We read with great interest the article by Naito and colleagues,¹ who recently proposed the autoantibodies against M3 muscarinic acetylcholine receptor (anti-M3R) as a new diagnostic marker for patients with Sjögren’s syndrome (SS).

We have been studying anti-M3R recently² and reviewed some theoretical aspects.³ The results of our work with the same 25-mer synthetic peptide (K-R-T-V-P-P-G-E-C-F-I-Q-F-L-S-E-P-T-I-T-F-G-T-A-I) as used by Naito et al showed a similar prevalence of anti-M3 in patients with SS (table 1⇓). Nevertheless, we did not draw the same conclusions and could not agree with the statement that antibodies against the 25-mer synthetic peptide might be a new diagnostic marker for SS.

We believe that the authors should mention a misleading fact in the article by Bacman et al,⁴ which was discussed by Cavill et al and Dawson et al^5,6—namely, the sequence of 25-mer synthetic peptide used by Bacman et al was in fact the amino acid sequence from the second extracellular loop of M4 muscarinic acetylcholine receptor. Neither of the two groups were able to detect the activity of anti-M3R with conventional immunological approaches. Furthermore, Gao et al constructed a CHO cell line expressing the human M3R gene and found positive anti-M3R antibodies in 9/11 patients with SS and in none of 11 healthy controls.⁷

The enzyme linked immunosorbent assay (ELISA) used by Naito et al was somewhat similar to our procedure. In our ELISA, Costar medium binding microtitre plates were coated with the same 25-mer peptide in absolute ethanol (10 mg/l), and incubated at 4°C for at least 3 hours. Serum samples were first diluted 1:100 in 1% bovine serum albumin (BSA)/phosphate buffered saline (PBS) (Naito et al 1:50 in 5% BSA/PBS) and incubated for 1 hour at room temperature (Naito et al for 2 hours at 37°C). Optical density values of anti-M3R were not normally distributed in any of our tested groups. Therefore, the cut off value was estimated at the 95th centile of 349 controls. Neither sensitivity nor specificity of the ELISA for SS was improved by binding the synthetic peptide to BSA by a cross linker (N-(γ-maleimidobutyryloxy)succinimide ester; Pierce).

In conclusion, it seems that the 25-mer synthetic peptide used in routine immunological techniques^2,5,6 does not disclose clinically relevant antibodies, suggesting that a short linear peptide does not depict an adequate epitope for the binding of anti-M3R. Data presented by Gao et al, applying native M3R protein, seem far more promising, but they should be verified on a larger group of patients and controls.