to BioTechniques free email alert service to receive content updates.
Cost-effective trapezoidal modified Boyden chamber with comparable accuracy to a commercial apparatus
Ruey-Hwang Chou, Kai-Chun Lin, Sheng-Chieh Lin, Ji-Yen Cheng, Cheng-Wen Wu, Wun-Shaing W. Chang
National Health Research Institutes, Academia Sinica, Taipei, Taiwan
BioTechniques, Vol. 37, No. 5, November 2004, pp. 724–726
Full Text (PDF)

The Boyden chamber was designed over 40 years ago (1). It was first designed to study chemotaxis but later modifications resulted in it becoming one of the most popular tools for assessing cell motility and invasion. The classic Boyden chamber consists of two compartments separated by a filter membrane, which, with its micro-porous surface, provides a barrier so that cells cannot pass through except by active migration. For chemotaxis analysis, cells are seeded onto the upper compartment, and the chemoattractant is placed in the lower compartment. After incubation for a period less than the time taken for cells to divide, the degree of chemotactic response can be microscopically determined by counting the percentage of transmigrated cells compared to the original number of seeded cells (1). In order to adapt the Boyden chamber to other applications such as quantitating the invasive potential of tumor cells, various extracellular matrix molecules can be coated onto the membrane. These coating materials can be laminin, collagen, natural basement membrane, or the most widely used substance known as Matrigel™ (BD Biosciences, San Jose, CA, USA; References 2,3,4,5). The porous membrane is occluded by Matrigel, therefore mimicking the extracellular environment. Only invasive cells that have digested the gel will reach the underside of the membrane.

Currently, there are several commercial and in-house versions of the Boyden chamber in use (6,7,8). All chambers are square or rectangular in shape and can be made in either a single- or multi-well format. Apart from a few newly developed but rather expensive commercial products, most of the Boyden assay systems are time-consuming, technically demanding, and cumbersome. They are prone to common errors such as: (i) the Matrigel coating is often irregular or disrupted, resulting in uneven cell invasion; (ii) the process of removal of the inserts sometimes leads to cross-contamination and variability in results; (iii) the recovery of transmigrated cells is notoriously difficult, especially for inexperienced users; and (iv) the manual fixing, cell staining, and cell counting further serve to amplify the inconsistency. As a consequence, many results of the Boyden chamber assays have been shown to vary significantly from one laboratory to another, and even within the same laboratory.

To help circumvent some of these problems, here we describe a few simple but very effective modifications to the Boyden chamber. Our intent is to provide researchers who wish to screen for anti-invasive or anti-metastatic agents a cheap yet accurate and reliable alternative.

First, we have modified the typical design of the Boyden chamber into a trapezoid mode ((Figure 1)A). Alteration of the top plate to 205 mm long and 95 mm wide enabled us to arrange a total of 24 experimental wells (12 mm in diameter per well) to provide a versatile apparatus ((Figure 1), B and C).

Figure 1.

Photographs and diagram of the trapezoid multi-well chamber. (A) The trapezoid shape of the chamber is shown in a lateral view, with the wells and sampling port in perfect alignment. (B) Diagram demonstrating the components of the chamber (displayed in mm units). Note that the experimental wells are arranged in an interlaced mode to save space and avoid cross-contamination of wells. (C) View from above the top and bottom plates of the chamber. (D) Photograph showing the extended length of the exhaust pipe and the broadening of the sampling port. Such a modification prevents cells from overflowing from the exhaust pipe and reduces experimental error. (E) The 2-piece alloy-made coating mold. There is a 0.1 mm difference between the height of the cast and mold that enables a smooth and even coating of Matrigel.

Of all the problems associated with the conventional assay, probably the most critical one is the loss of cells when washing and recovering them for further analyses. This problem is mainly due to the fact that the side-sampling ports are very narrow, causing difficulty in recovering transmigrated cells, and the exhaust pipe is too short (i.e., 10–13 mm in length), which results in the overflow of cell populations during washes and inaccurate cell counts. To solve these problems, we enlarged the bottom plate to 205 mm long and 135 mm wide to make the whole chamber appear as a trapezoid. This design allows us to extend the length of the exhaust pipe by approximately 3- to 4-fold, depending on the position of each well ((Figure 1), C and D). This also allowed us to broaden the diameter of the sampling port in a 45° angle ((Figure 1)D), thus completely solving the overflow problem. Finally, to facilitate a smoother and more homogeneous Matrigel coating, a 2-piece alloy-made coating mold was constructed, as shown in (Figure 1)E.

To examine the efficacy of this trapezoid model, the cell invasion of three tumor cell lines [cervical carcinoma HeLa, lung adenocarcinoma CL1-5 (9), and brain glioblastoma U-87MG cells] was assessed by using our modified apparatus and two existing Boyden assay systems. The study was performed in hexaplicate and repeated twice. For the modified chamber, a sheet of 10 µm polycarbonate membrane (GE Osmonics, Minnetonka, MN, USA) was placed between the upper and lower pieces of the ultraviolet (UV)-sterilized coating mold. A volume of 2 mL Matrigel at 5 mg/mL was then coated onto the membrane and allowed to dry. The dried insert was then removed from the mold and placed over the lower wells of the chamber that had been filled with culture medium. After assembling the upper and lower plates of the chamber, the cells (5 × 104 cells/well) were seeded onto the upper compartments and incubated at 37°C for 24 h. The cells that invaded to the underside of the membrane were thoroughly washed, harvested, and counted as previously described (10). As shown in (Figure 2), the results demonstrate that, with our modified chamber, the measured coefficients of variation for the invasive abilities of HeLa, CL1-5, and U-87 MG cells could be significantly decreased by 58%, 53%, and 65%, respectively, compared with values obtained with the conventional homemade chamber. Furthermore, among the three cell lines tested, the average percentage of error reduced by the trapezoid apparatus (58.7%) and the commercial product (61.7%) differed by only 3% ((Figure 2)). This indicates that our modified chamber can function as an equally accurate alternative system. More importantly, this multi-well trapezoid apparatus costs only approximately 20% of the commercial product.

Figure 2.

The experimental variation for the three different chambers. The in vitro Boyden invasion assays were performed using the unmodified chamber (7,8) (white bar), the modified trapezoid model (black bar), and the FluoroBlok™ invasion system (BD Biosciences Clontech, Palo Alto, CA, USA) (gray bar). Bar graphs represent the mean (⨱) of cells harvested plus or minus standard deviation (±SD) from hexaplicate experiments. The results show that with the modified chamber, the measured coefficient of variation (cv%) for the invasive abilities of HeLa, CL1-5, and U-87 MG cells was reduced by 58%, 53%, and 65%, respectively, compared to those obtained with the unmodified chamber. In addition, among the three cell lines tested, the average percentage of error reduced by our trapezoid apparatus (58.7%) and the commercial product (61.7%) differed by only 3%, indicating that our modified Boyden chamber has comparable accuracy to the commercial apparatus.

In the past few years, several excellent studies have been reported to render the Boyden assays more reproducible and simpler to perform (11,12,13,14,15). These modifications provide important refinements to key methods such as cell visualization and quantification, yet some limitations remain to be resolved. In this study, we have developed an improved chamber that overcomes the particularly troublesome steps of cell washing and recovery. Our model provides the benefits of accuracy, cost-effectiveness, high-throughput, and ease of manipulation. In our view, this reusable trapezoid apparatus, when combined with improvements in cell staining and counting, provides a substantial improvement to assessing cell invasion and migration in vitro.


This work was supported by the National Health Research Institutes (grant nos. 91A1-PPLAB-1 and 91A1-PPLAD-1) and in part by the National Science Council (grant no. NSC91-3112-P-001-045), Taipei, Taiwan. We are deeply indebted to Dr. Yi-Wen Chu and Ms. Yu-Rung Kao for their valuable comments on the design of the chamber. We also thank Dr. Ravi Mahadeva for his critical comments and Ms. Miao-Chong Joy Lin for her assistance in formulating the paper.

Competing Interests Statement

The authors declare no competing interests.

1.) Boyden S., The chemotactic effect of mixtures of antibody and antigen on polymorphonuclear leucocytes, J. Exp. Med., P453 - P466

2.) Kleinman K. H. McGarvey L. M. Liotta A. L. Robey G. P. Tryggvason K. Martin R. G., Isolation and characterization of type IV procollagen, laminin, and heparin sulfate proteoglycan from the EHS sarcoma, Biochemistry, P6188 - P6193

3.) Terranova P. V. Hujanen S. E. Loeb M. D. Martin R. G. Thornburg L. Glushko V., Use of a reconstituted basement membrane to measure cell invasiveness and select for highly invasive tumor cells, Proc. Natl. Acad. Sci. USA, P465 - P469

4.) Terranova P. V. Hujanen S. E. Martin R. G., Basement membrane and the invasive activity of metastatic tumor cells, J. Natl. Cancer Inst., P311 - P316

5.) Livant L. D. Linn S. Markwart S. Shuster J., Invasion of selectively permeable sea urchin embryo basement membranes by metastatic tumor cells, but not by their normal counterparts, Cancer Res., P5085 - P5093

6.) Albini A. Iwamoto Y. Kleinman K. H. Martin R. G. Aaronson A. S. Kozlowski M. J. McEwan N. R., A rapid in vitro assay for quantitating the invasive potential of tumor cells, Cancer Res., P3239 - P3245

7.) Gehlsen R. K. Wagner N. H. Hendrix C. M.J., Membrane invasion culture system (MICS), Med. Instrum., P268 - P271

8.) Hendrix C. M.J. Seftor A. E. Seftor B. R.E. Fidler J. I., A simple quantitative assay for studying the invasive potential of high and low human metastatic variants, Cancer Lett., P137 - P147

9.) Chu W. Y. Yang C. P. Yang C. S. Shyu C. Y. Hendrix C. M.J. Wu R. Wu W. C., Selection of invasive and metastatic subpopulations from a human lung adeno-carcinoma cell line, Am. J. Respir. Cell Mol. Biol., P353 - P360

10.) Shih Y. J. Yang C. S. Hong M. T. Yuan A. Chen W. J.J. Yu J. C. Chang L. Y. Lee C. Y., Collapsin response mediator protein-1 and the invasion and metastasis of cancer cells, J. Natl. Cancer Inst., P1392 - P1400

11.) Bloomfield L. K. Baldwin L. B. Harkin G. D. Tonissen F. K., Modification of the Boyden chamber to improve uniformity of cell invasion of Matrigel-coated membranes, BioTechniques, P1242 - P1246

12.) de la Monte M. S. Lahousse A. S. Carter J. Wands R. J., ATP luminescence-based motility-invasion assay, BioTechniques, P98 - P106

13.) Sasaki Y. C. Passaniti A., Identification of anti-invasive but noncytotoxic chemotherapeutic agents using the tetrazolium dye MTT to quantitate viable cells in Matrigel, BioTechniques, P1038 - P1043

14.) Vipra R. M. Chiplonkar M. J., Vital stain to study cell invasion in modified Boyden chamber assay, BioTechniques, P1200 - P1204

15.) Yamakawa S. Furuyama Y. Oku N., Development of a simple cell invasion assay system, Biol. Pharm. Bull., P1264 - P1266