Figure 14.  The ability of natural and artificial Qdm-analogue peptides to protect T2Q cells from lysis by Qa1R +ve NK cells.
	Qdm	AMAPRTLLL
	P1S	SMAPRTLLL
	P2Q	AQAPRTLLL
	P2L	ALAPRTLLL
	P2V	AVAPRTLLL
	P2K	AKAPRTLLL
	P2T	ATAPRTLLL
	Qdm-k	AMVPRTLLL
	P3S	AMSPRTLLL
	P4A	AMAARTLLL
	P5A	AMAPATLLL
	P6S	AMAPRSLLL
	P7V	AMAPRTVLL
	P8V	AMAPRTLVL
	P8F	AMAPRTLFL
	HLA-A2	VMAPRTLVL
	HLA-G	VMAPRTLFL
	HLA-Cw4	VMAPRTLIL
	HLA-B35	VTAPRTVLL
	HLA-B8	VMAPRTVLL
	mhsp60	GMKFDRGYI
	GroEL	GMQFDRGYL
Artificial or natural [derived from human cI molecules] peptides having single or multiple amino-acid
substitutions compared to Qdm were titrated for their ability to protect T2Q cells from lysis by Qa1R
+ve NK cells.  The figure shows the concentration in nM required for 50% maximal protection [note
the logarithmic scale].
Conclusions
1.  No analogue tested showed greater protective capacity than Qdm, and most were markedly inferior.
2.  The naturally occuring homologue of Qdm, Qdm-k, was as efficient as Qdm in sensitizing T2Q cells
to lysis by the CTL clone d12i [data not shown] but surprisingly was 30 fold less efficient at protecting
T2Q cells from NK cells.
3. The relative ability of analogue peptides derived from human cI molecules to protect T2Q cells from
lysis by Qa1R +ve NK cells was unrelated to their ability to protect against lysis by human NK cells as
reported in the studies of others.
4.  The peptides derived from mouse and bacterial heat shock proteins that are recognized by Qa1-
restricted CTL following infection with Salmonella [Lo et al. Nature Med. 6, 215, 2000] were unable to
protect T2Q cells from Qa1R +ve NK cells at any concentration tested.
5.  Comparison of the protective capacity of substituted Qdm analogue peptides and their ability to bind
to Qa1 as judged from sensitization to the CTL clone d12i suggested that residues at positions 4, 5,
and 8 of the Qdm sequence are important for the recognition of Qa1-Qdm complexes by inhibitory
CD94/NKG2 Qa1 receptors on NK cells.