Anti-C/EBPβ (E299) antibody from abcam (ab32358) detects three C/EBPb isoforms in myoblasts. To validate the abcam anti-C/EBPβ antibody in myogenic cultures, we performed western blot of C2C12 myoblasts that were retrovirally transduced to express the LAP isoform (C/EBPβ-LAP), the LIP (C/EBPβ-LIP) isoform, all C/EBPβ isoforms (β) or with empty virus (pLXSN). The antibody successfully detected endogenous LAP, and the LAP (~36 kDa) and LIP (~17 kDa) isoforms in C/EBPβ-overexpressing cells (Fig. 1A).
Anti-C/EBPβ antibody (ab32358) detects a non-specific band in differentiating myoblasts. We next evaluated the ability of ab32358 to detect endogenous C/EBPb expression in proliferating and differentiating myoblasts. C/EBPβ expression is highest in proliferating myoblasts, and is rapidly downregulated after induction to differentiate in low serum conditions. C2C12 myoblasts were expanded in growth medium (GM) for 48 hours then differentiated in differentiation medium (DM) for 96 hours. Consistent with our previous findings9–11, C/EBPb-LAP is the most predominantly expressed isoform in myoblasts and its expression decreased 24 hours after induction to differentiate (Fig. 1B). However, beginning at 48h of differentiation, a strong band, far more intense that the signal for C/EBPβ, was detected at ~23 kDa and this band was not detected in proliferating myoblasts even with longer exposure time (Fig. 1B). According to the data sheet for the antibody and BLAST analysis of the provided epitope, the antibody could detect C/EBPa and C/EBPε. However, the molecular weight of the observed band does not overlap with C/EBPa (42 and 30 kDa) or C/EBPε isoforms (32, 27 and 14 kDa). To determine if the 23 kDa band represented a novel C/EBPb isoform, we examined the expression of this band in myoblasts overexpressing C/EBPb (Fig. 1C). Over the course of differentiation, the band was only detected in empty vector control cells at 72 hours of differentiation, while it remained undetectable in cells overexpressing C/EBPβ (Fig. 1C). We next examined the expression of this band in primary myoblasts isolated from a conditional knockout mouse (cKO) in which Cebpb is excised in satellite cells. Satellite cells were cultured in growth medium for 48h and switched to differentiation medium for 48h (the time course for differentiation of primary myoblasts is shorter than in C2C12 cells). Knockout efficiency was confirmed by western blot (Fig. 1D) and C/EBPb-LAP expression in WT cells was downregulated with differentiation as previously reported9,10(Fig. 1D). Interestingly, the 23 kDa band was detected in differentiating WT and cKO myoblasts ruling out the possibility that this band is an isoform of C/EBPb (Fig. 1D). Since C/EBPb is an inhibitor of myogenesis9,10, the detection of the 23 kDa band correlates with myogenic differentiation (detected only in differentiating myoblasts) and not with C/EBPβ expression (Fig 1C and D).
ab32358 detects MYL4 in differentiating myoblasts. To identify the protein causing the 23 kDa band in differentiating myoblasts, we performed an immunoprecipitation (IP) of whole cell extracts from C2C12 myoblasts differentiated for three days using the ab32358 antibody or non-specific IgG. The 23 kDa band was successfully precipitated using the anti-C/EBPb antibody but not by the control IgG as detected by silver staining (Fig. 2A, red box). Western blot analysis of the input and the C/EBPb-IP sample confirmed the pull down of the 23 kDa band, and its absence in the control IP lane (Fig. 2B). The excised 23 kDa band was analyzed by mass spectrometry, which identified 16 mouse proteins with molecular weights between 19-23 kDa (Fig. 2C). Based on the spectrum counts, myosin light chain proteins (MYL4, MYL1/3 and MYL12b) were detected at higher levels than others. Similarly, myosin light chain proteins were more highly ranked based on the percentage of amino acids detected by the spectrum. Myosin light chain 4 was detected with 11 exclusive unique peptides, 13 exclusive unique spectra and 72 total spectra with 66% coverage (Fig. 2C). Myosin light chain 1/3 skeletal muscle isoform (MYL1) was detected with 11 unique peptides, 19 unique spectra, 66 total spectra and 82% coverage. MYL12b was also identified with 7 unique peptides and 40% sequence coverage. We used publicly available microarray data from proliferating (GM) and differentiated myoblasts (GSE24811)12 to determine the expression pattern of the myosin light chain genes. In parallel with myogenin and myosin heavy chain 8 and 3 (markers of terminal differentiation), Myl4 and Myl1 were upregulated with myogenic differentiation (Fig. 2D). Myl12b gene expression remained stable with differentiation (Fig. 2D).
To confirm the identity of the band detected with the ab32358 antibody, we incubated the membrane from Fig. 2B with antibody against MYL4, confirming the detection of MYL4 protein in the C/EBPb-IP sample at a similar level to the input (Fig. 2E). Furthermore, a western blot comparing extracts from undifferentiated myoblasts (GM), differentiated myotubes (D3), extracts from HEK293 cells (human) and recombinant myosin light chain proteins was performed (Fig. 3). The anti-C/EBPβ antibody failed to detect recombinant MYL12B, but did recognize recombinant MYL4 protein, as well as a band in HEK293 cells (Fig. 3). Interestingly, while HEK293 cells are not thought to express MYL4, human MYL4 is known to be slightly larger than the mouse protein. Loading of recombinant proteins was verified using an antibody that detects multiple isoforms of MYL (MYL1, MYL3, MYL4 and MYL6), which also detected a band in differentiated C2C12 extracts but not HEK293 cells, or against MYL12A/B. The anti-C/EBPβ antibody failed to recognize recombinant MYL1 (data not shown). Thus, these findings suggest that the previously unidentified band detected by ab32358 in differentiating myoblasts corresponds to MYL4.