Mathematical Modeling and Optimization Methods for De Novo Protein Design - Systems Biology: Networks, Models, and Applications - page 56

composition of the parent peptide sequence. Moreover, since the variation
in the residue composition for position 11 is predicted to be rather
broad, position 11 was restricted to be arginine in subsequent
sequences (except set C). The first set of sequences was constructed to
better analyze the effect of the tyrosine substitution at position 4, with
the justification to focus on this substitution being an attempt to assess
the unusually dominant selection of tyrosine at position 4. The consis-
tent element of the sequences belonging to set A is the assignment of
tyrosine to position 4. To further isolate any substitution with respect
to the parent peptide sequence, sequences A1, A2, and A3 assume the
parent compstatin composition of histidine at position 9. Moreover,
sequence A1 resembles the parent peptide sequence at positions 1 and
13 as well, while sequences A2 and A3 are constructed so as to add the
valine substitutions incrementally, first at position 13 for sequence A2
and then at both positions 1 and 13 for sequence A3. Sequences A1 and
A3 exhibit substantial increases in fold stability over the parent peptide
sequence (table 2.2). These results highlight the significance of the
tyrosine substitution at position 4, and may help to further clarify cer-
tain features of the proposed binding model for the compstatin-C3
complex [73].
To further explore the combination of position 9 substitutions with
the presence of tyrosine at position 4, several additional sequences
were constructed. The B1 and B2 constructions represent a reduction in
the number of simultaneous mutations from the parent peptide
sequence. In effect these two sequences correspond to the individual
combinations of sequence A2 with both sequence A4 and sequence A5
Table 2.2 Sequence and experimental relative activity of compstatin analogs
with improved activity that were identified by rational design,
experimental combinatorial design, and the novel in silico de novo
protein design approach
Relative
Peptide Sequence activity Reference
Compstatin I[ CV V QDWG HHR C ]T-NH 2 1 [70]
Ac-Compstatin Ac-I[ CV V QDWG HHR C ]T-NH 2 3 [72]
Ac-H9A Ac-I[ CV V QDWG AHR C ]T-NH 2 4 [73]
Ac-I1L/H9W/T13G Ac-L[ CV V QDWG WHR C ]G-NH 2 4 [74]
Ac-I1L/V4Y/H9F/T13V Ac-V[ CV Y QDWG FHR C ]V-NH 2 6 [24]
Ac-I1L/V4Y/H9A/T13V Ac-V[ CV Y QDWG AHR C ]V-NH 2 9 [24]
Ac-V4Y/H9F/T13V Ac-I[ CV Y QDWG FHR C ]V-NH 2 11 [24]
Ac-V4Y/H9A/T13V Ac-I[ CV Y QDWG AHR C ]V-NH 2 14 [24]
Ac-V4Y/H9A Ac-I[ CV Y QDWG AHR C ]T-NH 2 16 [24]
Ac-V4W/H9A Ac-I[CVWQDWGAHRC]T-NH 2 45 [81]
Boldface is used to indicate that amino acids were fixed. Brackets indicate the disulfide
bridge. Relative complement inhibitory activity is derived from IC 50 measurements