Troubles in IL-8 measurement of tissue cells cultured with E.coli - (Nov/27/2008 )

Hi,
I’m now doing the measurement of IL-8 level of cell culture medium with a IL-8 ELISA kit (Porcine CXCL8/IL-8 Quantikine ELISA Kit, Cat # P8000) which is from R&D system.
What I did is: culture the tissue cells in 6 well plates overnight which will reach 50% confluence, co-cultured 10000000000 and 100000000 of E.coli with tissue cells in 1 ml medium for 1h, 2h and 3h. After incubation, I collected the medium, centrifuge at 300xg for 5 min (discard cell debris), collected supernatant, then did another centrifuge at 6000xg for 5 min (discard bacteria). Finally I did IL-8 measurement.
The result was: IL-8 in medium which cultured tissue cells with 1000000000 E.coli was always lower than control cells (which means the medium with tissue cells only). It’s so strange that the medium with co-culture of tissue cells and 1000000000 E.coli should be at least the same as the control, why it’s always lower at each time point?
I was really confused about this. Has anyone have any experience of this?
Thanks,
Jing

Although I don't do measurements of IL nor grow cells...could your experimental line be producing a product that competes for binding and is not recognized by the ab of the enzyme conjugate?

If this is not the case you may be able to disregard the lower values if you can run many different samples at higher concentrations and use the intercept and slope (of a methods comparison plot) as offsets to determine your product concentrations.

good luck.