Summary. Background: Circulating endothelial cells (CECs) have been identified as markers of vascular damage in a variety of disorders, such as myocardial infarction, vasculitis, and transplantation. CD146-driven immunomagnetic isolation has gained widespread use, but the technique is hampered by the lack of a definition of CECs and the absence of a consensus for their enumeration. Aim: To evaluate several variables influencing immunomagnetic isolation of CECs, formulate a definition for CECs and propose a consensus protocol for their enumeration. Methods: We devised a protocol based on CD146-driven immunomagnetic isolation and a subsequent confirmatory step with Ulex-Europaeus-Lectin-1 staining. In a multi-center effort, we evaluated the preanalytical and analytical phases of this protocol. We evaluated the effects of storage, anticoagulation and density centrifugation, and compiled previous experience with this technique. Results: Our protocol permitted unequivocal identification of CECs with acceptable reproducibility. There was an effect of storage time in that median cell numbers declined to only 87.5% of their baseline values during 24 h of storage at 4°C. Recovery was lower with citrate than with ethylene-diamine tetra-acetic acid after 4 h of storage; density centrifugation was also associated with lower recovery. We provide a comprehensive list of technical recommendations and potential pitfalls. Finally, based on our experience with this protocol and a recent consensus workshop, we formulated a working definition for CECs. Conclusion: Our work represents an important step toward consensus regarding the CECs. Our recommendations represent the experience of three major centers and should now be scrutinized by others in the field.

SummaryChanges in gene expression, by application of H 2 O 2 , O 2°-generating agents (methyl viologen, digitonin) and gamma irradiation to tomato suspension cultures, were investigated and compared to the well-described heat shock response. Two-dimensional gel protein mapping analyses gave the first indication that at least small heat shock proteins (smHSP) accumulated in response to application of H 2 O 2 and gamma irradiation, but not to O 2°-generating agents. While some proteins seemed to be induced specifically by each treatment, only part of the heat shock response was observed. On the basis of Northern hybridization experiments performed with four heterologous cDNA, corresponding to classes I-IV of pea smHSP, it could be concluded that significant amounts of class I and II smHSP mRNA are induced by H 2 O 2 and by irradiation. Taken together, these results demonstrate that in plants some HSP genes are inducible by oxidative stresses, as in micro-organisms and other eukaryotic cells. HSP22, the main stress protein that accumulates following H 2 O 2 action or gamma irradiation, was also purified. Sequence homology of amino terminal and internal sequences, and immunoreactivity with Chenopodium rubrum mitochondrial smHSP antibody, indicated that the protein belongs to the recently discovered class of plant mitochondrial smHSP. Heat shock or a mild H 2 O 2 pretreatment was also shown to lead to plant cell protection against oxidative injury. Therefore, the synthesis of these stress proteins can be considered as an adaptive mechanism in which mitochondrial protection could be essential.