Automated Colony Counting: A Critical Requirement for Advanced Cancer Research

The colony counting method has long been recognized as the primary method for measuring the effects of radiation, chemotherapeutic drugs, and other agents on cell viability. However, increased emphasis on combinational studies has significantly increased the number of plates that need to be counted, and three-dimensional in vitro culture systems have made it more difficult for researchers to take an accurate count. Advanced cancer research requires a system that allows for fast, objective counts in 3-D assays with the flexibility to store and share data with other research facilities.

Issues with manual counting

Manual counting is a substantially time-consuming task. It is not uncommon for a count of 100 plates to take in excess of 9 hr to complete, and some assays may require the analysis of thousands of plates.¹ Not only does the count itself utilize expensive hours of researchers’ time, but those responsible for conducting the manual count need to be adequately trained, which in itself is expensive and time-intensive. Unfortunately, for all the time, money, and effort it takes to conduct such research, the results are potentially unrepeatable and often strewn with errors.

The manual counting procedure¹ requires the researcher to:

• Decide to accept or reject an object, depending on whether or not it fits the criteria of a colony
• If it is a colony, determine whether its size (diameter or maximum chord) exceeds the predetermined minimum size to qualify it for acceptance
• If these conditions are satisfied, add one more count to the running total and proceed to the next object.

Counting colonies by eye presents several key problems to the researcher. It is difficult, if not impossible, to determine the size of an acceptable colony. In the size range of interest set, the number of colonies changes very rapidly with size. In practice, “this means that an error of only ±10% in the visual estimate of colony size can result in +90% to –45% counting error. In visual counts, since small differences in size are very difficult to determine accurately, this problem can contribute to very significant counting errors.”¹

Figure 1 – GelCount automated colony counting system.

Optical distortion is another important problem. Most assays investigated in laboratories are two-dimensional; however, in “a living body, tumour growth occurs in three dimensions—that is, it involves anchorage-independent growth, which is actually one of the most important measures of tumorgenicity.”² Most laboratories must therefore assess anchorage-independent growth using soft agar (agarose) colony formations, by measuring cell proliferation through the manual counting of colonies in semi-solid culture media. Furthermore, despite the 3-D nature of tumor growth, most assay count analysis is confined to the flat portion of the agar surface, as the meniscus at the petri dish walls may cause optical distortion.¹

Figure 2 – Multiple-well cell culture plate set in the GelCount automated colony counting system.

Manual counts can also be severely compromised by the investigative bias of a researcher whose expectations about the outcome of the experiment can consciously or subconsciously affect the counting process. More simply, basic human error plus physical and mental fatigue can seriously affect the accuracy and reproducibility of results, particularly on a large count. Indeed, investigative fatigue is such a serious problem that “errors of 100% or more have been observed when two people count the same plate.”¹

Benefits of automation

Automating the detection, counting, and analysis of mammalian cell colonies offers significant benefits to cancer biologists processing tumor colony-forming assays. The most obvious advantage is the increase in speed and accuracy.

Objective and consistent “machine” counting is significantly faster than subjective and labor-intensive manual counting. A large assay might take up to 9 hr, whereas an automated count would take a fraction of the time. For example, an automated colony counting system such as GelCount™ (Oxford Optronix Ltd., Abingdon, U.K.) (see Figures 1 and 2) typically takes 5 or 6 min per four 6-well plates for adherent colonies. Even nonadherent colonies (in soft agar or equivalent) typically only take 12–15 min per four 6-well plates. Clearly, a machine does not suffer from fatigue, confirmation bias, or any other problem related to subjectivity.

In addition to the significant reduction in time and cost that result from automated colony counting, there are other, less obvious advantages. For instance, some equipment does not require the removal of dish or plate covers during automated counting; thus culture sterility can be maintained and the assay continued.

The automated image acquisition and analysis approach to colony counting has been validated and found to provide higher accuracy, reproducibility, and precision of results than manual counting, “which has been shown to have a significant degree of intra- and inter-observer variability, with coefficients of variation (CV) ranging from 8.1 to 40.0% and 22.7 to 80%, respectively. Standard CVs for cell viability assessment and progenitor (colony) type enumeration have been shown to range from 19.4 to 42.9% and 46.6 to 100%, respectively.”³

Features to consider when selecting an automated colony counting device/additional features

Data archiving

Automation also delivers unprecedented data archiving and retrieval capabilities. For example, colony counts and statistics can be automatically exported to a spreadsheet, while sample images can be stored in a generic bitmap format for printing, presentations, and so on.

In addition, the integrated approach permits raw colony images to be archived to disk, for a permanent archive of colony samples and in support of full-functionality “off-line” processing/reprocessing, even where the original samples are no longer available.

Now research laboratories can visually document findings quickly or submit captured images to multiple research teams or external agents for independent validation and/or processing and assessment. This ability to capture, process, and export data on- or off-line gives laboratories a fast, cost-effective way to perform objective and rigorous double-blind tests, and extends their ability to participate in global or large-scale trials or pursue new collaborative research endeavors.

Another major advantage of automation technology is that the anchorage-independent cultures can be imaged repeatedly without cell staining and therefore without damaging the colonies. This makes it possible for true growth curves and time-dependent effects to be quantified using colony number and colony size by area or by volume.¹

Conclusion

With its dramatic effect on throughput, automated colony counting is opening up new possibilities for ambitious new projects that will advance the frontiers of cancer research, including large-scale, combinational studies.

References

1. Kressner, B.E.; Morton, R.R. et al. Use of an image analysis system to count colonies in stem cell assays of human tumors; http://www.ncbi.nlm.nih.gov/pubmed/6736686.
2. Kajiwara, Y.; Panchabhai, S. et al. A new preclinical 3-dimensional agarose colony formation assay; http://www.tcrt.org/A-New-Preclinical-3-Dimensional-Agarose-Colony-Formation-Assay-p-329-334-p16759.html..
3. Standard Test Method for Automated Colony Forming Unit (CFU) Assays—Image Acquisition and Analysis Method for Enumerating and Characterizing Cells and Colonies in Culture, Active Standard ASTM F2944.

Dr. Michael Rau, Ph.D., is Director of Sales and Marketing, Oxford Optronix Ltd., 19-21 East Central, 127 Olympic Ave., Milton Park, Abingdon OX14 4SA, U.K.; tel.: +44 (0) 1235 821 803; fax: +44 (0) 1235 921 678; e-mail: [email protected] .