Using the HTML::Entities documentation:

#!/usr/bin/env perl -l

use strict;
use warnings;
use utf8;
binmode(STDOUT => ':utf8');

use HTML::Entities;

my $unsafe_chars = "<&>'\"[]\200-\377";
my $string = "<[Here's my \"2¢\" worth]>";
print $string;
print encode_entities($string, $unsafe_chars);
[download]

Output:

$ pm_html_ent_plus_brackets.pl
<[Here's my "2¢" worth]>
&lt;&#91;Here&#39;s my &quot;2&cent;&quot; worth&#93;&gt;
[download]

Update: Oops! just noticed  in the output (just before ¢). Fixed by adding:

use utf8;
binmode(STDOUT => ':utf8');
[download]

• use utf8; — because source code contains UTF-8, i.e. the ¢ character.
• binmode(STDOUT => ':utf8'); — doesn't change any of the character entity references but, without it, print $string; gives <[Here's my "2?" worth]> (note the ? instead of ¢) now that use utf8; has been added.