	

July 20, 2006

MEMORANDUM

SUBJECT:		Review of 2004 pink bollworm monitoring data and revised
monitoring protocol submitted by Monsanto for Bt cotton (Bollgard and
Bollgard II).  EPA Reg. No. 524-478, 524-522.  DP Barcode:  No barcode
assigned.  Decision:  363972.   MRID#:  467350-01.

TO:			Leonard Cole, Regulatory Action Leader

Microbial Pesticides Branch

Biopesticides and Pollution Prevention Division (7511C)

FROM:		Alan Reynolds, Entomologist   /s/

Microbial Pesticides Branch

Biopesticides and Pollution Prevention Division (7511C)

PEER

REVIEW:		Sharlene Matten, Ph.D., Biologist    /s/

Microbial Pesticides Branch

Biopesticides and Pollution Prevention Division (7511C)

Action Requested

BPPD has been asked to review pink bollworm resistance monitoring data
submitted by Monsanto for the Bollgard (Cry1Ac) and Bollgard II
(Cry2Ab2) Bt corn registrations (EPA Reg No. 524-478 and 524-522).  The
submitted report (MRID# 467350-01) includes the monitoring results
collected from the 2004 growing season for the Cry1Ac and Cry2Ab2
toxins. 

Background

The major target pests of Bollgard Bt cotton, tobacco budworm (TBW),
cotton bollworm (CBW), and pink bollworm (PBW) have been monitored for
susceptibility to Cry1Ac since the product was first registered in 1996.
 Monitoring for PBW susceptibility to Cry1Ac has been conducted in the
state of Arizona from 1996 by the University of Arizona/Extension
Arthropod Resistance Management Laboratory.   

All of the Cry1Ac monitoring data through the 2000 growing season have
been previously reviewed by the Agency during the 2001 Bt crops
reassessment (EPA 2001).  In this review, it was concluded that through
the 2000 season, there was no evidence of TBW, CBW, or PBW resistance to
the Cry1Ac delta endotoxin produced by Bollgard cotton cultivars under
field situations.  Since the 2001 reassessment was completed, BPPD
monitoring reviews have been completed for Cry1Ac data through the 2003
growing season (see BPPD 2004a, 2005a).  These reviews also concluded
that no field resistance has been detected for the three major target
pests.  

Bollgard II, which expresses Cry2Ab2 and Cry1Ac, was registered for the
2003 growing season with similar monitoring requirements for Cry2Ab2. 
Data for PBW and Cry2Ab2 were initially received with the 2003 report
and have been previously reviewed (see BPPD 2005a).  The 2003 data
showed that PBW were highly susceptible to the toxin at the two test
doses.  The higher test dose used, 10 µg/mL, appeared to be functional
as a “discriminating dose” for PBW.

For the 2004 growing season, Cry1Ac and Cry2Ab2 monitoring data were
submitted for TBW and CBW (reviewed in BPPD 2005b).  However, PBW data
were not submitted with TBW/CBW report because PBW are typically
collected late in the season and must be reared for several months to
generate progeny for the bioassays.  As such, a supplemental report was
submitted for PBW and is reviewed in this memorandum.  

Conclusions and Recommendations

1) Monsanto has submitted an acceptable report for the 2004 Cry1Ac and
Cry2Ab2 monitoring of PBW.  No further information is needed for the
2004 PBW monitoring efforts.  However, BPPD has a number of comments,
detailed below, that should be addressed in future monitoring assays and
study reports.

2)  The 2004 data showed that PBW remains highly susceptible to both
Cry1Ac and Cry2Ab2.  At the discriminating test concentration used (10
µg/ml), only one larva survived on Cry1Ac and no larvae survived on
Cry2Ab2.  At the lower test concentration (1 µg/ml), mortality to both
toxins exceeded 95%.  

3) A separate boll sampling program revealed low PBW infestation among
collected Bt bolls.  Of the sampled Bt bolls, 0.34% were infested with
PBW, although many of the putative Bt bolls were later determined to be
non-expressing off types.  The rate of boll infestation observed in 2004
is comparable to rates seen in previous years.

4)  It is recommended that in future monitoring efforts, follow-up
susceptibility testing be conducted with PBW larvae recovered from Bt
bolls (verified expressing the Bt toxin), if historical infestation
averages are exceeded for Bt cotton.  Also, follow-up testing should be
conducted on survivors of the discriminating test concentrations.  It is
noted that such PBW larvae may contain resistance genes (either
heterozygous or homozygous) that should be further analyzed.  For 2004,
such testing was not needed because:  a) the rate of boll infestation in
Bt cotton was within historical averages, and b) only one PBW larvae
survived the discriminating test concentrations.

Monitoring Results - PBW (2004)

The 2004 monitoring work for PBW was conducted in Arizona by researchers
at the University of Arizona and the Arizona Cotton Research and
Protection Council, who have been conducting the work since 1997.  The
methodology for the 2004 PBW assays was largely the same as in previous
years and utilized artificial diet tests with a 21-day observation
period.  The monitoring also included the continuation of Cry2Ab2
testing initiated in 2003 (applicable to the Bollgard II registration). 
To conduct the bioassays, a discriminating dose type approach in which
PBW mortality was assessed to two test concentrations for both Cry1Ac
and Cry2Ab2.  Baseline susceptibility (i.e. a LC50 or similar measure)
was not determined.  The two test concentrations of Cry1Ac and Cry2Ab2
used were 1.0 µg/ml and 10 µg/ml.  Negative controls (no toxin) were
also tested.  The Cry1Ac toxin used in the assays was obtained from Dow
AgroSciences (MVP-II Bioinsecticide) while the Cry2Ab2 toxin was
obtained from freeze dried corn powder provided by Monsanto.  These
toxin sources were also used for the 2003 monitoring.   In addition to
the laboratory bioassays, field efficacy was assessed in 2004.

	2004 Sampling and Assays

PBW were collected as larvae (from bolls brought to the laboratory) from
three western cotton-growing states:  Arizona (15 sites), California (2
sites), and New Mexico (1 site).  Sampling was limited in New Mexico and
Texas (no collections) due to ongoing PBW eradication efforts in those
areas.  All collection sites were represented in the Cry2Ab2 testing,
though two sample sites from Arizona were not used in the Cry1Ac
testing.  A susceptible laboratory strain was also used as an internal
standard for the experiments.  Additionally, for the Cry2Ab2 tests, a
Cry1Ac-resistant PBW laboratory colony was included.  Fourth instar
larvae emerging from bolls were reared to adulthood to produce progeny
for testing (F2 - F8 progeny were used in the tests).  The bioassays
were conducted with artificial diet incorporated with the two test
concentrations (an untreated control was also used).  Neonate larvae
were placed in one ounce cups with diet and observed for 21 days. 
Larvae that failed to develop past the third instar by the end of the
test were considered “dead” and Abbott’s formula was used to
obtain corrected mortality scores (i.e. to justify mortality in the
control groups).

2004 Cry1Ac Results

The results of the Cry1Ac assays from Arizona-collected PBW at the 1.0
µg Cry1Ac/ml dose were variable and revealed an average corrected
mortality of 96.5% (range 78.7 - 100%).   At the 10 µg Cry1Ac/ml dose,
the average mortality for the Arizona collections was 99.9% (range 99.7
- 100%), with only one population exhibiting less than 100% mortality to
the test concentration (only one larva survived to the fourth instar). 
Similar results were observed for the California and New Mexico
collections, with a mortality range at the 1.0 µg Cry1Ac/ml dose of
88.2 - 100% and 100% mortality for all three tested populations at the
10.0 µg Cry1Ac/ml concentration.  The susceptible laboratory colony
used as a control group showed 58.5% mortality at the 1.0 µg Cry1Ac/ml
concentration and 100% mortality to the 10 µg Cry1Ac/ml   In comparison
to previous years, the 2004 results at the higher 10.0 µg/ml dose
(99.9% overall mortality) were similar to the results from 2002 and 2003
(99.8% average mortality was observed in both years).  However, the
overall PBW mortality observed in 2004 at the 1.0 µg Cry1Ac/ml
concentration (95.4) was higher than in previous years (68.3% in 2003,
85.7% in 2002) (also see table 1 below).  In the report, the authors
pointed out that PBW mortality to the higher discriminating
concentration (10 µg Cry1Ac/ml) has been higher in recent years
(>99.4%) than was observed in the first year of Cry1Ac monitoring (94.1%
mortality in 1997).   Overall, the authors concluded that PBW remains
susceptible to Cry1Ac and that are no indications of resistance in the
field.  The 2004 results from Arizona are summarized and compared with
historical data in table 1 below.

2004 Cry2Ab2 Results

	

As with Cry1Ac, PBW were sensitive to Cry2Ab2 during the 2004 testing. 
At the 1.0 µg/ml test concentration, mortality (corrected) for Cry2Ab2
ranged from 97.9 to 100%  for Arizona collections (average 99.8%) and
95.2 to 100% for California and New Mexico collections (average 97.2%). 
At the 10.0 µg Cry2Ab2/ml dose, no survivors were observed from any of
the Arizona, California, or New Mexico samples (100% overall mortality).
 The susceptible laboratory colony used as control had average
mortalities of 82.7% to the 1.0 µg Cry2Ab2/ml concentration and 100% to
the 10.0 µg Cry2Ab2/ml concentration.  The Cry1Ac-resistant colony
showed somewhat less mortality to the lower 1.0 µg concentration
(70.7%), but like the other groups, had 100% mortality to the higher
10.0 µg concentration.  The 2004 assays represented the second full
year of monitoring for resistance to the Cry2Ab2 toxin and had similar
results to the 2003 monitoring (see table 2) and also to previous
baseline work with Cry2Ab2 done in 2001 and 2002 (see BPPD 2004b).  As
with Cry1Ac, the authors concluded that the sampled PBW populations
remained highly susceptible to Cry2Ab2.  The 2004 Arizona data are
summarized in table 2 below.

2004 Field Efficacy Studies

In addition to the susceptibility bioassays, the Arizona monitoring
group sampled large numbers of Bt and non-Bt cotton bolls throughout the
state (obtained from 40 pairs of Bt and non-Bt fields).  The procedures
were similar to the boll sampling that was also conducted during 2003. 
Bolls were examined for PBW larvae – of 39,500 Bt boll sampled, 133
PBW larvae were found (for an infestation rate of 0.34%).  By
comparison, the infestation rate for 2003 was 0.21%.   For non-Bt
cotton, 2082 larvae were found in 10,375 non-Bt bolls, an infestation
rate of 21.7% (compared with 27.4% in 2003).  As was the case in 2003,
subsequent analysis of the Bt bolls determined that many were
non-expressing off-types (33 out of 35 infested Bt bolls were found to
be off-types).  The authors noted that the infestation rate observed in
Bt fields has historically averaged less than 0.35%.

As with past reports, the PBW reports for 2004 was thorough and
well-organized.  Since the methodology has remained consistent
throughout the PBW monitoring efforts, the data can be placed in a
historical context to evaluate long-term shifts in susceptibility. 
Through 2004, eight years of monitoring data have now been tabulated for
Cry1Ac and two years for Cry2Ab2.  Overall, the authors note that PBW
susceptibility to both toxins remains high.  BPPD agrees with this
assessment, based on the susceptibility data compiled to date (see
tables 1 and 2 below) and the low infestation rates observed in the
field efficacy trials.  PBW mortality at the 10.0 µg/ml dose was
comparable to previous years (close to 100% mortality).  The 10.0 µg/ml
concentration is essentially a true discriminating dose, i.e. an LC99
that can be used to distinguish potentially resistant insects from
susceptible ones.  Infestation rates in the field efficacy study were
0.34% infested Bt bolls, which is higher than has been observed in other
years.  However, many of the collected Bt bolls with PBW larvae were
later determined to be non-Bt expressing off-types and the increased
infestation rate is unlikely to be the result of adaptation to Bt
toxins.  PBW susceptibility to Cry1Ac at the lower 1.0 µg/ml
concentration increased in 2004 after two years in which lower
mortalities were observed (see table 1).  The researchers had attributed
the reduced susceptibility in 2002 and 2003 to a potential decrease in
the potency of the Cry1Ac toxin used in the tests (similar reductions in
susceptibility were noted with the unselected laboratory colony).  It is
unclear from the report if the researchers obtained new, more potent
toxin or addressed the issue through other means, although the 2004
results were comparable to the data obtained prior to 2002.

The 2004 PBW report contained the second year of monitoring with the
Cry2Ab2 toxin (expressed in Bollgard II cotton).  PBW were found to be
highly susceptible to the toxin at both of the chosen concentrations:
1.0 µg/ml and 10.0 µg/ml.  The 10.0 µg/ml dose appears to be a
functional discriminating dose, with 100% mortality to all sampled PBW
populations.

It is recommended that in future field monitoring efforts, follow-up
susceptibility testing be conducted with PBW larvae recovered from Bt
bolls (verified expressing the Bt toxin) if historical infestation
averages are exceeded for Bollgard.  Past reports have noted that 0.300%
infested bolls (3 per 1000) is the average PBW infestation rate for Bt
cotton.  To date, PBW sampling has not revealed infestations in Bt
cotton bolls exceeding this historical rate after non-expressing off
types are discounted.  Since Bollgard and Bollgard II are considered
high dose for PBW, larvae recovered from Bt bolls may be heterozygous or
homozygous for Bt toxin resistance.  The determination that these larvae
are carrying heritable resistance traits could provide an early
indication of a resistance problem.  It is also noted that if widespread
infestation is observed in Bt fields (i.e. above historical averages),
additional action may be necessary, as prescribed by remedial action
plans for the registrations.  It is also recommended that follow-up
testing be conducted on survivors of the 10 µg/ml Cry1Ac and Cry2Ab2
discriminating concentrations.  These larvae may also be homozygous for
Cry1Ac or Cry2Ab2 resistance alleles and warrant additional scrutiny. 
On the other hand, survivors of the 1 µg Cry1Ab/ml dose may be
heterozygotes with one resistance allele.  For the 2004 data, only one
larva was observed to survive the discriminating 10 µg/ml Cry1Ac
concentration (no larvae survived the Cry2Ab2 discriminating
concentration).  As such, no follow-up testing was needed for the 2004
monitoring work.  However, given that some PBW have survived both test
concentrations in in previous years, it is possible that resistance
alleles are relatively common in PBW populations in western cotton
growing regions.  Considering the high Bt cotton adoption in these
regions, it will be imperative to closely monitor PBW and Bt cotton for
increases in tolerance to Cry1Ac and possible unexpected field damage.

Table 1. Field-Collected PBW Mortality to Discriminating Concentrations
of Cry1Ac from 1997 to 2004 (taken from Dennehy et al. 2002 - 2005) 

Year	Average Mortality of Field Collected PBW (%)1

	1.0 µg Cry1Ac/ml dose 	10 µg Cry1Ac/ml dose

1997	57.4	94.1

1998	90.6	99.9

1999	97.9	100

2000	97.4	100

2001	94.8	99.4

2002	85.7	99.8

2003	68.3	99.8

2004	95.4	99.9

1 Mortality values are corrected for mortality observed in control
groups.

Table 2. Field-Collected PBW Mortality to Discriminating Concentrations
of Cry2Ab2 from 2003 to 2004 (taken from Dennehy et al. 2004, 2005) 

	Year		Average Mortality of Field Collected PBW (%)1

		1.0 µg Cry2Ab2/ml dose 		10 µg Cry2Ab2/ml dose

2003		97.3		99.9

2004	99.1	100

1 Mortality values are corrected for mortality observed in control
groups.

References

Pink Bollworm Monitoring

  SEQ CHAPTER \h \r 1 Dennehy, T., L. Shriver, M. Sims, D. Holley, Y.
Carriere, and B. Tabashnik, 2002.  Monitoring for Cry1Ac protein
susceptibility among field populations of pink bollworm during the 2001
cotton growing season.   Unpublished study submitted to EPA.  MRID # 
457065-01.

Dennehy, T., S. Brink, B. Wood, D. Holley, G. Unnithan, Y. Carriere, and
B. Tabashnik, 2003.  Susceptibility of pink bollworm to Cry1Ac:  final
results of 2002 season studies.   The University of Arizona Cooperative
Extension paper.  No MRID #.

Dennehy, T., G. Unnithan, S. Brink, B. Wood, Y. Carriere, B. Tabashnik,
L. Antilla, and M. Whitlow, 2004.  Susceptibility of pink bollworm to
Cry1Ac and Cry2Ab2:  final results of 2003 season studies.   Unpublished
study submitted to EPA.  MRID # 464419-01.

Dennehy, T., G. Unnithan, S. Brink, B. Wood, Y. Carriere, and B.
Tabashnik, 2005.  Susceptibility of Bt toxins Cry1Ac and Cry2Ab2 to
southwestern pink bollworm in 2004.   Unpublished study submitted to
EPA.  MRID #  .

Other

EPA, 2001.  Bt plant-incorporated protectants October 15, 2001
biopesticides registration action document.  Available at
http://www.epa.gov/pesticides/biopesticides.

BPPD, 2003.  Technical review of Monsanto’s submissions: resistance
monitoring protocols for monitoring of Cry1Ac susceptibility among field
populations of pink bollworm, tobacco budworm, and cotton bollworm.  S.
Matten memorandum to L. Cole, September, 2, 2003.

BPPD, 2004a.  Review of 2001-2 monitoring data submitted by Monsanto for
Bt cotton (Bollgard).  A. Reynolds memorandum to L. Cole, September 8,
2004.

BPPD, 2004b.  Review of preliminary monitoring data submitted by
Monsanto for Bollgard II Bt cotton.  A. Reynolds memorandum to L. Cole,
May 26, 2004.

BPPD, 2005a.  Review of 2003 monitoring data submitted by Monsanto for
Bt cotton (Bollgard and Bollgard II).  A. Reynolds memorandum to L.
Cole, June 8, 2005.

BPPD, 2005b.  Review of 2004 tobacco budworm/cotton bollworm monitoring
data and revised monitoring protocol submitted by Monsanto for Bt cotton
(Bollgard and Bollgard II).  A. Reynolds memorandum to L. Cole, December
1, 2005.

 PAGE   - 6 - 

 PAGE   - 1 - 

 

