"Microenvironmental contaminations" induced by fluorescent lipophilic dyes used for noninvasive in vitro and in vivo cell tracking.

Determining how normal and leukemic stem cells behave in vivo, in a dynamic and noninvasive way, remains a major challenge. Most optical tracking technologies rely on the use of fluorescent or bioluminescent reporter genes, which need to be stably expressed in the cells of interest. Because gene transfer in primary leukemia samples represents a major risk to impair their capability to engraft in a xenogenic context, we evaluated the possibility to use gene transfer-free labeling technologies. The lipophilic dye 3,3,3',3' tetramethylindotricarbocyanine iodide (DiR) was selected among 4 near-infrared (NIR) staining technologies. Unfortunately we report here a massive transfer of the dye occurring toward the neighbor cells both in vivo and in vitro. We further demonstrate that all lipophilic dyes tested in this study (1,1'-dioctadecyl-3,3,3',3'-tetramethylindotricarbocyanine perchlorate [DiI], DiD, DiR, and PKH26) can give rise to microenvironmental contamination, including when used in suboptimal concentration, after extensive washing procedures and in the absence of phagocytosis or marked cell death. This was observed from all cell types tested. Eventually, we show that this microenvironmental contamination is mediated by both direct cell-cell contacts and diffusible microparticles. We conclude that tracking of labeled cells using non-genetically encoded markers should always be accompanied by drastic cross validation using multimodality approaches.