Spread of Laboratory-generated Antibiotic Drug Resistance in Chinese Rivers

A generation ago, Jeremy Rifkin famously convinced the Cambridge city council to ban genetic engineering in that city, using the fear of "tinkering" with nature and producing a "superbug" that would hurt the public health and thereby (for a time) reduced progress in biological research at Harvard and elsewhere in the city environs.  Mr. Rifkin was not alone in his concern over potential problems from recombinant DNA; indeed, the scientist who first produced a recombinant plasmid, Paul Berg (Jackson et al., 1972, "Biochemical Method for Inserting New Genetic Information into DNA of Simian Virus 40: Circular SV40 DNA Molecules Containing Lambda Phage Genes and the Galactose Operon of Escherichia coli," Proc. Natl. Acad. Sci. U.S.A. 69(10): 2904-09), was instrumental in organizing a conference in 1975 on the issue at Asilomar, California, which resulted in guidelines (both physical and biological) from the National Institutes of Health for the safe practice of the new technology (Guidelines for research involving recombinant DNA molecules," 41 Fed. Reg. 27911-43 (1976)).  But in the intervening decades, widespread distribution of laboratory-derived recombinant DNA introduced into the environment has not been reported as a threat to humans or the environment.

CoverUntil now, that is.  In a report in the journal Environmental Science and Technology entitled "A Survey or Drug Resistance bla genes Originating from Synthetic Plasmid Vectors in Six Chinese Rivers," scientists from Sichuan University and the Chinese Institute of Health and Environmental Medicine report that drug resistance genes (specifically, the beta-lactamase (bla) gene that provides resistance to ampicillin) have been detected in the Pearl, Yangtze, Yellow, Hai He, Sunggari, and Huangpu rivers in mainland China (Chen et al., 2012, Environ. Sci. Technol. 46(24): 13448–54).  These rivers drain the majority of the Chinese mainland in all provinces and have the cities of Guangzhou, Nanjing, Ji'nan, Tianjin, Harbin, and Shanghai, respectively, along their banks.  The authors took samples from these rivers containing naturally occurring bacteria and assayed using polymerase chain reaction (PCR) and a variant, quantitative real-time PCR, to assess the frequency with which beta-lactamase DNA could be detected.  The assay was specific for the bla gene that comprises most recombinant plasmid strains, such as pBR322 (one of the oldest and most widespread plasmids used for molecular biological research) as well as pUC19.  PCR was performed using three pairs of "universal" primers for the bla gene as well as sequencing primers used for detecting exogenous DNA in cloning sites in the plasmid vectors.  In addition, the bacterial isolates obtained from the river samples were assayed for antibiotic resistance using assays recommended by the U.S. Clinical Laboratory Standards Institute.

Plasmid-derived bla-encoding DNA was detected in all six rivers assayed, detected using both the "universal" primers as well as the plasmid sequencing primers (the latter results further supporting the conclusion that the samples were of laboratory origin).  The detection rate varied from 21.9% (in the Hai He River samples) to 36.4% (in the Yangtze River samples), and the total copy number of bla copies in the samples ranged from 6.7 +/- 5.0 x 101 copies/mL (in the Yellow River samples) to 4.6 +/- 0.6 x 103 copies/mL (in the Pearl River samples).  The Pearl and Hai He rivers showed the widest spectrum of cephalosporin resistance from the bla gene present in bacterial samples, extending to 3rd- and 4th-generation drugs like cefotaxime and cefoperazone, while the spectrum was narrower (e.g., cefalotin, cephazolin, cefmetazole, and cefoxitin) in samples from the other rivers tested.  Sequence analysis confirmed that sequences "neighboring" the bla sequences detected in the river samples "most frequently represented artificial or synthetic constructs, including cloning, expression, shuttle, gene-fusion, and gene trap vectors" derived from recombinant laboratory plasmid vectors, most strongly with pBR322.

Es-2012-02760s_0006Abstract, Chen et al., 2012.

An interesting ancillary result reported in the paper was that tetracycline resistance-conferring DNA was also detected in samples from all six rivers, and some river samples also showed genes that provided resistance to other commonly used laboratory antibiotic selection markers such as gentamicin.  Detection of gene sequences encoding tetracycline resistance is consistent with detection of pBR322, which was engineered to contain a tetracycline resistance gene as well as the bla gene.  The authors also report that detection of high levels of laboratory plasmid-derived antibiotic resistance genes in the Pearl River was consistent with the levels of antibiotic pollution know to exist in that river from human and animal sources.

This report raises serious questions, of course, over the consequences of widespread use of antibiotic resistance-encoding genetic elements.  It is particularly disturbing because genetic engineering is not limited to the lab anymore, being used in such technologies as biofuels, agriculture, and bioremediation.  The risk of antibiotic resistance gene contamination of the environment has been recognized previously and steps taken to minimize it; for example, Monsanto has developed vectors encoding other markers such as lacZY for use in its recombinant seed.  These efforts take on greater significance, and relevance for human and animal well-being, in view of this latest report.