RESEARCH ARTICLE Summary. The preferential detection of cells with intact membranes by sample treatment with propidium monoazide (PMA) in combination with PCR amplification is gaining in popularity. This study evaluates the effect of PMA on 454 pyrose- quencing profiles of environmental water samples from a canal in Amsterdam and seawater (with sediment) left untreated or exposed to elevated temperatures (50, 60, or 85°C) for 10 min. Community analysis was based on the extraction of genomic DNA followed by PCR amplification of 16S rRNA genes using universal bacterial primers. Whereas the highest temperature in combination with PMA treatment completely suppressed PCR amplification, PCR products from the other samples were subjected to massively parallel tag sequencing. PMA treatment did not substantially affect the sequence profiles of non-heated samples, but heat exposure resulted in a clear difference in the relative proportions of certain groups. This difference was sig- nificantly more pronounced in heated seawater than in heated canal water. The effect of the chosen experimental conditions on the membrane integrity of cells was supported by BacLight LIVE/DEAD staining in combination with flow cytometry, which confirmed an increase in the uptake of propidium iodide in samples exposed to high temperatures. [Int Microbiol 2010; 13(2):59-65] Keywords: live/dead cells distinction · viability · propidium monoazide · 454 pyrosequencing Introduction The lack of information on cell viability is a major shortcom- ing of methods based on DNA analysis. The reason lies in the persistence of DNA for significant time periods after cells have lost their viability, depending on the environmental con- ditions [7,12]. When a suspension of heat-killed Salmonella cells is introduced in a seawater microcosm (initial cell con- centration of 10 5 –10 6 cells/ml), DNA is detectable for up to 10 days in summer seawater kept at 20°C and up to 55 days in winter seawater kept at 10°C, as measured by PCR [5]. Free DNA spiked into these microcosms persists for a shorter period, but is still detectable for 3–8 days at 10°C and for 2–4 days at 20°C. Similar results were presented by Novitsky [17] regarding the persistence of microbial biomass DNA contained in marine beach sand. Biomass in the sand was labeled with radioisotopes and killed by the addition of chlo- roform, followed by the addition of untreated sand and/or seawater [17]. The degradation rate of DNA has been esti- mated at 5–16% per day. INTERNATIONAL MICROBIOLOGY (2010) 13:59-65 DOI: 10.2436/20.1501.01.111 ISSN: 1139-6709 www.im.microbios.org *Corresponding author: F. Schuren TNO Quality of Life Business Unit Food and Biotechnology Innovations Microbial Genomics Group Utrechtseweg 48. Zeist, Netherlands Tel. +31-306944930. Fax +31-306944466 E-mail: frank.schuren@tno.nl Andreas Nocker, Tim Richter-Heitmann, Roy Montijn, Frank Schuren,* Remco Kort TNO Quality of Life, Business Unit Food and Biotechnology Innovations, Microbial Genomics Group, Zeist, Netherlands Received 28 December 2009 · Accepted 29 May 2010 Discrimination between live and dead cells in bacterial communities from environmental water samples analyzed by 454 pyrosequencing