The reason I said I need to add it is that the current code is not specifying it at all. Previously we were using encoding only for DotNetOpenMail but it is a different kind of thing like you would use koi8-r for Russian whereas the .NET ones use either ascii utf8 utf32 or unicode. But currently it is not specified at all so it is always using ascii at the moment.
If you verified that you have valid email addresses both in the to and from and you think the body and subject encoding has any effect on it I can make a build that will use utf-8 utf-32 or unicode and you can see if that solves it but, I do not think it is an encoding issue.
Best,
Joe