Mammalian cell growth versus biofilm formation on biomaterial surfaces in an in vitro post-operative contamination model Guruprakash Subbiahdoss, Roel Kuijer, Henk J. Busscher and Henny C. van der Mei Correspondence Henny C. van der Mei h.c.van.der.mei@med.umcg.nl Received 12 April 2010 Revised 1 June 2010 Accepted 28 June 2010 Department of Biomedical Engineering, University Medical Center Groningen and University of Groningen, PO Box 196, 9700 AD Groningen, The Netherlands Biomaterial-associated infections are the major cause of implant failure and can develop many years after implantation. Success or failure of an implant depends on the balance between host tissue integration and bacterial colonization. Here, we describe a new in vitro model for the post- operative bacterial contamination of implant surfaces and investigate the effects of contamination on the balance between mammalian cell growth and bacterial biofilm formation. U2OS osteosarcoma cells were seeded on poly(methyl methacrylate) in different densities and allowed to grow for 24 h in a parallel-plate flow chamber at a low shear rate (0.14 s ”1 ), followed by contamination with Staphylococcus epidermidis ATCC 35983 at a shear rate of 11 s ”1 . The U2OS cells and staphylococci were allowed to grow simultaneously for another 24 h under low- shear conditions (0.14 s ”1 ). Mammalian cell growth was severely impaired when the bacteria were introduced to surfaces with a low initial cell density (2.5¾10 4 cells cm ”2 ), but in the presence of higher initial cell densities (8.2¾10 4 cells cm ”2 and 17¾10 4 cells cm ”2 ), contaminating staphylococci did not affect cell growth. This study is believed to be the first to show that a critical coverage by mammalian cells is needed to effectively protect a biomaterial implant against contaminating bacteria. INTRODUCTION Biomaterial-associated infections (BAI) can develop from the peri-operative microbial contamination of implant surfaces during implantation, immediately post-surgery during hospitalization or by late haematogenous spreading from infections elsewhere in the body. Both peri-operative and post-operative contamination can cause BAI long after implantation, as bacteria can stay dormant on an implant surface for several years (Singh et al., 2009; Proctor et al., 2006). Micro-organisms involved in BAI are resistant to antibiotics and the host immune system due to their biofilm mode of growth, and biomaterial implants with a biofilm have to be removed in most cases (Gristina et al., 1976, 1990; Habash & Reid, 1999). Irrespective of the route of infection, the fate of a biomaterial implant depends mainly on the outcome of the so-called ‘race for the surface’ between successful tissue integration of the bio- material implant and biofilm growth (Gristina, 1987). If this race is won by tissue cells, then the biomaterial surface is fully integrated by tissue cells and less vulnerable to bacterial biofilms. On the other hand, if the race is won by bacteria, the implant surface becomes colonized by bacteria and tissue cell functions are hampered by bacterial virulence factors and toxins (Gristina, 1987; Gristina et al., 1988). In the concept of the race for the surface, a full surface coverage of a biomaterial in vivo by a viable tissue cell layer, an intact cell membrane and functional host defence mechanisms resist bacterial colonization (Gristina, 1994). Previously an in vitro experimental model for the peri- operative bacterial contamination of implant surfaces was put forward and the effects of bacterial presence on the adhesion, spreading and growth of mammalian cells were determined in a single experiment (Subbiahdoss et al., 2009). The outcome of the race for the surface between contaminating Staphylococcus epidermidis and mammalian cells on glass appeared to be dependent on the number of bacteria present prior to mammalian cell seeding and the absence or presence of fluid flow. The mammalian cells lost the race for the surface in the absence of flow due to the accumulation of bacterial toxins, but were able to grow under flow conditions due to the continuous supply of fresh medium to, and removal of endotoxins from, the interface on all commonly used biomaterial surfaces (Subbiahdoss et al., 2010). Abbreviations: BAI, biomaterial-associated infections; CLSM, confocal laser scanning microscopy; PMMA, poly(methyl methacrylate); TRITC, tetramethylrhodamine isothiocyanate. Microbiology (2010), 156, 3073–3078 DOI 10.1099/mic.0.040378-0 040378 G 2010 SGM Printed in Great Britain 3073