In the original publication of the 1QF6 structure (Sankaranarayanan et al., 1999), the authors described how they learned which specific residues in the anticodon-binding domain were responsible for binding the codon of tRNAs that specify Thr (but not other tRNAs). They included this busy figure, which shows the anticodon-binding domain in red at the bottom. The tRNA is extending down from the top in blue (only part is shown) with the anticodon loop and in particular the sequence U32, U33, C34, G35, U36, [modified]37, and A38 showing.    The anticodon is at positions 34-36, with 35 and 36 being essential for specifying Thr (and 34 being referred to as the 'Wobble' position due to its less essential nature; anticodons 5'-[any]-G-U-3' specify Thr). The entire loop of nucleotides #32-37 were each found to make at least one H-bond with residues in the anticodon-binding domain of the synthetase.   The authors referred to a mutagenesis experiment that involved tRNAs (not the synthetase) - looking to see whether mutation in the anticodon was sufficient to allow a synthetase to recognize a different tRNA. The experiment showed that even if all of the other sequence remains the same, if you switch anticodons (perhaps changing only one nucleotide of the entire tRNA!) you can direct it to a different synthetase and have a different amino acid added.   In that experiment, the two anticodons at positions 34-36 (which means 5' -> 3') were CGU (for Thr) and CAU (for ______).   Take a look at a codon table Links to an external site. to determine which amino acid should fill in the blank (and keep in mind that it's a codon table, not an anticodon table! Figure out which codon binds so that you can use the table...).