The following uses look behind (?<=...) with a match set [KR]and negative look ahead (?!P) that rejects a "following P" match in a split to slice up the protein:

use strict;
use warnings;

my @proteins = qw(
    DAAAAATTLTTTAMTTTTTTCKMMFRPPPPPGGGGGGGGGGGG
    ALTAMCMNVWEITYHKGSDVNRRASFAQPPPQPPPPLLAIKPASDASD
    DAAAAATTLTTTAMTTTTTTCK
    );

for my $protein (@proteins) {
    my @peptides = split /(?<=[KR])(?!P)/, $protein;

    next if @peptides < 2;
    
    print "Protein: $protein\n";
    print "Peptides:\n";
    print "   $_\n" for @peptides;
}
[download]

Prints:

Protein: DAAAAATTLTTTAMTTTTTTCKMMFRPPPPPGGGGGGGGGGGG
Peptides:
   DAAAAATTLTTTAMTTTTTTCK
   MMFRPPPPPGGGGGGGGGGGG
Protein: ALTAMCMNVWEITYHKGSDVNRRASFAQPPPQPPPPLLAIKPASDASD
Peptides:
   ALTAMCMNVWEITYHK
   GSDVNR
   R
   ASFAQPPPQPPPPLLAIKPASDASD
[download]