www.raupp.info

The advent of agricultural biotechnology has raised many biosafety
concerns over the past decade. One of the more interesting concerns has
been the potential for horizontal gene transfer (HGT), October 2005 by
C. Neal Stewart, Jr. and Ayalew Mentewab.

The movement of a transgene from plant to microbe could pose a
significant risk, especially if an antibiotic resistance gene,
originally from a bacterium, could be transferred to a pathogenic
bacterium, causing new antibiotic resistance problems for human health.
These concerns have prompted regulators and companies alike to look
askance at the use of antibiotic resistance genes in transgenic plants.

Researchers have expended much energy and resources into developing
alternative transgenic plant production schemes such as herbicide
tolerance, positive selection1, and marker-free selection2. All these
alternatives have prominent drawbacks. After all, approximately 70% of
all transgenic plants have been produced using the neomycin
phosphotransferase II (nptII) gene from Escherichia coli for good
reason: it works, especially in most dicot species.

Besides alternative selection technologies, research has also focused on
numerous technologies to remove markers using site-specific
recombination3. While site-specific transgene removal might well be the
ultimate solution to minimizing exogenous DNA in transgenic plants, it
is far from routine in today?s laboratory. Thus, most researchers and
smaller companies still use antibiotic selectable markers to produce
transgenic plants for at least two reasons: 1) the high efficiency of
using antibiotic selection; and 2) intellectual property constraints and
technology availability of other techniques. Thus, new, and perhaps
safer, selectable markers using established and proven selection schemes
would be attractive for many plant biotechnologists.

We recently described a plant gene that confers kanamycin resistance to
transgenic plants4. In this article we will discuss the relevance of
this new antibiotic resistance gene to discussions of HGT biosafety,
public acceptance and regulatory concerns, and its comparison to nptII.
Many people contend that nptII is safe, but others have recently argued
that if HGT occurred one trillion-fold less often than current risk
assessment literature presumes, HGT could still have negative impacts5.
Thus, the concern over HGT, especially of antibiotic resistance genes,
perhaps warrants a closer look.

There are dozens of ATP binding cassette (ABC) transporters in
Arabidopsis thaliana, and the discovery of a unique function of the ABC
transporter Atwbc19 was quite accidental. We performed microarray
experiments in which A. thaliana was exposed to the explosive chemical,
trinitrotoluene (TNT)6. Atwbc19 was one of several upregulated genes
revealed, and we decided to perform additional experiments to assess the
opportunity to use this gene in transgenic plants for potential
explosives detection and remediation6. We noticed that a T-DNA
insertional knockout mutant did not grow in media containing kanamycin,
so we decided to produce transgenic plants with the ABC transporter,
with and without an nptII cassette, to test whether Atwbc19 could be
used as an nptII substitute. We found that its conferred resistance to
kanamycin (and only kanamycin) was similar to nptII in transgenic
plants, thus leading to the characterization of the first plant gene
endowing resistance to antibiotics4. The initial experiments were in
tobacco, but we are far enough along in experiments to produce
transgenic Brassica species to conclude that it is effective in this
genus as well.

The availability of the Atwbc19 gene could enable efficient and safe
(with regards to HGT) production of transgenic plants. After all, this
gene has presumably been in plants for eons, and there is no evidence
that it has been transferred to bacteria during the course of evolution.
Bacteria, like all organisms, have ABC transporters, but database
searches have yielded no ABC transporter gene hits with plant-like codon
use patterns in bacteria. Indeed, this illustrates one conceptual hurdle
in the entire plant-to-microbe HGT argument: there simply are not many
examples of plant-like genes found in bacteria, indicating that HGT from
plants to microbes is quite rare. The inverse is true, however; that is,
there are well documented examples of bacteria-like genes (and indeed
entire microbial symbionts) being co-opted into the plant nuclear,
chloroplast, or mitochondrial genome. And this is no surprise
really?after all, Agrobacterium tumefaciens and its relatives naturally
transform plants inserting bacterial DNA7.

So, this begs the question, how serious is the threat of HGT from
transgenic plants to bacteria? Transgenic plant-to-microbe HGT has been
shown to occur under experimental conditions when the bacteria already
contained a form of the plant transgene (experimental tricks?see
reference 7 for a discussion), but gene transfer from transgenic plants
to bacteria has never been shown occurring in the field7. Nonetheless,
there is a lack of data on the real rates of HGT between transgenic
plants and microbes, and hence prestigious groups around the world, such
as WHO, FAO, and the NAS, are urging researchers to produce transgenic
plants without antibiotic resistance markers8. Does the availability of
the plant Atwbc19 change anything, or should it be grouped, biosafety
risk-wise, with nptII and other selectable markers of bacterial origin?

While the absolute risk of HGT is unknown, we believe there are at least
four reasons why Atwbc19 would carry relatively less risk than nptII and
other antibiotic resistance genes of bacterial origin. First, Atwbc19 is
very specific for kanamycin. Unlike nptII, it does not confer resistance
to geneticin, neomycin, or other aminoglycoside antibiotics4, which are
used clinically more often than kanamycin. Second, at 2.2 kb, Atwbc19 is
approximately 2.75 times larger than nptII; thus the chances of it being
integrated intact into a bacterial recipient would, at least, be that
much lower. Third, unlike genes of bacterial origin, Atwbc19 has plant
codon usage. Thus, if it were to be introgressed into a bacterial
genome, it would likely be expressed less than a bacterial gene.
Finally, and perhaps most importantly, even if Atwbc19 were transferred
into bacteria and were expressed, it might not lead to an antibiotic
resistant phenotype. This particular factor is not mentioned in the HGT
literature. Our hypothesis is that Atwbc19 is targeted to the vacuole
membrane of the plant cell, and its mode-of-action involves the active
transport of kanamycin into the vacuole where it is sequestered4. While
our data are equivocal with regards to precise targeting, this
mode-of-action is consistent with published data on other ABC
transporters. If this were, indeed, the mode-of-action, we would expect
that Atwbc19 would not confer kanamycin resistance to bacteria since
they lack a prominent central vacuole for sequestration of toxins. The
same is true for mammalian cells. We have unpublished preliminary data
showing that kanamycin resistance is not conferred to Escherichia coli
when the gene is placed under the gal promoter, and we are following-up
these findings with formal experiments with this and other bacterial
species and mammalian cell cultures.

References

1. Reed J et al. (2001) Phosphomannose isomerase: an efficient
selectable marker for plant transformation. Vitro Cell Dev Biol-Plant
37, 127-132

2. de Vetten N et al. (2003) A transformation method for obtaining
marker-free plants of a cross-pollinating and vegetatively propagated
crop. Nat Biotechnol 21, 439-442

3. Baszczynnski CL et al. (2003) Site-specific recombination systems and
their uses for targeted gene manipulation in plant systems, pp157-178 in
Stewart CN Jr (ed) Transgenic Plants: Current Innovations and Future
Trends. Horizon Scientific Press: Wymondham, UK

4. Mentewab A & Stewart C N Jr. (2005) Overexpression of an Arabidopsis
thaliana ABC transporter confers kanamycin resistance to transgenic
plants. Nat Biotechnol 23, 1177-1180

5. Heinemann JA & Traavik T (2004) Problems in monitoring horizontal
gene transfer in field trials of transgenic plants. Nat Biotechnol, 22,
1105-1109

6. Mentewab A, Cardoza V & and Stewart CN Jr. (2005) Genomic analysis of
the response of Arabidopsis thaliana to trinitrotoluene as revealed by
cDNA microarrays. Plant Sci 168, 1409-1424

7. Broothaerts W et al. (2005) Gene transfer to plants by diverse
species of bacteria. Nature 433, 629-633

8. Davison J (2004) Monitoring horizontal gene transfer. Nat Biotechnol
22, 1349


www.checkbiotech.org

------------------------------------------
Posted to Phorum via PhorumMail