In vitro molecular structure of N-terminal B-type natriuretic peptide: monomer or oligomer?

In the circulation, the A-type natriuretic peptide and the B-type natriuretic peptide (BNP) regulate cardiovascular homeostasis. Each hormone is derived from biosynthetic precursors that are processed to the active C-terminal hormones and the N-terminal fragments (1). Measurements of both BNP and N-terminal proBNP in plasma are recommended as diagnostic tools for heart failure. The peptides are measured by immunoassays that use antibodies directed against specific epitopes. A basic understanding of the native peptide structure in terms of posttranslational modifications is thus paramount, because changes in posttranslational processing will affect assay measurement and clinical interpretation drastically. A recent report in Clinical Chemistry addressed the possibility of natriuretic peptide oligomerization in a study that used a recombinant N-terminal proBNP peptide (possessing a glycine–serine N-terminal extension) spiked into 3 matrices (2 ). Buffer, plasma, or serum was incubated for various times and temperatures; the material was then applied to a reversedphase column, a gel-filtration column, and a reversed-phase column, respectively. All 3 columns were eluted with acidic acetonitrile. Antibodies directed against linear epitopes (residues 1–22, 10 – 29, and 57–76) were used to assay fractions for N-terminal proBNP immunoreactivity with the intent of delineating the peptide sequences. We note that the immunoreactivity of endogenous N-terminal proBNP from unspiked plasma does not yield the same response with the different assays [see Fig. 1A of (2 )], suggesting that the antibodies possess different affinities for full-length N-terminal proBNP. We believe this possibility is worth considering, given that immunoreactivity differences are a primary means the authors used to infer analyte structure. In other words, how are the different affinities taken into account in the elution profiles? An experiment showing the relative affinities for full-length N-terminal proBNP in buffer alone seems warranted. The authors also note from Fig. 1 in their report that chemical deglycosylation decreases the immunoreactivity for all of the studied epitopes, as well as for N-terminal pro–A-type natriuretic peptide. We suggest this result is due to Nto O-acyl migration in the presence of the harsh acid they used, which likely could be reversed by treatment with base. Alternatively, enzymatic deglycosylation may mitigate these deleterious effects. The matrix-incubation experiments were used to evaluate the effect of protease activity on N-terminal proBNP. It is, however, unclear whether stable fragments could simply be a consequence of either the absence of protease-sensitive sites or the presence of sites that might require proteases not found in the matrix of study. Our major concern is that the authors conclude that a region of N-terminal proBNP oligomerizes and that this region contains the stable domain deduced from the matrix-incubation experiments. This conclusion is derived from the c immunoactivity results from the columns, including mass spectrometric data for the fractions from the third column. When gelfiltration columns are eluted with 300 to 500 mL/L acetonitrile, the ensuing denaturation destroys noncovalent protein–protein interactions (3 ). Thus, the eluted material described in the authors’ Fig. 3B is extremely unlikely to retain quaternary structure. We have tested the effect of denaturing conditions on the elution of synthetic proBNP 1–76 (Fig. 1). Elution under benign conditions produces an aberrant elution pattern that is sometimes interpreted as oligomerization but is more likely due to a highly unfolded, disordered structure (4, 5). The addition of 6 mol/L guanidine in this case apparently produced a slightly more compact structure and thus a later elution. Furthermore, rigorous biophysical measurements have revealed only monomeric N-terminal proBNP for solid-phase synthesized 1–76 N-terminal proBNP (4) and methionine-1–76 N-terminal proBNP (5). We note that each peptide gave a single band on a protein-stained denaturing gel (data not shown), in contrast to the authors’ antibodystained immunoblot of recombinant N-terminal proBNP [Fig. 4C in (2)]. It would be informative to perform an additional Western blotting experiment with the other antibodies they used. By their very nature, immunoassays are defined by the choice of antibodies. Our ability to extract deFig. 1. Superdex 75 chromatographic profile for synthetic 1–76 proBNP eluted with neutral phosphate buffer (black circles) or 6 mol/L guanidine-HCL (gray circles). Arrow indicates elution of the 12.5kDa protein. Clinical Chemistry 57:6 000 – 000 (2011) Letters to the Editor