N blocking buffer in 1:100, 1:500, and 1:100 dilutions, respectively. After a gentle wash, the cells were incubated with species- or isotype-specific, AlexaFluor-conjugated secondary antibodies (Life Technologies/Thermo Fisher Scientific) diluted in blocking buffer (1:250). Green (505?25 nm) and far red (>660 nm) fluorescence images of the SGC2085 immunostained samples were acquired at 488 and 633 nm excitations, respectively. An Olympus FV500-IX confocal laser scanning microscope with PLAPO 60?(1.4) oil immersion objective were used for imaging.Detection of apoptotic cellsFor quantitative assessment of apoptosis, the cells were incubated in Annexin V binding buffer containing 20-fold diluted Alexa Fluor 488-labeled Annexin V (Life Technologies/Thermo Fisher Scientific) and 5 M Hoechst33342 (Life Technologies/Thermo Fisher Scientific) for 3 min. At least 3 separate fields of view were acquired for each condition by confocal microscopy using an UPLAPO 40?(0.85) dry objective. Annexin V positive cells and total cell number were determined on the basis of green and blue fluorescence, respectively. The results were expressed as percentage of total cell number. For co-staining of apoptotic cells and ABCG4, the cells were PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/21251029 first subjected to Alexa Fluor 647- conjugated Annexin V (Life Technologies/Thermo Fisher Scientific) (20-fold dilution, 3 min), then fixed and immunostained as described above. To detect caspase-3 activity and Annexin V positivity in same experiment, the cells were incubated with 10 M PhiPhiLuxG2D2 (Calbiochem) and 10 FCS for 20 min, washed twice, then subjected to Alexa Fluor 488-labeled Annexin V in Annexin V binding buffer for 3 min. Results are expressed as mean ?S.E.M. For statistical analysis, Student’s t-test was used to evaluate significant differences. (p < 0.01) in comparison with controls.Results Dimer formation of ABCGTo detect ABCG4 expression, we generated a monoclonal antibody against ABCG4 with an approach that we previously used to raise an antibody against ABCG1, utilizing the GST-fused N-terminal domain of the ABC protein for immunization [24]. The specificity of the monoclonal antibody was confirmed by Western blot analysis with whole cell lysates from HEK293 cells transfected with ABCG1, ABCG2, or ABCG4, as well as from MDCKII cells transducedPLOS ONE | DOI:10.1371/journal.pone.0156516 May 26,4 /Functional Cooperativity between ABCG4 and ABCGwith c-myc-tagged ABCG5 and HA-tagged ABCG8 (Fig 1A). For negative controls, cell lysates from the parental cell line was used. The anti-ABCG4 antibody recognized a single band at the expected molecular weight (app. 60 kDa) in the ABCG4-containing sample, and did not crossreact with any other ABCG proteins. Noteworthy, even ABCG1 was not recognized by the anti-ABCG4 antibody, despite the 72 amino acid identity of these proteins. To explore whether the ABCG4 protein is glycosylated, ABCG4-expressing cells were treated with tunicamycin (5 g/ml, 24 h). Since ABCG2 is known to be glycosylated, ABCG2-expressing cells were used for positive control. In HEK cells, ABCG2 is fully glycosylated as demonstrated in our previous study [27] and also by the appearance of a single band in Fig 1A. We found that tunicamycin treatment altered the apparent molecular weight of ABCG2, but had no effect on the migration of ABCG4 (data not shown), implying that ABCG4 is not glycosylated. This finding is in agreement with previous results using a tagged variant of ABCG4 and glycosydase digestion.