A concerted performance evaluation and molecular characterization of embryonic stem cells (ESC) will enable us to select cell lines more precisely. However, most of the current ESC studies focus on one or the other approach. Our aims were to compare the targeting efficiencies and gene expression profiles of mouse R1 (Nagy et al. 1993 PNAS 90, 8424–8428) and HM-1 (Selfridge et al. 1992 Somat. Cell Mol. Genet. 18, 325–336) ESCs. Cells were cultured on mitomycin C-treated mouse embryonic fibroblast feeder cell layer and grown in standard ESC medium changed daily [high glucose DMEM (Invitrogen, Carlsbad, CA, USA) supplemented with 0.1 mm 2-mercaptoethanol (Sigma, St. Louis, MO, USA), fetal bovine serum (15% v/v; HyClone, Logan, UT, USA), 1000 U mL–1 murine-LIF (CHEMICON International, Temecula, CA, USA), and antibiotics (penicillin: 50 U mL–1, streptomycin: 50 µg mL–1 (Sigma)]. Kinase-inactive murine Tyk2 (Tyk2K923E) vector was constructed by mutating the invariant lysine in the ATP-binding site of the kinase domain. From both ESC lines, 3 � 106 cells each were nucleoporated using the Amaxa A23, A24, and A30 programs (Amaxa Biosystems, Cologne, Germany), and the percentages of positive clones after PCR and Southern blot analyses were scored. For expression profiling, total RNA was isolated from aliquots of R1 ESC at passage 13 and HM-1 ESC at passage 23 using RNeasy Midi kit (Qiagen, D�sseldorf, Germany) procedures. Fifteen µg of total RNA each from the contrasting samples were used for reverse transcription, labeled with either Cy3 or Cy5 dye (Amersham, Freiburg, Germany), and hybridized to the cDNA arrays containing over 21 000 sequences. The microarray results of four independent hybridizations were analyzed to identify differentially regulated genes, and four randomly selected genes were verified by real-time PCR analyses. There was no significant difference at program A23 for the percentage of positive clones [(2.5% R1) and (2.4% HM-1)]. However, as the program number increased (A24, A30), the efficiency of R1 ESC improved (4.7% and 6%, respectively), while that of HM-1 ESC worsened (2.4% and 0%), and these differences between the cell lines were significant (P < 0.05). The gene expression profiles revealed a number of calcium-binding proteins that were significantly (P < 0.01) down-regulated in the HM-1 ESCs compared to R1 ESC, and these were verified by using independent analyses of real-time PCR that confirmed the microarray results. Thus, the variations in the targeting efficiency can be partially explained by differences in the calcium-binding protein levels of these ESCs. An earlier study (Fierro and Liano 1996 J. Physiol. 496, 617–625) described the tolerance and buffering capacity of other cell types with higher calcium-binding protein levels.
Neuropeptide and calcium-binding protein gene expression profiles predict neuronal anatomical type in the juvenile rat
