Reply to “Letter to the editor: Comments on Stuart et al. (2016): ‘Myosin content of individual human muscle fibers isolated by laser capture microdissection’”

reply: Dr. Schiaffino is a highly respected and experienced investigator who has been very involved in muscle fiber type characterization for nearly three decades. In his comments (2) about our previous publication (4), he made several points. He is concerned about our data showing that specific muscle fibers contain a mixture of myosin heavy chain isoforms, and that most fibers contain detectable amounts of MYH4 and MYH6.

Unfortunately, some confusion over detailed techniques was expressed in the letter. There were concerns included about the mass spectroscopy method including total proteins rather than unique peptides, but in fact, all of our mass spectroscopy data were from the number of unique peptides identified (see page C384 of ref. 4). Additionally, our methods stated that each sample analyzed by mass spectroscopy contained ∼300 transversely cut sections lifted from microscope slides, adding up to 2.0 mm² area (10 μm thick). In materials and methods, we described a variation of the laser capture microdissection (LCM) technique for quantifying mRNA content of fibers from unstained sections. We found that the staining technique caused a major decrease in quantifiable mRNA, so for fiber samples for mRNA, alternate sections were stained to identify fiber type and the unstained slide was the source of samples where mRNA degradation was minimized. All LCM samples for mass spectroscopy and Western blots were obtained directly from dual-stained slides, after we verified that the staining technique did not alter the mass spectroscopy data or the Westerns.

The finding in our report that individual fiber types contain more than one MYH isoform is not unique in that others have found similar results (3, 5), including the paper by Marta Murgia and coworkers (1). Murgia isolated individual fibers by teasing them out of soleus and EDL muscles excised from mice. The identification of fiber type in their studies was inferred from the mass spectroscopy-determined myosin isoform content (their Supplemental Methods). They, in fact, rejected identification of fibers that included too many isoforms. This approach is circular and will tend to confirm their bias.

When they looked at two halves of the same fiber, the MYH isoform content was quite variable in some fibers (1). The table included in the letter from Schiaffino and coworkers actually shows unique peptide primary data that are quite similar to our report and it does not show fibers that express only one MYH isoform. The calculated MYH content proportions in the bottom section of the table do not match the data in the top section for number of total and unique peptides. Their unique peptide data show that the type 1 fiber contained 48.8% MYH7, 12.5% MYH2, 20.0% MYH1, and 18.8% MYH4. The type 2B fiber had 16.5% MYH7, 17.4% MYH2, 16.5% MYH1, and 49.5% MYH4. It is unclear how the total/unique peptide data can lead to the calculated MYH content percentage data. Reviewing the methods described in Murgia's supplemental files did not provide an explanation for the nearly single isoform expression in the type I and type IIB fibers that they report.

Our calculations of percentage composition of each LCM-collected specific fiber type sample were based on the unique peptides for each isoform being representative of the isoform content. This is a reasonable approximation since the trypsin-generated unique peptides from these four myosin heavy chain isoforms represent 19, 21, 24, and 27% of the entire amino acid sequences. In contrast, unique peptides from the myosin light chains are from a larger portion of the full-length proteins (57-88%). Making adjustments to the fiber isoform percentage composition by these differences in unique peptide content (which we did not do in the paper) would have only a small effect and would still result in clear evidence of mixed MYH and MYL isoform content across all of the fiber types.

The presence of MYH4 and MYH6 in specific fiber samples apparently concerned Schiaffino, but the Western blots and the mass spectroscopy studies independently corroborated the presence of these isoforms across all three human fiber types. In contrast to the immunogens that were used to generate the monoclonal antibodies used in the Behan method of fiber typing, the antibodies used for MYH1, MYH2, MYH4, MYH6, and MYH7 Western blots were all supposed to be raised to proteins specific to each of these isoforms. However, since our report (4), we have indeed found that the “specific” antibodies obtained from Sigma did cross react across more than one MYH isoform. As Schiaffino and coworkers pointed out, the antigen used to generate in rabbits an anti-MYH6 antibody was a 143 amino acid peptide that was 87% identical to the corresponding sequence of MYH7 protein. We, therefore, must agree that the immunoblots showing evidence of MYH6 in Figs. 2 and 4 could be due to cross-reaction with MYH7. However, MYH6 was identified as unique peptides in the mass spectroscopy data of Figs. 3, 5, 6, and 7, indicating that this specific isoform of myosin heavy chains was indeed present in these skeletal muscle samples. MYH4 was also clearly present in the samples analyzed by mass spectroscopy unique peptides, even though the MYH4 antibody might cross-react with other isoforms in the immunoblot studies.

Finally, the pH dependence of the ATPase reaction used to identify fiber types over many years was difficult to attribute to one MYH isoform because of the mixtures seen across the fiber types. Our explanation was highly speculative but does fit the data. Verification of this speculation or of another explanation awaits additional molecular investigations.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

C.A.S. drafted manuscript; C.A.S., M.F.B., W.L.S., and M.H.S. edited and revised manuscript; C.A.S., M.F.B., W.L.S., and M.H.S. approved final version of manuscript.