Inheritance of Resistance to Bacillus thuringiensis Toxin (Dipel ES) in the European Corn Borer

See allHide authors and affiliations

Science  07 May 1999:
Vol. 284, Issue 5416, pp. 965-967
DOI: 10.1126/science.284.5416.965


Resistance in the European corn borer, Ostrinia nubilalis (Hübner), to a commercial formulation ofBacillus thuringiensis (Bt) Berliner toxin, Dipel ES, appears to be inherited as an incompletely dominant autosomal gene. This contrasts with the inheritance of resistance to Bt in other insects, where it has usually been characterized as a recessive trait. The proposed high-dose/refuge strategy for resistance management in Bt maize depends on resistance being recessive or partially recessive. If field resistance turns out to be similar to this laboratory resistance, the usefulness of the high-dose/refuge strategy for resistance management in Bt maize may be diminished.

Maize and several other crops have been bioengineered to express, in plant tissues, endotoxins derived from Bacillus thuringiensis (Bt) Berliner (1, 2). Such transgenic maize hybrids are known asBt maize (Bt corn) hybrids. Bt maize hybrids have been developed to protect the crop against corn borers such as the European corn borer, Ostrinia nubilalis(Hübner) (order Lepidoptera, family Crambidae). Ostrinia nubilalis ranks among the most important pests of maize in North America, causing losses in excess of $1 billion annually (1). The efficacy of Bt crops against this insect has been impressive and is resulting in widespread adoption of this bioengineered technology. Selection for pest resistance toBt is expected to be intense and is likely to result in the evolution of resistance to Bt endotoxins. An effective resistance management program will be needed to preserve the long-term utility of this technology. The U. S. Environmental Protection Agency (EPA) has approved conditional registrations for severalBt maize transformations and is requiring the development of a scientifically sound resistance management strategy by the year 2001 (1, 2). The currently favored resistance management strategy for Bt maize is the “high-dose/refuge strategy” (1). Implicit in this strategy is the assumption that genes promoting resistance in the insect will be recessive or partially recessive (1, 2).

We analyzed resistance to Dipel ES in a laboratory colony of O. nubilalis (3). Dipel ES is a commercial formulation ofBt endotoxins (4). Our results suggest that resistance to Dipel ES in O. nubilalis is inherited as an incompletely dominant autosomal gene. In contrast, the inheritance of resistance to Bt in most other insects is controlled by recessive genes (5, 6). If field resistance toBt turns out to be similar to this laboratory resistance, the usefulness of the high-dose/refuge strategy will be greatly diminished, and its application in the proposed resistance management plans required by EPA for the re-registration of Bt maize hybrids will need to be reevaluated.

Two O. nubilalis strains were used in this analysis: an unselected control strain (KS-SC-S) and a Dipel ES–resistant strain (KS-SC-R) (3, 7). The resistant strain has demonstrated 70-fold resistance to Dipel ES (3). These tests began when the strains were in the eighth and ninth generations in culture (seventh selected generation). Pupae were divided by sex before eclosion. Females from one population were mass-crossed with males from the other population. Four types of crosses were made: (i) reciprocal parental crosses between resistant (R) and susceptible (S) moths, (ii) F1 × F1 crosses, (iii) backcrosses of F1 with susceptible (S) moths (8), and (iv) three successive backcrosses between heterozygous (RS) and susceptible (S) moths (9). The susceptibility of corn borer neonates to Dipel ES was determined with a bioassay (10). The dose/mortality data were analyzed with probit regression.

Maternal effects on Dipel ES resistance were examined by comparing median lethal concentrations (LC50's) of progeny derived from the reciprocal parental crosses. The LC50's for the two reciprocal crosses did not differ significantly, based on overlapping 95% confidence intervals (CIs) (Table 1). This result confirms that Dipel ES resistance is inherited autosomally and that it is not sex-linked (11).

Table 1

Dose/mortality results for the progeny of susceptible crosses (S), resistant crosses (R), reciprocal crosses R × S (F1), F1 crosses (F2), and F1 × susceptible backcrosses of O. nubilalis in bioassays with Dipel ES incorporated in the diet. The RS and SR are heterozygous strains of O. nubilalis. The first letters in crosses represent the female. Numbers in parentheses indicate the 95% CI.

View this table:

Examination of Dipel ES resistance inheritance indicates that the trait is incompletely dominant. The pooled LC50 for the F1 crosses was significantly higher than that for the susceptible strain and significantly lower than that for the resistant strain, based on a nonoverlap of 95% CIs (Table 1). The dose/mortality regression line for the F1 was closer to that of the resistant strain than to that of the susceptible strain (Fig. 1). Stone's method (12) indicates a dominance value D of 0.721 ± 0.02, consistent with incomplete dominance (12).

Figure 1

Dose/morality regression lines for the progeny of susceptible strain (open squares, straight solid line), resistant strain (open circles, long-dashed line), reciprocal crosses R × S (F1) (solid triangles, dotted line), F1 crosses (F2) (solid circles, angled solid line), and F1× susceptible backcrosses (solid squares, angled dashed line) ofO. nubilalis in bioassays with Dipel ES incorporated in the diet.

We tested the standard monogenic inheritance model by comparing observed and expected mortality at different Dipel ES concentrations (6, 13). There was no significant deviation between observed and expected mortality for six of seven concentrations for the pooled backcross data and for five of seven concentrations in the pooled F2 generation. The minimum number of effective genes calculated with a modification of Lande's method (14) yielded a calculated number of genesn E of 0.57. Both results suggest that one gene (or a few genes) influenced Dipel ES resistance in this European corn borer strain.

Repeated backcrosses between heterozygous (RS) and susceptible (S) populations (9) showed no significant increase in mortality over the three successive backcross generations, and the mortalities for the three backcrosses were not significantly different from expected values based on criteria for monofactorial inheritance (Table 2). This result also suggests the presence of a single gene governing Dipel ES resistance in this strain of O. nubilalis (13).

Table 2

Dose/mortality results for the progeny of three successive backcrosses of the heterozygote to the susceptible strain ofO. nubilalis in bioassays with Dipel ES incorporated in the diet. Expected mortality for monogenic inheritance at 2.43 ml of Dipel ES per kilogram of diet = 59.5%. Numbers in parentheses indicate percentage mortality.

View this table:

There was a somewhat higher genetic variation for the backcross and F2 populations than for the resistant parent, susceptible parent, or F1 progeny, based on the slopes of the probit regression lines (Table 1 and Fig. 1). This increased variance suggests that the number of loci that affected Dipel ES resistance in the strain is small (6, 15).

The inheritance of resistance to Bt has been assessed in several other insects. In Plodia interpunctella(Hübner), resistance was autosomal and recessive or partially recessive (16). In Heliothis virescens(Fabricius), resistance to a delta endotoxin of Bt subsp.kurstaki strain HD-1 was autosomal, incompletely dominant, but controlled by several genetic factors (17). The high level of resistance in H. virescens to CryIAc and to CryIAb was partially recessive and controlled by one or a few loci, but the low level of resistance to CryIIA was more dominant (18). In four colonies of Plutella xylostella(L.) (6, 19), inheritance of resistance to Btsubsp. kurstaki was recessive or incompletely recessive. Resistance was most likely autosomal and controlled by one or a few major loci. The genetic bases for resistance to Bttoxins appear to be different in different insects (20) and may even differ for different toxins. Our results agree with other studies in that Dipel ES resistance in O. nubilalis appears to be caused by a single autosomal gene (or by a few genes). However, our results differ in that resistance in this strain appears to be incompletely dominant rather than recessive.

These results could have important implications for resistance management of O. nubilalis in Bt maize. The currently proposed resistance management strategy for Btmaize, the high-dose/refuge strategy (1, 2, 21), requires (i) that plant tissue be very toxic so that heterozygotes for resistance are killed, (ii) that the resistance alleles be very rare, and (iii) that susceptible insects are within an effective mating distance of resistant insects. The high-dose/refuge strategy would not be useful for resistance management if the trait is dominant. However, the genetic dominance we observed is expressed over a specific range of Dipel ES doses; at higher dosages, all individuals are expected to be susceptible. The practical importance of this genetic dominance will depend on whether these insects can survive on Bt maize. The high-dose/refuge strategy is subject to a number of stringent prerequisites that may be difficult to meet in practice. More robust resistance management options are needed.

  • * To whom correspondence should be addressed. E-mail: rhiggins{at}


View Abstract

Navigate This Article