Scientists and companies could be understandably reluctant to allow their popular products to be compared directly with less widely used ones, let alone to toil away in efforts that might boost a competitor's status. To put it bluntly, how would I feel if a whole group of scientists found that my favorite embryonic stem cell (ESC) line didn't perform particularly well? What about lines that my colleagues or company had derived or had a commercial interest in?

Yet these are the risks taken by the members and supporters of the International Stem Cell Initiative (ISCI) who recently completed a collaborative characterisation of nearly five dozen ESC lines from 11 countries and 17 laboratories (1).

Since launching that project, ISCI has grown to include more countries and laboratories. Even though these are not the problems most biologists dream of solving, coordinating funding, intellectual property, and shipments across disparate time and legislative zones is time-consuming and intellectually demanding. Still, we do all want to know whether we can really believe the results of our experiments. In my field, this means knowing in what ways various human ESC lines are essentially the same and how and what makes them behave differently. Will the experiment performed on one line hold across various lines created in different laboratories under different conditions across the world?

Our first project validated commonly used markers and gene expression across 59 cell lines and turned up a few differences as well. A recently launched project will determine which media work best for which lines. Another project is planned to catalogue the abnormalities that various ESC lines tend to accumulate over time. This could help us understand why abnormalities occur as well as how to detect and prevent them more easily.

In January 2004, the International Stem Cell Forum approved approximately $800,000 for comparing a variety of ESC lines. Next came many, many discussions. We had to set up multiple agreements. All ESC lines examined had to have been derived according to common ethical guidelines. Many of the lines tested are owned by universities or companies. Each had to consent that ISCI data would be made public even though, theoretically, the project could produce data showing some lines would be inferior.

Then we had to allocate tasks and establish common protocols and sources of material to make sure results on different lines were comparable. Invitrogen supplied all the media. Applied Biosystems designed a custom, low-density array for assessing gene expression. Chemicon provided one of the antibodies. Gene Service Ltd carried out the gene expression analysis. Various academic groups agreed to carry out a range of activities on behalf of the consortium. These included studies of imprinting, testing for contamination by microorganisms, histopathology of xenograft tumors, and statistical analyses of the data.

One key concern was the antibodies used to define the cells. The hybridomas producing these antibodies were owned by different institutions around the world, and had often been licensed to various supply companies. We persuaded the owning institutions to archive a stock of each of those hybridomas for future reference at the UK Stem Cell Bank, and to allow the Bank to provide a common preparation of each antibody to all the laboratories participating in the ISCI project. The Bank also acted as a hub to receive various materials from the participating laboratories and to pass them on to the laboratories that carried out specific tests for the consortium as a whole.

Planning the meeting made me a bit nervous. I worried discussion could turn acrimonious if the data we expected to be similar varied in different people's labs.

Once ISCI had launched its first project, the participants plus some others met at the Jackson Laboratory in Bar Harbor, Maine, to look at the preliminary results and review progress. Planning the meeting made me a bit nervous. Rather than formal scientific presentations, we held informal workshops. I hoped we would be able to look at each other's cell data cooperatively, but I worried discussion could turn acrimonious if the data we expected to be similar varied in different people's labs. There was also the worry that if we got people together for freeform chats nothing would happen. But it did!

Broadly speaking, all the lines expressed similar patterns of antigens and genes. Although the study could not ascertain whether or not lines showed more subtle differences—such as in their capacity to differentiate—no subgroups of lines with substantially different phenotypes were identified. The results form the basis for the public ISCI database located on the Forum's website . Plans for phase II of ISCI include making provisions for additional ES cell lines to be evaluated in the same way and added to the database. In fact, one of ISCI's goals is make the registry self-sustaining.

Collaborating across countries and laboratories takes more than phone calls and emails between scientists. ISCI is funded by ISCF (www.stemcellforum.org), which was established in 2003 by a group of national medical research funding agencies to encourage international collaboration that supports stem cell research. The current ISCF membership encompasses 21 agencies from 19 countries. The UK's Medical Research Council chairs the Forum and provides the secretariat. The impetus to establish the Forum came from those agencies who wanted to accelerate research in this area, to share rare and high-profile resources, and to spread best practice. The ISCF links funders and policy makers, and it brainstorms ways to help realize the potential of stem cell biology. Apart from ISCI, the Forum also funds projects related to ethics, intellectual property, regulation, and banking of lines.

Funding agencies in each country must hew to their own policies. Coordinating regulations requires creativity and patience. We must consider not just which laboratories are best equipped to carry out experiments, but also which are allowed to receive funding and experimental supplies. Projects had to be shaped to take account of the different positions of each country. For example, the United States has a restrictive regulatory environment, while the United Kingdom has clear, permissive policies; Singapore has experience generating and maintaining lines under a company production model.

At its last meeting in Singapore, the Forum approved funding for additional work. The $2 million phase II of ISCI to evaluate media and track chromosomal abnormalities is now underway. The initiative is organized by a scientific steering group comprising myself at the University of Sheffield in the United Kingdom, Nissim Benvenisty of the Hebrew University of Jerusalem, Barbara Knowles at the Jackson Laboratory in the United States, Ron McKay at the US National Institutes of Health, Steve Oh from A*STAR in Singapore, Martin Pera at the University of Southern California, Janet Rossant at the Hospital for Sick Children in Toronto, Henrik Semb at Lund University in Sweden and Glyn Stacey at the UK Stem Cell Bank.

As with our first project, all participants will provide assurance that the lines they study have been ethically sourced and will also agree to place their results in the public domain. Meanwhile a third workshop will be held at the Jackson Laboratory in Bar Harbor in October 2007 to track progress so far and to consider possible plans for future studies.

Unlike the meeting that launched the first ISCI project, I am not afraid this time that scientists will find little to say.