Science in Society Archive

Bio-remediation Without Caution

A bacterium living inside plants could be improved for cleaning up environmental pollutants without genetic modification. Prof. Joe Cummins and Dr. Mae-Wan Ho reveal that this seemingly beneficial development is beset with danger, as the bacterium concerned is a known pathogen.

Water soluble and volatile organic environmental pollutants, such as benzene, toluene, ethylbenzene and xylene compounds, chlorinated solvents and nitrotoluene ammunition wastes, are being cleaned up using plants in combination with microorganisms that naturally live inside the plants (endophytes).

Endophyte bacteria live within the tissue of the plant without harming it. They are found in most plant species, and many can colonize the vascular system. The highest densities of bacteria are usually found in the roots, less in the stem, and least in the leaves. The plants take up the pollutants through their roots, and the bacteria break these down within the roots or in other parts of the plant.

This natural process is inefficient because the compounds tend to get transported up the plant faster than the bacteria can break them down. Once transported up, the plants metabolize the contaminants, and some of the metabolites as well as the contaminant can be toxic. For example, trichloroethane is metabolized into trichloroacetic acid, both of which are toxic. Worse still, plants tend to release volatile pollutants and their metabolites into the atmosphere via evaporation from the leaves, which turns bio-remediation into bio-pollution.

A recent article in Nature Biotechnology reports how phytoremediation could be greatly improved by engineering an endophyte bacterium Burkholderia cepacia [1], a natural resident of the yellow lupine. Researchers from Linburgs University in Diepenbeek, Belgium and Brookhaven National Laboratory in New York, USA, created a strain of B. cepacia that has enhanced ability to degrade toluene within the plant, enabling the plant to tolerate high levels of toluene, and also substantially reduced the amount of toluene released into the atmosphere.

The engineered strain of bacterium carries marker genes for kanamycin resistance and nickel resistance and is derived from the natural endophyte. By adding to this endophyte strain a toluene-degrading plasmid from another strain of B. cepacia that normally lives in the soil through natural conjugation (bacterial reproduction) between the strains, a new endophyte strain is created that can live in the plant and degrade toluene taken up by the plant [1,2].

Plants inoculated with the engineered bacterium grew much better than plants that were not inoculated; or else inoculated either with the control strain lacking the plasmid, or with the strain that normally lives in the soil. More impressively, the plants inoculated with the engineered bacterium reduced toluene evaporation into the atmosphere to about 50% of the control. This looks very promising, as the researchers point out, the experiment could have been done without any genetic modification. The plasmid containing all the toluene degrading enzymes belonged to a natural soil bacterium, and an endophyte host without the marker genes could easily have been used to receive the plasmid by conjugation.

A non-GM bacterial endophytic strain created in this way may well be the very first really useful and beneficial product from the industry. So what's wrong?

The research paper did not address obvious safety issues, such as, what metabolites of toluene are generated in the plant, and whether they are toxic, nor how the plants are to be disposed of. There are three lupine species cultivated for fodder - blue, white and yellow - and along with these there are a number of wild species. The wild species contain alkaloid chemicals that are very toxic to cattle and sheep while the cultivated species are edible for farm animals, provided care is taken to treat the seeds in such a way as to remove the toxins. Lupines thrive on poor soil and provide ground cover and green manure as well as fodder for animals [3].

More importantly, the research report failed to mention that B. cepacia has the ability to cause fatal disease in humans.

The groundwater of Wichita, Kansas was found to be polluted with the chemical solvents dichloroethylene and trichloroethylene, and was remediated using a natural strain of B. cepacia [4]. Neither special public health measures nor epidemiological follow up seemed to have been implemented after the remediation.

The United States Environmental Protection Agency (EPA) has considered the problems associated with approval of B. cepacia as a plant pesticide; for, not only is the bacterium used to fight plant pests but is itself a pest as it is a disease agent in humans. EPA, through a Scientific Advisory Panel (SAP), reviewed B. cepacia as a plant pesticide and acknowledged that it is linked to human disease [5, 6]. The SAP risk assessment peculiarly noted "Bc [B. cepacia] has been referred to as an opportunistic human pathogen. However, as might be expected, the strains registered or proposed for use as biopesticides were isolated from the soil or plant roots, rather than from human patients"[7].

In reality, the SAP comment offered cold comfort because the B. cepacia strains isolated from patients proved essentially undistinguishable from strains isolated from the roots of crops such as corn. The American Phytopathological Society has produced a useful review of the risks from plant disease or human disease along with the benefits in remediating chemical pollution and fighting some plant diseases [8]. Unfortunately, there has been no clear and simple way to differentiate between the 'evil' and the beneficial strains of B. cepacia, and no way of preventing the two from exchanging genes.

B. cepacia has an unusual genetic makeup; it has a relatively large amount of DNA (about twice that of E. coli) and unlike most bacteria, which usually have a single chromosome, B. cepacia strains have as many as five large replicons (chromosomes) and the different chromosomes are rich in insertion sequences that allow for extensive gene exchange between different strains, and insertion of disease related genes from other bacterial species [8]. B. cepacia is a prominent cause of death among cystic fibrosis patients, the bacterium frequently reaches epidemic proportion among such patients and an epidemic related strain was identified in soil samples in USA [9]. It is believed to be a complex species made up of seven distinct genomic subspecies all of which are capable of infecting humans. Even though genomovar III (subspecies) is most frequently associated with epidemics, all of the disease-related genomovars were isolated from maize rhizosphere (root zone) [10]. The disease is difficult to contain because disease bacteria may be replenished continually from the soil and plant material.

Nosocomial (hospital based) B. cepacia epidemics appeared among patients with diabetes, malignancy, heart failure and chronic obstructive pulmonary disease [11]. A nosocomial B. cepacia outbreak appeared in an intensive pediatric care unit [12], and B. cepacia infection was common among renal transplant patients [13]. Different B. cepacia clones showed differing infectivity among cystic fibrosis patients and patients with different complaints [14,15]. Antibiotic resistant B. cepacia infection was the most common cause of death among lung transplants for cystic fibrosis patients [16]. B. cepacia causes feared infections because the strains tend to be antibiotic resistant. Bacteria isolated from different infections were found to be resistant to all seven tested antibiotics but were sensitive to treatment with honey [17].

Do lupines pose a threat to people with compromised immune systems or cystic fibrosis? Yellow lupines, and perhaps the other commercial species as well, contain potentially disease-causing B. cepacia endophytes so their presence in hospitals and homes of compromised people is unwise. The bacteria may be transferred by direct contact with broken plant stems or petals along with the dust and debris associated with the plant; a gift of lupines could be fatal.

There is clearly a large literature on the threat of B. cepacia infection and its death toll among compromised patients. The existing evidence indicates that the bacterial infections may pass from the ecosystem to the hospital ward and there seems no way of ensuring that the B. cepacia strains used in biotechnology are unable to infect compromised humans.

Article first published 25/05/04


  1. Barac T, Taghavi S, Borremans B, Provoost A, Oeyen L, Colpaert J, Vangronsveld J and van der Lelie D. Engineered endophytic bacteria improve phytoremediation of water-soluble, volatile, organic pollutants. 2004 Nature Biotech online 2004, 11 April doi:10.1038/nbt960
  2. Glick BR. Teamwork in phytoremediation. Nature Biotech 2004, 22, 526-7.
  3. Grieve M. " A modern Herbal: Lupins" 1995 (Correction/Update - 2/16/01) pp1-6
  4. Mostellar D, Borguin A and Olsen,Camp Dressler and McKee Inc. Aerobic bioremediation of TCE contaminated ground water: bioaugmentation with Burkholderia cepacia PR301. Bioremedation Kansa B 1997, pp1-4
  5. United States EPA "Report FIFRA scientific advisory panel meeting july 20-23,1999 session1-Burkholderia cepacia : risk assessment of a biopesticide with afffinities to a human opportunistic pathogen" 1999 SAP Report No. 99-04
  6. United States EPA "Risk assessment of Burkholderia cepacia based biopesticides and other pathogens related to oppoetunistic human pathogens"
  7. Parke J. Burkholderia cepacia :friend or foe? The education center of the American Phytopathological Society 2000, pp1-8
  8. Wigley P and Burton N. Multiple chromosomes in Burkholderia cepacia and B. gladioli and their distribution on clinical and environmental strains of B. cepacia. J. Applied Microbiology 2000, 88,914-8
  9. LiPuma J, Spilker T, Coenye T. and Gonzalez C. An epidemic Burkholderia cepacia complex strain identified in soil. Lancet 2002, 359, 2002-3
  10. Fiore A, Laevens S, Bevivino A, Dalmastri C, Tabacchioni S, Vandamme P and Chiarini L. Burkholderia cepacia complex: distribution of genomovars among isolates from the maize rhizosphere in Italy. Environmental Microbiology 2001, 3, 137-43
  11. Huang C, Jang T, Liu C, Fung C, Yu K and Wong W. Characteristics of patients with Burkholderia cepacia bacteremia. J Microbiol Immunol Infect. 2001, 34,215-9
  12. Bureau-Chalot F, Piednoir E, Pierrat C, Santerne B and Bajolet O. Nosocomial Burkholderia cepacia outbreak in an intensive pediatric care unit. Archives de Pédiatrie 2003, 10, 882-88
  13. Li F, Chan K, Chan T and Lai K. Burkholderia urinary tract infection after renal transplantation. Transplant Infect Dis 2003, 5,59-61
  14. Agodi A, Barchitta M, Giannino V, Collura A, Pensabene T, Garlaschi M,Pasquarella C, Luzzaro F, Sinatra F, Mahenthiralingam E. and Stefani, S. Burkholderia cepacia complex in cystic fibrosis and non-cystic fibrosis patients: identification of a cluster of epidemic lineages. J.Hosp.Infect. 2002, 50,188-95
  15. Manno G, Dalmastri C, Tabacchioni S, Vandamme P, Lorini R, Minicucci L, Romano L, Giannattasio A, Chiarini L.and Bevivino A. Epidemiology and clinical course of Burkholderia cepacia complex infections, particularly those caused by different Burkholderia cenocepacia strains, among patients attending an Italian Cystic Fibrosis Center. Journal of Clinical Microbiology 2004, 42,1491-7.
  16. Dobbina C, Maleyb M, Harknessc J, Bennd R, Maloufe M, Glanvillee A and . Byea P. The impact of pan-resistant bacterial pathogens on survival after lung transplantation in cystic fibrosis: results from a single large referral center. Journal of Hospital Infection 2004, 56, 277-82.
  17. Cooper R, Wigley P and Burton N. Susceptibility of multi resistant strains of Burkholderia cepacia to honey. Letters in Applied Microbiology 2000, 31, 20-4.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a tripod?