TABLE 2.

Sequence relationships between EspF and the EspF-like open reading frames^a

EspF homolog	Sequence length (no. of amino acids)	No. of proline-rich repeats	% Identify between sequencesb
			EPEC EspF	EHEC EspF	EHEC U-EspF	EHEC M-EspF
EPEC EspF	206	3	100	87	X	X
EHEC EspF	248	4	75	100	X	X
EHEC U-EspF	337	7	24	28	100	97
EHEC M-EspF	189	5	28	27	55	100

^aPairwise global alignment of the predicted amino acid sequences with the Blosum 62 scoring matrix is shown in normal type. Sequence similarity between EspF and U- and M-EspF occurs primarily in the proline-rich sequences. Pairwise local alignment (boldface type) without gaps reveals the closest sequence relationship between the LEE-encoded EspF of EPEC and EHEC and between the U- and M-EspF sequences. Alignments were performed by using the Align program (Scientific and Educational Software, Durham, N.C.)

^bX, absence of extended similarity between the respective sequences.