Check the following Re^2: splitting data. Changing the code slightly will give you trees based on levels of parantheses.

use strict;
use warnings;

while(my $input = <DATA>){
  print "input: $input\n";
  my $level  = 0;
  my $tab    = "| ";
  my %action = (
        '('       => sub { print "\n", $tab x ++$level, shift },
        ')'       => sub { print "\n", $tab x $level--, shift },
        'default' => sub { print shift },
           );
  
  ( $action{$_} // $action{'default'} )->($_) for $input =~ /./g;
  print "\n\n\n";
}

__DATA__
A (B C D)
A (B C D) (E F G)
A B (C (D E) (F G)) (H I)
[download]

with the following output

input: A (B C D)

A 
| (B C D
| )


input: A (B C D) (E F G)

A 
| (B C D
| ) 
| (E F G
| )


input: A B (C (D E) (F G)) (H I)

A B 
| (C 
| | (D E
| | ) 
| | (F G
| | )
| ) 
| (H I
| )
[download]

not based on regex though... but if you now replace parantheses on level 1 with quotes and put quotes around words on level 0 you should have what you want. Just change the subs in %action as required.