Try adding "charset=none" in the connection string for dbremote on all nodes. John's comment that dbremote is assuming OS charset when reading a message is correct, and Volker points to very relevant section of the readme file.

Here's a note to SQL Remote and charsets from the newest v12 EBF (3942).

In my understanding, it doesn't describe a bugfix but a "how to" - and possibly that might work for you, too:

    ================(Build #3850  - Engineering Case #730270)================

SQL Remote always assumes that all databases involved in replication share 
    the same character set. By default, SQL Remote will always apply source CHAR 
    data to a target database using the default character set for the operating 
    system it is running on, ignoring the source data character set.

When using a database character set that is different than the default character 
    set for the operating system, dbremote must be instructed to perform explicit 
    data conversion to that character set on its connection string:

e.g. dbremote -c “CHARSET=utf8;…”

or instruct dbremote to always use the CHAR character set 
    of the target database to apply the remote CHAR data:

e.g. dbremote -c “CHARSET=none;…”

I think you are right. Something is likely interpreting the data as UTF8 on Linux. In cp1252, 'ä' is encoded as 0xE4 which introduces a 3-byte character and that will end up gobbling up the 'r' and the closing quote as part of that character. That's why you see the syntax error. In cp1252 'ü' is encoded as 0xFC which is not a valid lead byte (or follow byte for that matter) so it just gets passed through as a single byte and doesn't cause problems.

I don't know anything about the dbremote or email side of things though. I expect that the emails being sent don't have the encoding specified in the header and therefore the other side is assuming OS charset? Perhaps you can add such a header yourself?