To test this hypothesis, Coviella and Trumble grew cotton plants in twenty 3-liter pots in each of six Teflon-film chambers in a temperature-controlled greenhouse, three of which chambers were maintained at an atmospheric CO2 concentration of 370 ppm and three of which were maintained at 900 ppm CO2.  In addition, half of the plants in each chamber received high levels of nitrogen (N) fertilization, while half received low levels (30 as opposed to 130 mg N/kg soil/week).  After 45 days, leaves were removed from the plants and dipped in a solution containing Bacillus thuringiensis, after which known amounts of treated leaf material were fed to Spodoptera exigua larvae and their responses measured and analyzed.  These protocols revealed that plants grown in the elevated CO2 chambers did indeed have significantly lower foliar nitrogen concentrations than plants grown in the ambient CO2 chambers under the low N fertilization regime; but this was not the case under the high N regime.  Also, older larvae fed with foliage grown in elevated CO2 with low N fertilization consumed significantly more plant material than those fed with foliage grown in ambient CO2; but, again, no differences were observed in the high N treatment.  In addition, and "consistent with the effect of higher Bt toxin intake due to enhanced consumption," it was found that "insects fed on low N plants had significantly higher mortality in elevated CO2," but again, no such effect was evident in the high N treatment.  As a result, Coviella and Trumble concluded that "increasing atmospheric CO2 is making the foliar applications more efficacious," especially in the case of soils low in nitrogen.

In addition to applying Bt toxins to plants by spraying, the ability of crops to produce them internally has been achieved by means of genetic engineering; but there is concern that foreign genes from agricultural plants may be transferred into wild relatives of transgenic crop lines and thereby upset the "balance of nature."  In one of the first studies to address this concern, Coviella et al. (2000) grew cotton plants in the same manner as Coviella and Trumble, i.e., in twelve Teflon-film chambers in a temperature-controlled greenhouse, where six of the chambers were maintained at an atmospheric CO2 concentration of 370 ppm and six were maintained at 900 ppm.  Half of the plants in each chamber were of a transgenic line containing the Bt gene for the production of the Cry1Ac toxin that is mildly toxic to Spodoptera exigua, while the other half were of a near isogenic line without the Bt gene.  In addition, half of the plants in each chamber received high levels of nitrogen (N) fertilization, while half received low levels (30 as opposed to 130 mg N/kg soil/week).  Between 40 and 45 days after emergence, leaves were removed from the plants and fed to the S. exigua larvae, after which a number of larval responses were measured and analyzed, along with various leaf properties.

At the end of the experiment, it was learned that the low-N plants in the elevated CO2 treatment had lower foliar N concentrations than the low-N plants in the ambient CO2 treatment; and the transgenic plants from the low-N, high CO2 treatment produced lower levels of Bt toxin than did the transgenic plants from the low-N, ambient CO2 treatment, while the high level of N fertilization only partially compensated for this latter high-CO2 effect.  In the ambient CO2 treatment there was also a significant increase in days to pupation for insects fed transgenic plants; but this difference was not evident in elevated CO2.  In addition, pupal weight in ambient CO2 was significantly higher in nontransgenic plants; and, again, this difference was not observed in elevated CO2.

As for the implications of these results, Coviella et al. say they "support the hypothesis that the lower N content per unit of plant tissue caused by the elevated CO2 will result in lower toxin production by transgenic plants when nitrogen supply to the plants is a limiting factor."  They also note that "elevated CO2 appears to eliminate differences between transgenic and nontransgenic plants for some key insect developmental/fitness variables including length of the larval stage and pupal weight."  These results, in turn, suggest that in the case of inadvertent Bt gene transference to wild relatives of transgenic crop lines, elevated levels of atmospheric CO2 will tend to negate certain of the negative effects the wayward genes might otherwise inflict on the natural world.  Hence, the ongoing rise in the air's CO2 content could be considered to constitute an "insurance policy" against this potential eventuality.  Indeed, it could even be considered a natural manifestation of the Precautionary Principle, which could give some comfort to those who are advocates of genetic engineering within the context here discussed.

On the other hand, Coviella et al.'s results suggest that transgenic crops designed to produce Bt-type toxins may become less effective in carrying out the objective of their design as the air's CO2 content continues to rise.  Coupling this observation with the fact that the foliar application of Bacillus thuringiensis to crops should become even more effective in a higher-CO2 world of the future, as per the findings of Coviella and Trumble, one could argue that the implantation of toxin-producing genes in crops is not the way to proceed in the face of the ongoing rise in the air's CO2 content, which reduces the effectiveness of the genetic implantation technique at the same time that it increases the effectiveness of foliar application.  Hence, it is important to see what other researchers have learned about the subject.

Makino et al. (2000) grew three types of rice from seed - a wild type (WT) and two transgenic varieties, one with 65% wild-type Rubisco (AS-77) and one with 40% wild-type Rubisco (AS-71) - for a period of 70 days in controlled-environment chambers maintained at 360 and 1000 ppm CO2, after which they harvested the plants and determined their biomass.  This work revealed that the mean dry weights of the WT, AS-77 and AS-71 varieties grown in air of 360 ppm were, respectively, 5.75, 3.02 and 0.83 g.  In air of 1000 ppm CO2, on the other hand, the corresponding mean dry weights were 7.90, 7.40 and 5.65 g, so that the CO2-induced percentage increases in plant dry weight for the three varieties were 37% (WT), 145% (AS-77) and 581% (AS-71).  As a result, although the growth rates of the genetically-engineered rice plants were far inferior to that of the wild type when grown in ambient air of 360 ppm CO2, when grown in air of 1000 ppm CO2 they experienced far greater CO2-induced increases in growth.  Hence, whereas the transgenic plants were highly disadvantaged in air of 360 ppm CO2, they were found to be on a much more equal footing in highly-CO2-enriched air, which finding bodes well for the application of this type of genetic engineering in a future world of higher atmospheric CO2 concentration.

Chen et al. (2005b) grew well watered and fertilized cotton plants of two varieties (GK-12, expressing Cry1A (c) genes from Bacillus thurigiensis, and Simian-3, a non-transgenic cultivar from the same recurrent parent) in pots in open-top chambers maintained at either 376 or 754 ppm CO2 in Sanhe County, Hebei Province, China, from planting in mid-May to harvest in October.  Throughout this period, several immature bolls were collected and analyzed for chemical characteristics.  Others were stored under refrigerated conditions for later feeding to larvae of the cotton bollworm, when various parameters related to bollworm growth and development were monitored.  This work revealed that the elevated CO2 treatment increased immature boll concentrations of condensed tannins by approximately 22% and 26% in transgenic and non-transgenetic cotton, respectively (see Tannins in our Subject Index for a discussion of the significance of this observation).  In addition, Chen et al. report that elevated CO2 slightly decreased the body biomass of the cotton bollworm and reduced moth fecundity.  The Bt treatment was even more effective in this regard; and in the combined Bt-high-CO2 treatment the negative cotton bollworm responses were expressed most strongly of all.

In a parallel study, Chen et al. (2005a) grew transgenic Bacillus thuringiensis cotton (GK-12) plants from seed for a period of 30 days in well watered and fertilized sand/vermiculite mixtures in pots located in controlled-environment chambers maintained at atmospheric CO2 concentrations of 370, 700 and 1050 ppm.  Three generations of cotton aphids (Aphis gossypii) were subsequently allowed to feed on some of the plants, while a subset of the aphid-infected plants was additionally supplied with predatory ladybugs.  In response to these several operations, Chen et al. report that (1) "plant height, biomass, leaf area, and carbon:nitrogen ratios were significantly higher in plants exposed to elevated CO2 levels," (2) "more dry matter and fat content and less soluble protein were found in A. gossypii in elevated CO2," (3) "cotton aphid fecundity significantly increased ... through successive generations reared on plants grown under elevated CO2," (4) "significantly higher mean relative growth rates were observed in lady beetle larvae under elevated CO2," and (5) "the larval and pupal durations of the lady beetle were significantly shorter and [their] consumption rates increased when fed A. gossypii from elevated CO2 treatments."  Hence, they concluded that their study "provides the first empirical evidence that changes in prey quality mediated by elevated CO2 can alter the prey preference of their natural enemies," and in this particular case, they found that this phenomenon could "enhance the biological control of aphids by lady beetle," while at the same time enhancing control by means of negative Bt-induced effects on the aphids.

In conclusion, it would appear that genetic engineering to enable the production of Bt toxins within crops will either be little affected or actually somewhat benefited by the ongoing rise in the air's CO2 concentration.  Nevertheless, it would be wise to conduct many more studies of the subject.

References
Chen, F., Ge, F., and Parajulee, M.N.  2005a.  Impact of elevated CO2 on tri-trophic interaction of Gossypium hirsutum, Aphis gossypii, and Leis axyridis.  Environmental Entomology 34: 37-46.

Chen, F., Wu, G., Ge, F., Parajulee, M.N. and Shrestha, R.B.  2005b.  Effects of elevated CO2 and transgenic Bt cotton on plant chemistry, performance, and feeding of an insect herbivore, the cotton bollworm.  Entomologia Experimentalis et Applicata 115: 341-350.

Coviella, C.E., Morgan, D.J.W. and Trumble, J.T.  2000.  Interactions of elevated CO2 and nitrogen fertilization: Effects on production of Bacillus thuringiensis toxins in transgenic plants.  Environmental Entomology 29: 781-787.

Coviella, C.E. and Trumble, J.T.  2000.  Effect of elevated atmospheric carbon dioxide on the use of foliar application of Bacillus thuringiensis.  BioControl 45: 325-336.

Makino, A., Harada, M., Kaneko, K., Mae, T., Shimada, T. and Yamamoto, N.  2000.  Whole-plant growth and N allocation in transgenic rice plants with decreased content of ribulose-1,5-bisphosphate carboxylase under different CO2 partial pressures.  Australian Journal of Plant Physiology 27: 1-12.