Publication Date


Document Type


Committee Members

Norma Adragna (Committee Member), Nadja Grobe (Committee Member), Peter Lauf (Advisor)

Degree Name

Master of Science (MS)


The nearly complete inhibition of Na+/K+ ATPase (NKA) in fetal human lens (FHL) epithelial cells by chelerythrine (CHE), a Bcl-2 homology (BH)3-mimetic quaternary benzophenanthridine alkaloid was proposed to be the consequence of CHE binding at a BH1-like hydrophobic groove at the cytosolic aspect of NKA. This conclusion was based on in silico analysis showing at least two motifs in the N-terminal domain of NKA’s a1 subunit- i) aa 59-71 (ARAAEILARDGPN) and ii) aa 42-48 (DELHRKY) homologous to the BH1 and BH3 motifs of Bcl-2 (B-cell lymphoma) respectively, leading to a novel hypothesis that NKA could interact with these proteins by virtue of BH1/BH3 motifs (Lauf et al, 2013). Subsequently, based on the findings of co-immunoprecipitation (co-IP) and immuno-cytochemical (ICC) colocalization, NKA was proposed to directly interact with Bcl-XL (Bcl extra large) and a different spliced Bak (Bcl-2 antagonist killer) isoform (Lauf et al, 2015). In the early studies, NKA pulled down Bcl-XL and a 4 kDa higher molecular weight protein band purported to be an isoform of Bak (MW 23 kDa), however, the reciprocal co-IP of NKA by these Bcl-2 proteins couldn’t be established (Lauf et al, 2015). The initial objective of the present work was to confirm the significance of the BH1 like motif in the a1 subunit of isolated and purified pig renal NKA (prNKA) for its interaction with Bcl-2 proteins. The attempt to demonstrate, by co-IP, the in vitro binding of a Bak eicosapeptide containing a conserved BH3 motif to prNKA (prepared according to Klodos et al, 2002) serendipitously revealed a ~25 kDa band when the immunoblot was probed with a combination of anti-Bak as primary and Fc fragment specific-goat anti-rb IgG as secondary antibody. Though this protein band was initially thought to be prBak, presumably contaminating the preparation as a complex with prNKA, the use of Veriblot secondary antibody against non-reduced IgG could not confirm this assumption revealing instead a ~25 kDa protein which seems to be the light (L) chain of an IgG contaminant of prNKA. Substituting the previously used L chain-cross reactive secondary antibody with Veriblot, the co-IP between NKA and Bak was reassessed in FHL cells. Again, in this preparation, NKA failed to pull down Bak and the previously observed ~27 kDa band in the NKA co-IP lane appears to be L chain. The co-IP between Bcl-XL and NKA in FHL cells was revisited using a similar approach. However, under the present conditions, Bcl-XL did not co-IP with NKA. Because of the absence of Bcl-XL in Bak co-IP (serving as a positive control), the credibility of these findings is diminished and shows the difficulty in establishing the BH1/BH3 motif interaction. Furthermore, NKA could not co-IP Mcl-1, a higher molecular weight anti-apoptotic Bcl-2 protein, from the FHL lysate. However, ICC colocalized again unequivocally NKA, separately, with Bak, Mcl-1, and Bax, as with Bcl-XL before in FHL cells consistent with our previous conclusion that these proteins lie in close proximity possibly interacting with each other. To explain the discrepancy between the co-IP and ICC findings, it is proposed that the putative BH1/BH3 interaction of NKA with Bcl-2 proteins is conformation-dependent requiring optimum ionic and hydrophobic environments and the presence of specific ligands, conditions which are very likely to be compromised in the process of a co-IP technique.

Page Count


Department or Program

Department of Pharmacology and Toxicology

Year Degree Awarded