3/14-1 cellspassed through the center of the Matrigel-coated filters after 24 h (A, C), and a largeamount of them passed through it after 72 h (E, G). In the modified chambers, very fewerythrocytes passed through the Matrigel-coated filters after 24 h (B, D) and 72 h (F)), andNIH3T3/14-1cellsinvadedthroughitappearingasthescatteredpatternofdarkspotsafter72 h (H).

0.8‘0

@ 0.6

:@0.4

&

40 50 60 70 80 90Volume of Matrigel (p1)

100

Fig. 4. Erythrocyte assay of both Matrigel concentrations of 1.0 mg/mI (0, •)and 0.2mg/mI (0, U) which were used for coating the filter. Horizontalline, amount of Matrigel;vertical line, percentage of erythrocytes which passed through the Matrigel-coated ifiterand fell into the lower well. In the conventional chambers (0, 0), over 5% of erythrocytespassed through the Matrigel-coated filter with any amount of Matrigel except for 90 and100 @alof the 1.0-mg/mIMatrigel.In themodifiedchambers(•,U), theywerelessthan0.02% with any amount of Matrigel except for 50 and 60 @alof the 0.2-mg/nil Matrigel.Each point is the mean ±SD (bars) of three experiments. Error Bars not shown werecontainedwithinthe symbols.

Matrigel. When 50 and 60 @alof the 0.2-mg/nil Matrigel were used, afew erythrocytes passed through the Matrigel-coated filter (Fig. 4).

Macroscopic views of erythrocytes and NIH3T3/14-1 cells on theMatrigel-coated filter of the assay chambers were observed. In theconventional assay chamber, erythrocytes and NIH3T3/14-1 cellspassed through the center of the Matrigel-coated filter after 24 h ofincubation (Fig. 5, A, C). In the modified assay chambers, however,very few erythrocytes fell into the lower wells. They remained equallydistributed on the Matrigel-coated filters. Also, few NIH3T3/14-1cells passed through the Matrigel-coated filters (Fig. 5, B, D). After 72h of incubation, a large amount of both kinds of cells passed throughthe center of the Matrigel-coated filter in the conventional assaychambers (Fig. 5, E, G). With the modified assay chambers, however,the erythrocytes remained evenly on the filter even after 72 h. On theother hand, NIH3T3/14-1 cells showed a scattered pattern of invasioninto the Matrigel-coated filters. The invasion pattern on this filterobviously differed from that of the conventional chamber (Fig. 5, F,H). The water-repellenttreatmentwaseffectivein preventingthemeniscus form of the Matrigel layer and also in making a thin uniformlayer of Matngel.

MT'r Assay. The relationshipbetweenthe numberof BALB/3T3clone Mi cells and the amount of MTF-formazan is shown in Fig. 6.The absorbance was directly proportional to the number of cellswithin the range of 600 to 2 X iO@cells/well. This result indicated thatMiT assay was useful for determining a small number of invadedcells.

Invasion-MTT Assay. In the modified assay chambers, Vero cellsas a normal cell line revealed no increase of invasion after 72 h ofincubation compared to the increase observed during the first 5 h andinvaded into the Matrigel-coated filters sparsely. In the conventionalassay chambers, however, Vero cells passed through the center of theMatrigel-coated filter gradually, and then a large number of cellsseemed to invade into the Matrigel-coated filter after 72 h of incubation (Fig. 7).

This invasion-MTF assay using the modified assay chamber wasapplied to various kinds of cell lines. NFSaY83 and NIH3T3/14-1cells, which were a malignant and transformed cell line, respectively,invaded through the Matrigel-coated filter and their absorbance increased gradually with time. The absorbance of the normal cells,

0.2

0

0 5000 10000 15000 20000

Number of cells

Fig. 6. Relationship between the number of BALB/3T3 clone A31 cells and absorbanceby MTI' assay. This cell line was used as a normal cell line; also very few cells passedthrough the Matrigel-coated filter of the modified chamber. The absorbance was directlyproportional to the number of cells within the range of 600 to 2 X 10@cells/well using anenzyme-linked immunosorbent assay reader. Each point is the mean ±SD (bars) of fiveexperiments.Error bars not shownwerecontainedwithinthe symbols.

3622

on May 25, 2021. © 1994 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSAYFORQUANTITATINGTUMORCELLINVASION

into the lower well. This result indicates that the Matrigel layer on thefilter of the conventional chamber was not uniform and even due tothe meniscus phenomenon. It has been reported that 1% or less oftumor cells have invasive ability in a tumor mass ( 19, 20). If over0.1% of erythrocytes passed through the Matrigel-coated filter, thisinvasion assay would be unsuitable for determining the number ofinvaded cells. As the result ofthe erythrocyte assay using the modifiedchamber, however, the number of erythrocytes that passed through theMatrigel-coated filter decreased markedly to less than 0.02%. Theuniformity of Matrigel coating in the modified chamber was alsoconfirmed by the differences of the invasive pattern between theconventional and the modified chamber using NIH3T3/l4-l and Verocells. These results indicate that the water-repellent treatment of theinside wall of the assay chamber with paraffin was effective forpreventing the meniscus phenomenon and for making the uniformMatrigel coating.

In the conventional invasion assay using Matrigel, the invaded cellswere quantified by a visual counting under the light microscope (15,18, 19) and by assessing the number of radiolabeled cells (20). In1989, Hendrix et a!. (29) reportedthat the visual assessmentwas asgood as the radiolabeled method and was easier to use. However, ithas been difficult to accurately evaluate the invasive abilities ofmalignant cells. A MTT assay described by Mossman (22) in 1983 isa quantitative colonmetric method that uses the activity of the mitochondrial dehydrogenase and the absorbance of the dark blue productformazan. Therefore, this assay is directly proportional to the numberof living cells. Schlechte et al. (23) has applied a MY!' assay to theinvasion assay using the conventional large chambers. In this study, aMU assay was used to obtain the accurate evaluation for measuringthe invaded cells. A problem in applying the MT! assay was that thehigh absorbance of the background would interfere with the estimation of a small number of invaded cells, but the problem was solvedby dissolving the formazan with dimethyl sulfoxide using another24-well culture plate for a MTF reaction. Thus, the modified chamberand MU assay were useful for quantitating the invaded cells accuratelyand quickly.

The invasiveness depended on the concentration and volume ofMatrigel. Using either the l.O-mg/ml Matrigel or at least 80 pi of the0.2-mg/ml Matrigel, the tumor cells could not invade through theMatrigel-coated filter. In this study, the adequate concentration andvolume of Matrigel were 0.2 mg/ml and 70 p.1,respectively. Furthermore, the incubation time also was an important factor for the inva

MRC-5 Hela NFSaY83MRC-5 VEROSV1TG1

Fig. 9. Comparison of the percentage of invasion calculated from the absorbance by theinvasion-MU assay. The percentage of invasion of malignant cell lines (U) was significantly higher than that of normal cell lines (0) (P < 0.01). Columns, means of triplicatedeterminations; bars, SD.

0.5

0.4.

0

0.1-

— I I I J @I@ I I

0 10 20 30 40 50 60 70 80Incubation time (h)

Fig.7. Comparisonof the numberof Verocellswhichpassedthroughthe Matrigelcoated filter of the conventional or the modified chambers by the invasion-MiT assay.Verocellswereusedasa normalcellline.Thefilterofeachchamberwascoatedwith70id of the 0.2-mg/mI Matrigel, and 2 X lO@cells were seeded onto it. After the invasionassay. the absorbance of passed or invnied cells was measured by MU assay. In theconventionalchambers(0), the absorbenceincreasedgradually.In contrast,in themodified chambers (•),absorbance showed no increase in incubation time. Eachpoint isthe mean ±SD (bars) of triplicate determinations. Error bars no shown were containedwithin the symbols.

0.5

0.4

‘i::,@I I I I I I I I

0 10 20 30 40 50 60 70 80Incubation time (h)

Fig.8. Relationshipbetweenthe incubationtimeandtheabsorbanceof the invasionMiT assay. The absorbance ofmalignant cell lines, NIH3T3/14-l (•)andNFSaY83 (U),increased gradually. In contrast, the absorbance of normal cell lines. MRC-5 (0) andhumanumbilicalendothelialcells(0). revealedvirtuallyno changeandextremelylowabsorbance.The differencebetweenthe absorbanceof malignantand normalcell lineswas the most significant after 72 h of incubation. E@h point is the mean ±SD (bars) oftriplicate determinations. Error bars not shown were contained within the symbols.

72 h of incubation (Fig. 8). The adequate incubation time for thisassay was 72 h.

Percentage of Invasion. The percentage of invasion of the malignant cell lines exceeded those of the normal cell lines, and it wassignificant statistically (P < 0.01) (Fig. 9).

DISCUSSION

The uniformity of the Matrigel layer on the filter has been certifiedby protein staining (18—20)and microscopic observation (18, 19). Inthis study, the erythrocyte assay was useful for evaluating the uniformity of a Matrigel layer on the filter. Erythrocytes derived fromhuman peripheral blood are about 7.5 @.unin diameter. Therefore, ifthe Matrigel layer was not coated on the filter uniformly, the erythrocytes could pass easily through some parts of the filter. In theconventional chamber without the water-repellent treatment, a largeamount of erythrocytes passed through the center of the filter and fell

3623

1.

C0Cl)Cu>C

Ca)C,a)0.

on May 25, 2021. © 1994 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

ASSAY FOR QUANTITATING TUMOR CELL INVASION

amniotic membrane: requirement for a proteinase cascade. Cell, 47: 448—498,1986.13. Starkey, J. R., Hosick, H. L., Stanford, D. R., and Liggitt, H. D. Interaction of

metastatic tumor cells with bovine lens. Cancer Res., 44: 1585—1594,1984.14. Marcel, M. M., DeBruyre, 0. K., Vandesande, F., and Dragonetti, C. Immunohisto

chemicalstudyof embryonicchickheartinvadedby malignantcellsin threedimensional culture. Invasion Metastasis, 1: 195—204,1981.

15. Terranova, V. P., Hujanen, E. S., Loeb, D. M., Martin, G. R., Thornburg, L, andGluushko, V. Use of a reconstituted basement membrane to measure cell invasivenessand select for highly invasive tumor cells. Proc. Nail. Aced. Sci. USA, 83: 465—469,1986.

16. Kleinmann, H. K, McGarvey, M. L, Hassell, J. R., Star, V. L, Cannon, F. B., Laurie,G. W., and Martin, 0. R. Basement membrane complexes with biological activity.Biochemistry, 25: 312—318,1986.

17. Kleinmann, H. K., McGarvey, M. L, Liotta, L A., Robey, P. 0., Tryggvason, K., andMartin, 0. R. Isolation and characterization of type IV procollagen, laminin, andheparmnsulfate proteoglycan from the EHS sarcoma. Biochemistry, 21: 6188—6193,1982.

18. Albini, A., Iwamoto, Y., Kleinmann, H. K., Martin, 0. R., Aaronson, S. A., Kozolowski, J. M., and McEwan, R. N. A rapid in vitro assay for quantitating the invasivepotential of tumor cells. Cancer Res., 47: 3239—3245, 1987.

19. Hendrix, M. J. C., Seftor, E. A., Seftor, R. E. B., and Fidler, I. J. A simple quantitativeassay for studying the invasive potential of high and low human metastatic variants.Cancer Left., 38: 137—147,1987.

20. Repesh, L. A. A new in vitro assay for quantitating tumor cell invasion. InvasionMetastasis, 9: 192—208,1989.

21. Kramer, R. H., Bensh, K. K., and Wrong, J. Invasion of reconstituted basementmembranematrixby metastatictumorcells.CancerRes.,46: 1980—1989,1986.

22. Mosmann, T. Rapid colorimetric assay for cellular growth and survival: applicationto proliferation and cytotoxicity assays. J. Immunol. Methods, 65: 55—63,1983.

23. Schlechte, W., Brattain, M., and Boyd, D. Invasion ofextracellular matrix by culturedcolon cancer cells: dependence on urokinase receptor display. Cancer Commun., 2:173—179,1990.

24. Carmichael, J., Degraff, W. B., Gazdar, A. F., Minna, J. D., and Mitchell, J. B.Evaluation of tetrazolium-based semiautomatic colorimetric assay: assessment ofradiosensitivity. Cancer Res., 47: 943—946,1987.

25. Denizot, F., and Lang, R. Rapid colorimetric assay for cell growth and survival:modifications to the tetrazolium dye procedure giving improved sensitivity andreliability. J. Immunol. Methods, 80: 271—277,1986.

26. Ando, K., Hunter, N., and Peters, L J. Immunological nonspecific figure enhancement of artificial lung metastasis in tumor bearing mice. Cancer Immunol., 6:151—156,1979.

27. Jacobs, J. P., Jones, C. M., and Baffle, J. P. Characteristics of a human diploid celldesignated MRC-5. Nature (Lond.), 7: 168—170,1970.

28. Rhim, J. S., Schell, K., Creasy, B., and Case, W. Biological characteristics and viralsusceptibility of an African green monkey kidney cell line (Vero). Proc. Soc. Exp.Biol. Med., 132: 670—678, 1969.

29. Hendrix, M. J. C., Seftor, E. A., Seftor, R. E. B., Misiorowski, P. Z., Sundareshan, andWelch,D. R. Comparisonof tumor cell invasionassay: humanamnionversusreconstituted basement membrane barriers. Invasion Metastasis, 9: 278—297,1989.

30. Noel, A. C., Calle, A., Emonard, H. P., Nusgens, B. V., Simar, L, Foidart, J., Lapiere,C. M., and Foidart, J. M. Invasion of reconstituted basement membrane matrix is notcorrelated to the malignant metastatic cell phenotype. Cancer Res., 51: 405—414,1991.

3624

sion assay. The malignant cells revealed almost no invasion into theMatrigel-coated filter for the first 5 to 24 h of incubation, similar to

the observations of Kramer et aL (21). After 48 h of incubation, theabsorbance of malignant cells increased gradually with time; thenafter 72 h of incubation, the difference in absorbance between thenormal and malignant cells was obvious. These results indicate thatthe tumor cell invasion into the Matrigel layer depends on the incubation time.

The percentage of invasion was useful for comparing the invasiveabilities of various kinds of cell lines in this study. The percentage ofinvasion of malignant cell lines was significantly higher than that ofnormal cell lines. Noel et a!. (30), however, have reported that theinvasion into Matngel was not correlated to the malignant metastaticcell phenotype by microscopic study. Further study concerning therelationship between the metastatic potential and the percentage ofinvasion is now in progress using the modified invasion-MIT assay.

REFERENCES

1. Fidler, I. J. Tumor heterogeneity and the biology of cancer invasion and metastasis.Cancer Rca., 38: 2651—2660, 1978.

2. Nicolson, G. L Cancer metastasis: organ colonization and the cell surface propertiesof malignant cells. Biochim. Biophys. Acts, 695: 113—176,1982.

3. Liotta,L.A.Tumorinvasionandmetastasis:roleofthebasementmembrane.Am.J.Pathol., 117: 339—348,1984.

4. Jones, P. A., and DeClerck, Y. A. Destruction of extracellular matrices containingglycoproteins, elastin and collagen by metastatic human tumor cells. Cancer Res., 40:3222—3227, 1980.

5. Terranova, V. P., Liotta, L A., Russo, R. 0., and Martin, 0. R. Role oflaminin in theattachment and metastasis of murine tumor cells. Cancer Res., 42: 2265—2269,1982.

6. Malinoff, H. L, and wisha, M. S. Isolation of cell surface receptor protein for lamininfrom murine fibrosarcoma cells. J. Cell Biol., 96: 1475—1479,1983.

7. Kramer, R. H., and Vogel, K. G. Selective degradation of basement membranemacromolecules by metastatic melanoma cells. J. Nail. Cancer Inst., 72: 8889—8899,1984.

8. Hart, I. R., and Fidler, I. J. An in vitro quantitative assay for tumor cell invasion.CancerRca.,38: 3218—3224,1978.

9. Poste, G., Doll, J., Hart, I. R., and Fidler, I. J. In vitro selection of murine B16melanoma variants with enhanced tissue invasive properties. Cancer Res., 40: 1636—1644, 1980.

10. Liotta, L A., Lee, w. C., and Morakis, D. J. New method for preparing large surfaceof intact basement membrane for tumor invasion studies. Cancer Left., 11: 141—147,1980.

11. Hendrix, M. J. C., Gehlsen, K. R., wagner, H. N., Rodney, S. R., Misiorowski, R. L,and Meyskens, J. R. In vitro quantification of melanoma tumor cell invasion. Clin.Exp. Metastasis, 3: 221—233,1985.

12. Mignatti, P., Robbins, E., and Rifkin, D. B. Tumor invasion through the human

on May 25, 2021. © 1994 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1994;54:3620-3624. Cancer Res   Hiroshi Imamura, Sonshin Takao and Takashi Aikou  Bromide Assay for Quantitating Tumor Cell InvasionInvasion-3-(4,5-Dimethylthiazole-2-yl)-2,5-diphenyltetrazolium A Modified

  Updated version

  http://cancerres.aacrjournals.org/content/54/13/3620

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/54/13/3620To request permission to re-use all or part of this article, use this link

on May 25, 2021. © 1994 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from