M. Cancino-diaz et al., Recognition of B cells epitopes of the Klebsiella pneumoniae GroEL-like protein by HLA-B27 positive subjects, MICROB PATH, 28(4), 2000, pp. 211-220
The presence of antibodies against antigens of K. pneumoniae in HLA-B27 pos
itive patients with ankylosing spondylitis (AS), has been well documented.
We have previously reported that sera from HLA-B27 positive subjects react
with the K. pneumoniae GroEL-like protein (HSP60Kp) and have higher titers
than HLA-B27 negative individuals. We cloned the gene that codes for this p
rotein, determined hydrophilic regions by computer analysis of the predicte
d amino acid sequence and found that residues 389-397, 360-368 and 282-290,
were possible B cell epitopes. To test this prediction, and to determine i
f the HLA-B27 positive and negative AS patients recognize the same or diffe
rent epitopes, we truncated the hsp60Kp gene, from the 3' terminal nucleoti
de, to obtain fragments having or not the predicted epitopes. Four polypept
ides of 40, 37, 30 and 18 kDa were obtained and analysed, by ELISA and inhi
bition of ELISA, for their reactivity with IgG antibodies from three high r
esponders HLA-B27 positive AS patients and three HLA-B27 negative subjects
who recognized the rHSP60Kp. Sera from both HLA-B27 positive and negative s
ubjects reacted equally well with rHSP60Kp or with the 40 and 37 kDa peptid
es, which do not have residues 389-397 and 360-368, respectively, but react
ivity was lost with the 30 kDa peptide, which also lacks residues 282-290.
Contrary to what we expected, antibodies from HLA-B27 negative and positive
individuals recognized the same epitope of the HSP60Kp. Our results indica
te that the important epitope for B cells could be the 282-290 region and t
hat the contribution of the two other predicted regions is minimal. We also
conclude that the differences in response to the HSP60Kp in HLA-B27 positi
ve AS patients and HLA-B27 negative individuals is not qualitative, but onl
y quantitative. (C) 2000 Academic Press.