Aedes aegypti OX513A
Investigational Trial Protocol for the Evaluation of Efficacy
Evaluation of OX513A Aedes aegypti ability to transfer #OX513 rDNA construct by mating, resulting in the progeny having increased mortality.
Protocol July 2016 V12
Oxitec Ltd, 71 Innovation Drive, Milton Park, Oxfordshire, UK OX14 4RQ
Oxitec Ltd CONFIDENTIAL Page 1 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
1 STATEMENT OF CONFIDENTIALITY This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd.
Oxitec Ltd CONFIDENTIAL Page 2 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
2 LIST OF ABBREVIATIONS/GLOSSARY OF TERMS IN THIS DOCUMENT
Abbreviation/Term Description ACL2 Arthropod Containment Level 2 CI Confidence interval DNA Deoxyribonucleic acid DsRed2 Fluorescent marker from Discosoma species EA Environmental Exp exponential FKMCD Florida Keys Mosquito Control District Functional Adulthood Adults that are fully eclosed and able to maintain flight are assumed to be fully functional in ability seek hosts for blood meal and mate in order to reproduce. Adults that die on eclosion or are fully eclosed but incapable of maintaining flight, are not considered to have reached functional adulthood. GLP Good Laboratory Practice GIS Geographical information System GPS Global Positioning Systems Ha Hectare Hr (s) Hour (s) HRU Hatch and Release Unit IRR Initial Release Rate Km Kilometre L1 First instar larvae L3 Third instar larvae LPS Larval Pupal Sorter Mating Fraction Proportion of local females that mate released OX513A relative to local males. Determined and monitored through counting the numbers of fluorescent larvae and non fluorescent larvae hatched from collected eggs from ovitraps. The proportion of fluorescent larvae relative to non fluorescent larvae is the ‘Mating fraction’. mL Millilitre NB Nota bene ( Note) PCR Polymerase Chain Reaction RH Relative Humidity RR Relative Risk rDNA Recombinant DNA QC Quality Control Site Area A designated area in which OX513A releases will take place comprising of both the TA and the UCA Sqrt Square root SOP Standard Operating Procedure TA Treatment Area
Oxitec Ltd CONFIDENTIAL Page 3 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd.
Investigational field protocol Effective Date 2...
efficacy claims relating to the use of OX513A Ae.aegypti for vector control purposes. Efficacy of the system will be demonstrated if the stated primary objectives are achieved.
8.1 Primary Objective part 1: To determine whether released OX513A mate with local females resulting in their progeny inheriting a copy of the #0X513 rDNA construct.
8.2 Primary Objective part 2: To determine whether progeny inheriting the #0X513 rDNA construct have the expected phenotype resulting in significantly increased mortality
.
8.3 Secondary Objective: To determine whether sustained release of OX513A males results in a statistically significant suppression (≥50% with 95% confidence interval (CI)) of the local populations of Ae.aegypti relative to an untreated comparator area.
The Primary Objective part 1 and 2 will be used to support the following proposed product claim; OX513A mates with females of wild Aedes aegypti in a population so that progeny carry a copy of the #0X513 rDNA construct and result in mortality of these #0X513 rDNA construct bearing progeny before they reach functional adulthood.
9 STANDARDS APPLIED TO THE CONDUCT OF THE STUDIES (GLP, INVESTIGATOR, OR OTHER) This field trial is conducted for the purpose of obtaining data on claim validation (i.e., animal drug efficacy) in the environment and will not conducted under GLP. Standards used will be suitable for the publication of the trial in the peer‐reviewed scientific literature of international standing.
10 STUDY SCHEDULE 10.1 Proposed date of initiation: Q3 2016, depending on FDA‐CVM review
10.2 Schedule of events: The schedule of study events is described in Table 1:
Oxitec Ltd CONFIDENTIAL Page 5 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
Table 1: Overview of Study Schedule Event Method Timeframe PREPARATION PHASE Technology transfer Import eggs OX513A eggs shipped by air from Pre‐trial shipments for staff Oxitec UK laboratory to local rearing training in rearing/identification facility Maximum of for duration of trial. Establish local mass rearing Optimize rearing methodology to 4‐6 weeks local conditions Implement local mosquito Ovitraps deployed Minimum 8 weeks prior to population monitoring in Treatment rangefinder Area (TA) Untreated Comparator Area (UCA ) Ovitraps mimic natural oviposition sites. Proportion of ovitraps traps with eggs reflects adult Ae.aegypti density in population. BG‐ Sentinel adult traps will sample from 20 locations/week in each of TA and UCA. BG‐Sentinel adult traps attract host seeking adults. Numbers captured reflect adult population density. RANGEFINDER PHASE Obtain site specific estimate of Releases will be initiated for a period mating fraction achieved with a of (up to releases/week) known constant release rate of at a constant rate. Release will be a OX513A males. function of human population and Field derived egg sampling for estimated Ae.aegypti infestation primary objectives parts 1 and 2. level in treatment area at start of Data collected for the initiation of releases (Table 4). the Secondary objective (suppression phase). Release of OX513A males OX513A males to be released from predetermined GPS grid of release points to ensure even coverage of TA. Releases up to times/week
Oxitec Ltd CONFIDENTIAL Page 6 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd.
Investigational field protocol Effective Date 2...
Captured adults can be identified directly, allowing direct species identification and species specific population monitoring where species other than Ae.aegypti (eg. Ae albopictus) are/or expected to be present. Population density monitoring data collection for Secondary objective. POST TRIAL MONITORING After the cessation of releases following ovitraps are used for post‐trial cessation of release monitoring in the treatment area (TA immediately) for . Additional monitoring will be carried out for with an interim trapping by ovitraps period if fluorescence confirmed at the period. with an interim trap at if fluorescence is confirmed
10.3 Proposed date(s) of completion: Release phase will be from Table 1, not including the preparation phase) from the initiation of releases but may stop earlier if the primary and secondary objectives have been met. Analysis of data will be conducted throughout the study making it possible to determine if the primary and secondary objectives have been met before . If they have been met they will be recorded and a decision to stop the trial will be made by Study Director and Sponsor, in conjunction with FKMCD.
10.4 Post‐trial monitoring: After the cessation of releases, monitor will occur in the treatment area (TA), to maximise the likelihood of detection, immediately post‐trial. The monitoring tool will be ovitraps as they are more sensitive than adult traps and will also provide evidence of potential mating. Ovitraps will be set at the same trap density used in the trial and traps processed as discussed in Section 11.3 of the protocol. following the immediate post‐trial period, there will be an additional monitoring period of , where ovitraps at the same trap density used in the trial will be set in the treated area for , collected and screened using visual screening of larvae for fluorescence. If the visual fluorescence is in doubt, then samples will be shipped back to the UK for PCR analysis. If fluorescence is confirmed during this period, then a further screening will be conducted later, using the same techniques. If there is no fluorescence confirmed at the ‐trapping period, then the time‐point survey will not be conducted. A further screening will be
Oxitec Ltd CONFIDENTIAL Page 8 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
conducted at following the immediate post‐trial period, regardless of whether fluorescence is confirmed at the trapping periods.
The periodic post‐trial monitoring will conclude at (past the initial ) as this will have covered a “mosquito season” which typically in the Florida Keys runs from May to November and therefore includes monitoring of the previous years’ egg bank.
We expect to detect fluorescent samples in the immediate period. The results of this post‐trial survey will be reported to FDA within of the last trap being collected and analysed for fluorescence. If during the course of the subsequent post‐trial monitoring at (conditional on fluorescence confirmation at ), and post continuous monitoring, fluorescence is confirmed from trap samples, Oxitec will notify FDA‐CVM within of results being confirmed. The OX513A line of Ae. aegypti mosquitoes carries a repressible dominant lethality trait that prevents progeny inheriting the #OX513 rDNA construct from surviving to functional adulthood in the absence of tetracycline. Therefore, OX513A mosquitoes are not expected to establish at the proposed trial site. FKMCD will continue to undertake routine mosquito control operations in the area as required, and OX513A insects are susceptible to the chemicals used in those control programs.
STUDY LOCATION AND DESIGN
The study will be conducted in Monroe County, Florida. The proposed release site is located within Monroe County, on Key Haven. All areas in the study fall within the Florida Keys Mosquito Control District (FKMCD) and may receive on‐going mosquito abatement measures independent of this study (please refer to 16.7 of this protocol)
10.5 A designated Site Area: (site for evaluation of OX513A) will be divided into two areas of similar size separated by a buffer zone as in Figure 1. The will form the Treatment Area (TA) and the Untreated Comparator Area (UCA) with the buffer area .
Oxitec Ltd CONFIDENTIAL Page 9 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. SP‐00030 ED 1.a OX‐USA‐1 Investigational field protocol Effective Date 2...
Figure 1: Site Area for Investigational Use of OX513A
In summary the two blocks in the experiment are:
Treatment Area (TA) ‐ receives sustained release of OX513A males with a buffer zone . Untreated Comparator Area (UCA) ‐ receives no released OX513A males
The rationale for the study design and methodology along with additional background information are provided in the Supplementary Information Document (Appendix A).
10.6 Study Design: The study is designed to provide claim validation (Section 7), and will be conducted in two blocks as described above.
Oxitec Ltd CONFIDENTIAL Page 10 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd.
Investigational field protocol Effective Date 2...
Local larvae (non‐fluorescent)
10.9 Secondary Objective: To determine whether sustained release of OX513A results in a statistically significant suppression of the local populations of Ae.aegypti relative to UCA. 10.9.1 Null Hypothesis: Sustained release of OX513A will not affect the Ae.aegypti population in TA or will result in suppression of the Ae.aegypti population in the TA by relative to UCA. 10.9.2 Endpoint: Suppression of the Ae.aegypti population in the TA by relative to untreated population in UCA where the mean of multiple trap collections (Ovitraps and BG‐Sentinel adult traps) before suppression is compared with mean from a period, calculated as rolling average, from the suppression phase. 10.9.3 Experimental outline: Ovitrap and BG‐Sentinel adult trap surveys will be used to monitor the changing population densities of Ae.aegypti in TA and UCA. The population density of Ae.aegypti will be estimated before initiation of releases. Subsequent changes in relative population density between TA and UCA will be used to assess impact of sustained OX513A male releases on local population in the TA.
10.10 Replication in the study design: Primary objective (parts 1 and 2) will be assessed at level of individual insects. Secondary objective is assessed at the level of Ae.aegypti population density within the TA relative to UCA. There is no replication in the study because there is one contiguous treated area.
11 STUDY PROCEDURES 11.1 General procedures: 11.1.1 UK Egg Production: Sufficient eggs of homozygous OX513A for the conduct of the trial will be imported from Oxitec Ltd, UK by air and ground transport. It is anticipated that a maximum of shipment/week will be required for the duration of the trial and some pre‐trial shipments will be required to facilitate initial rearing optimization and staff training. The eggs are produced in the UK under regulated conditions for the use of genetically engineered organisms. The OX513A strain is subject to quality control procedures ) to ensure it remains homozygous for the introduced genes and there is no loss of phenotype. 11.1.2 Local Mass Rearing: Eggs will be received in a dedicated facility designed to Arthropod Containment Level 2 (ACL2) standards1 in Marathon, FL. Eggs will be
1http://online.liebertpub.com/toc/vbz/3/2 [accessed 29 Jan 2014]
Oxitec Ltd CONFIDENTIAL Page 12 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
reared to functional adults for the study. Methods in the SOP’s may require optimization for specific local conditions during the preparation phase of the trial. Deviations from the procedures will be noted and addressed in SOP revisions. 11.1.3 Staff training: Staff will be trained on the SOP’s and mosquito species identification using taxonomic keys prior to the start of the trial in the preparation phase. 11.1.4 Mosquito release and dispersal: OX513A mosquitoes will be deployed in a systematic manner from a pre‐determined georeferenced grid of release points at regular time intervals, for even and consistent coverage of the TA. Release points will be spaced approximately apart, with a maximum spacing of , and releasesl wil occur up to 3 times per week. Release points will be georeferenced using Global Positioning Systems (GPS), and the TA mapped with spatial data incorporated into an appropriate Geographical Information System (GIS). 11.1.5 Numbers to be released: The release numbers will be proportional to the local population of Ae.aegypti and the initial rate will be determined from the Rangefinder phase (see section 12.2 for details) in order to achieve a target mating fraction . Target mating fraction is determined and monitored through counting the numbers of fluorescent larvae and non‐fluorescent larvae hatched from collected eggs from ovitraps. The proportion of fluorescent larvae to non–fluorescent larvae in the population is the ‘Mating Fraction’.
11.2 Adaptive management of release rates: Mating fraction allows release rates to be adapted to the local Ae.aegypti population. As the population declines the ratio of released OX513A to local males will increase, as will the mating fraction. Hence, a lower release rate can be used while maintaining mating fraction in the target range of ≥0.5. In the anticipated event of Ae.aegypti populations decreasing, release numbers may also be revised downward correspondingly (section 12.5). If there are rapid increases in the local Ae.aegypti populations associated with seasonality over the course of the study, the release rate may be sustained at the same rate or an increased to maintain the mating fraction≥ in larvae to achieve suppression of the local Ae.aegypti population. Monitoring procedures: Two trap types (Ovitraps and BG‐ Sentinel adult traps) will be used to monitor the Ae.aegypti population in TA and UCA. Ovitraps provide an indirect measure of female abundance allowing assessment of Ae.aegypti population density, without interference from released OX513A (see Supplementary Information, Appendix A for more details). Eggs will be collected, hatched and the resultant progeny analyzed as set out in this protocol (Sections 11.3.3 and 11.3.4) assuming the possible presence of mosquito species other than Ae.aegypti (e.g. Ae albopictus). BG‐Sentinel traps directly capture adults. Number of female Ae.aegypti will be assessed; reflecting local Ae.aegypti population density, without
Oxitec Ltd CONFIDENTIAL Page 13 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
interference from OX513A released males. Ovitraps and BG‐Sentinel adult traps will be located predominately by domestic dwellings, although other sites may also be included. Consent for the placing and servicing of the traps will be obtained from the property owner/occupant. As far as possible, ovitraps and BG‐Sentinel adult traps will remain in same location for the duration of the study. However, it may be necessary to redeploy some traps within the same area, for example if property owner consent is removed. In this event traps will receive a new number corresponding to its revised location.
Oxitec Ltd CONFIDENTIAL Page 14 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
11.3 Ovitrap: Ovitraps will monitor changes in relative abundance of local Ae.aegypti population between the TA and UCA and allow the collection of eggs for evaluation of mating between released OX513A males and local females. 11.3.1 Deployment: A minimum of ovitraps will be used in each area (TA and UCA) with a trap density of . All trapping locations will receive a unique number and be georeferenced using a GPS system. 11.3.2 Servicing: traps will be checked . Ovitraps will be serviced by water being topped up/replaced if needed. If necessary, the whole trap may be replaced. Each time traps are serviced oviposition substrate will be replaced with new one. Collected oviposition substrate will be labelled and transferred to the laboratory for further analysis on this schedule. 11.3.3 Ovitrap processing: Oviposition substrate will be examined for presence of Aedes eggs. As species and paternity identification is not possible at the egg stage, eggs will be matured by allowing them to dry at room temperature (range 20‐30°C) for before hatching. Hatch eggs by submerging oviposition substrate and eggs under water in individual pots identified by the ovitrap from which they were collected. 11.3.4 Larval screening for OX513A fluorescent marker: Ae.aegypti larvae (L1) will be visually screened for the presence of the fluorescent marker (Phuc et al. 20072) in eggs collected from ovitraps in TA and UCA during the study. Fluorescence is visible through a suitable fluorescent microscope (e.g. Leica MZ10F) with a DsRed2 filter set (e.g. Chroma) with excitation 520‐560 nm and emission 580+ nm) in a darkened room. Each larva will be scored as fluorescent or non‐ fluorescent. Fluorescent larvae will be assumed to be Ae.aegypti and non‐ fluorescent larvae subsequently identified as Ae.aegypti/Non Ae.aegypti from their morphological traits using taxonomic keys3. 11.3.5 Species identification of non‐fluorescent larvae: Larvae will be maintained at room temperature (range 20‐30°C) and will be provided with ground Tetramin® fish food ad libitum. Positive species identification of Ae.aegypti will be carried out at larval (L3 or above) and/ or adult stage from key morphological features, using taxonomic keys. 11.3.6 Calculation of Ovitrap index: Ovitrap index will be calculated as measure of abundance in TA and UCA:
2Phuc H, Andreasen M, Burton R, Vass C, Epton M, et al. (2007) Late‐acting dominant lethal genetic systems and mosquito control. BMC Biology 5: 11.
3Taxonomonic keys: http://fmel.ifas.ufl.edu/key/ [accessed 31 July 2014]
Oxitec Ltd CONFIDENTIAL Page 15 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd.
Investigational field protocol Effective Date 2...
11.5.6 Before pupation begins, examine pots daily. 11.5.7 If dead individuals are present, collect each individual separately into a 1.5mL microcentrifuge tube containing 0.5mLs 70% ethanol. Label each tube with the cohort description and date. Store at ≤‐15°C. Record the numbers of dead larvae and dead pupae. 11.5.8 Once pupation begins, examine pots daily: Pick all live pupae from each pot and transfer into an eclosion container such as a weigh boat (100mL) assigned to that pot. Each day live pupae should be added to this eclosion container. Place the eclosion container into the adult cage assigned to that pot. If dead individuals are present in the rearing container or the eclosion container, collect each individual separately into a microcentrifuge tube containing 0.5mLs 70% ethanol. Label each tube with the pot description and date. Store at ≤‐15°C. Record the numbers of dead larvae, dead pupae and live pupae. 11.5.9 Once eclosion begins, examine cages daily: Collect any dead adults, including those on the surface of the eclosion container and place individually into a microcentrifuge tubes containing 0.5mLs 70% ethanol. Label each tube with the cohort description and date. Store at ≤‐15°C 11.5.10 Terminate the experiment after pupation begins: From rearing pots collect remaining dead and live larvae/pupae and store individually in microcentrifuge tubes containing 70% ethanol at ≤‐15°C. Remove the eclosion container from the cage. Count and record the number of dead pupae and dead adults on the surface of the water. Store individually in microcentrifuge tubes containing 70% ethanol at ≤‐15°C. Count and record the number of dead adults on the bottom of the cage, and non‐functional adults. Store individually in microcentrifuge tubes containing 70% ethanol at ≤‐15°C. Freeze cage at ≤‐15°C until adults are dead. Count and collect remaining adults (functional adults). Store individually in microcentrifuge tubes containing 70% ethanol at ≤‐15°C.
The recording category definitions for mortality are in Table 3 below:
Oxitec Ltd CONFIDENTIAL Page 18 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
Table 3: Recording Categories for Mortality from Larval Rearing/Eclosion Dead larvae Larvae that die
Dead Pupae Pupae that die in rearing pot or eclosion
container, including those partially eclosed
Dead adults on Fully eclosed adults that have died on the water surface of the water in the eclosion container. Dead adults in Fully eclosed adults that have left eclosion cage container and subsequently died.
Live larvae at end Live larvae at end of study
of study
Live pupae at end Live pupae at end of study of study Non‐functional Fully eclosed live adults that are unable to adults maintain flight
Functional adults Fully eclosed live adults able to maintain flight
11.6 PCR Analysis of Mosquito Larvae and Adults: 11.6.1 Ship samples collected as part of the ‘Testing Pre‐Functional‐Adult Mortality’ to Oxitec Ltd. UK. NB: Samples will be shipped double contained in accordance with all governing regulations. 11.6.2 Carry out DNA extractions according to Oxitec Ltd SOP: New Genomic DNA extraction using Purelink Genomic DNA Kit by Invitrogen (Appendix E) 11.6.3 Carry out PCR analysis according to Oxitec Ltd SOP: OX513A Quality Control Protocol for Colony Genotyping (Appendix B) Record the genotype of each insect with respect to OX513A alleles. Results will confirm the presence or absence of #OX513 rDNA construct in the test groups and the zygosity of #OX513 rDNA construct in the samples. The majority of sample genotypes are expected to be either local or hemizygous (OX513A:local) resulting from the mating of released OX513A males with local females.
Oxitec Ltd CONFIDENTIAL Page 19 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd.
Investigational field protocol Effective Date 2...
subjective, based on the judgement of the Study Director. Ae.aegypti INITIAL INFESTATION LEVEL will be categorised as VERY LOW, LOW, MEDIUM, or HIGH (as described in Table 4).
12.2.2 Calculation of the Rangefinder Release rate: Rangefinder Release rate = Human population density * treatment area estimated infestation level as in Table 4:
Table 4: Guideline Minimum Initial Release Rate (OX513A males/person/week) according to Estimated Infestation Level.
Initial Ovitrap Minimum infestation Index Release Rate (IRR) per level of person in treatment area Ae.aegypti HIGH
MEDIUM
LOW VERY LOW
12.3 Suppression phase: the objective of the suppression phase of the trial is sustained release of OX513A resulting in a statistically significant suppression of the local populations of Ae.aegypti relative to untreated comparator area thereby gathering data for the Secondary objective of the study. Additionally, field collected eggs will be used to examine Primary Objective parts 1 and 2 if insufficient were collected during the range finder phase.
12.3.1 A mating fraction is estimated to be sufficient to suppress the target population of Ae.aegypti relative to untreated comparator area. This is based on models of mosquito population dynamics giving elimination thresholds as 0.13‐0.574 and previous experience from field release of OX513A in Grand Cayman and Brazil, which gave suppression of local Ae.aegypti populations with a mating fraction of .
4Dye C (1984) Models for the population dynamics of the yellow fever mosquito, Aedes aegypti. Journal of Animal Ecology 53:247‐268. Harris AF, et al. (2011) Field performance of engineered male mosquitoes. NatBiotechnol 29(11):1034‐1037. Phuc H, et al. (2007) Late‐acting dominant lethal genetic systems and mosquito control. BMC Biology 5: 11.
Oxitec Ltd CONFIDENTIAL Page 21 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
12.3.2 Additional ovitraps outside the Site Area. Additional ovitraps not already proposed in the protocol, will be placed outside the designated Site Area as described in the protocol (Section 11.3). To detect migration of OX513A mosquitoes outside of the Site Area, we propose to monitor urban areas with ovitraps
Figure 2 shows the proposed areas, which will include parts of the Florida Keys Community College, the hospital and an area of the golf course that are in closest proximity to the perimeter of the proposed TA in Key Haven.
If during the course of the monitoring fluorescence is confirmed from trap samples, Oxitec will notify FDA‐CVM within of the confirmation. The OX513A line of Ae. aegypti mosquitoes carries a repressible dominant lethality trait that prevents progeny inheriting the #OX513 rDNA construct from surviving to functional adulthood in the absence of tetracycline. Additionally, FKMCD will continue to undertake routine mosquito control operations in the area as required, and OX513A insects are susceptible to the chemicals used in those control programs.
Oxitec Ltd CONFIDENTIAL Page 22 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd.
Investigational field protocol Effective Date 2...
fraction is calculated from total fluorescent over non‐fluorescent larvae recovered for TA over these .
12.5 Ongoing adaptive management of release rates: Mating fraction will be evaluated every during suppression phase. Release rates will be adjusted according to Table 5 to ensure mating fraction is maintained in target range . In the anticipated event of Ae.aegypti populations decreasing, a corresponding increase in mating fraction will result and therefore release numbers may also be revised downward according to Table 5. If release rate needs to be changed the revised release rate NC will be calculated as described in Section 11.2, with values for mating fraction (PR) and release rates (NR) derived from previous 6‐8 week assessed period instead of rangefinder phase.
Table 5: Adaptive Management of Release Rate based on Mating Fraction
Mating Fraction Release rate adjustment Increase release rate Maintain release rate Reduce release rate* (optional) * Decision is based on logistical considerations including rate of suppression desired, and number of insects available for release.
12.5.1 Releases will occur up to times a week using the adaptive release rate calculated to achieve the Mating Fraction target range. Releases will last up to but may stop earlier if Primary and Secondary objectives have been met.
12.6 Post‐trial monitoring After the cessation of releases, Oxitec proposes to monitor the treatment area (TA), to maximise the likelihood of detection, weekly for a period of immediately post‐ trial. The monitoring tool will be ovitraps as they are more sensitive than adult traps and will also provide evidence of potential mating. Ovitraps will be set at the same trap density used in the trial and traps processed as discussed in Section 12.3 of the protocol.
following the immediate post‐trial period, there will be an additional monitoring period of , where ovitraps at the same trap density used in the trial will be set in the treated area for , collected and screened using visual screening of larvae for fluorescence. If the visual fluorescence is in doubt, then samples will be shipped back to the UK for PCR analysis. If fluorescence is
Oxitec Ltd CONFIDENTIAL Page 24 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
confirmed during this period, then a further screening will be conducted later, i.e. at following the post‐trial period, using the same techniques. If there is no fluorescence confirmed at the ‐trapping period, then the time‐point survey will not be conducted. A further screening will be conducted at following the immediate post‐trial period, regardless of whether fluorescence is confirmed at the trapping periods.
We propose to conclude the periodic post‐trial monitoring at as this will have covered a “mosquito season” which typically in the Florida Keys runs from May to November and therefore includes monitoring of the previous years’ egg bank.
We expect to detect fluorescent samples in the immediate period. The results of this post‐trial survey will be reported to FDA within of the last trap being collected and analysed for fluorescence. If during the course of the subsequent post‐trial monitoring at and post continuous monitoring, fluorescence is confirmed from trap samples, Oxitec will notify FDA‐CVM within of results being confirmed. The OX513A line of Ae. aegypti mosquitoes carries a repressible dominant lethality trait that prevents progeny inheriting the #OX513 rDNA construct from surviving to functional adulthood in the absence of tetracycline. Therefore, OX513A mosquitoes are not expected to establish at the proposed trial site. FKMCD will continue to undertake routine mosquito control operations in the area as required, and OX513A insects are susceptible to the chemicals used in those control programs.
13 DISPOSAL OF UNUSED INSECTS Unused insects and recaptured insects not required for further analysis will be sealed in leak‐ proof containers for appropriate disposal (e.g. autoclaving or freezing at ≤‐15oC for >12 hours) and then disposal via clinical or biohazard waste disposal.
14 OTHER VARIABLES TO BE RECORDED 14.1 Weather Data: A weather station will be located in the vicinity of the Site Area to record weather during the course of the study. Metrological data recorded will include temperature, rainfall, wind speed and direction. 14.2 Sex Sorting Efficiency: Physical sex separation at the pupal stage is used to remove females from OX513A males. A minimum of pupae will be checked from every batch prior to release. If numbers of females exceed the batch will be resorted and only released if ≤ females are detected according to Appendices F and G Sex Sorting of Pupae for Release and Sex Sorting Criteria for Release). Therefore, estimated percentage of females in every
Oxitec Ltd CONFIDENTIAL Page 25 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
batch released, and total number released will be recorded and compiled in final report. 14.3 Alternate mosquito control activities: Mosquito abatement measures may be applied independently of this study by FKMCD in TA and UCA. If these measures are used they will be recorded and included in the final report. 14.4 Community engagement: Communication will be maintained with vector control operatives (FKMCD staff) to address any concerns that might arise in the Site Area during the period of the trial.
15 DATA ANALYSIS 15.1 Primary Objective part 1: To determine whether released OX513A mate with local females resulting in their progeny inheriting a copy of the #OX513 rDNA construct. Ovitrapping will be used to collect eggs from Site Area (TA and UCA). Visual fluorescent screening will be used as primary method for assessing genotype of Ae.aegypti progeny collected from TA and UCA. Fluorescence will indicate presence of the #OX513 rDNA construct. Fluorescent screening cannot distinguish between hemizygous and homozygous individuals for #OX513 rDNA construct. PCR analysis on a subset (those used for assessment of pre‐functional‐adult mortality in Section 11.5) of fluorescent insects will be conducted to differentiate homozygous from hemizygous insects. Detection of OX513A hemizygous progeny from field will serve as the metric to determine the ability of OX513A males to mate with local females in field. The metric for primary objective part 1 will be encompassed in the determination of Primary objective part 2, where fluorescent larvae collected from field will subsequently be reared and genotyped (Section 11.5). As long as fluorescent larva, subsequently confirmed by PCR as being hemizygous, is detected the null hypothesis will be rejected. Samples for PCR will be stored in 70% ethanol and shipped to the UK for the analysis.
15.1.1 Null hypothesis: released OX513A males will fail to mate with local females in environment resulting in progeny not inheriting the #OX513 rDNA construct. 15.1.2 Outcome: Primary Objective part 1 will be met if null hypothesis is rejected by detection of OX513A hemizygous progeny from field collected eggs by fluorescence and PCR methods.
15.2 Primary Objective part 2: To determine whether the progeny inheriting the #0X513 rDNA construct have the expected phenotype resulting in significantly increased mortality whereby there is at least a increase in the proportion of the OX513A progeny dying before they reach functional adulthood compared to progeny from the local Ae.aegypti. The test statistic will be a function of the proportion of larvae not developing to functional adults. We will test the hypothesis that larvae inheriting
Oxitec Ltd CONFIDENTIAL Page 26 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
#0X513 rDNA construct from released males are as likely as local mosquitoes to die before reaching functional adults. The mortality of each larva will be assumed independent of all others, as they are reared at a low density to avoid intraspecific competition, thus the comparison is a straightforward comparison of binomial proportions. The observed relative risk (RR) is the ratio of the proportion of larvae hemizygous for the #0X513 rDNA construct dying before becoming functional adults relative to the proportion of local larvae dying before becoming functional adults. Thus
RR = (XO/NO)/(XL/NL)
where:
NO = sample size of larvae hemizygous for the #0X513 rDNA construct
NL = sample size of local larvae
XO =number of larvae hemizygous for the #0X513 rDNA construct observed to die before reaching functional adulthood
XL = number of local larvae observed to die before reaching functional adulthood
To test the null hypothesis that there is less than or equal to a increase in mortality in the progeny inheriting the #OX513 rDNA construct relative to local comparator larvae without rDNA construct, a log‐binomial model will be used, with genotype (#OX513 rDNA construct vs. local comparator larvae without #OX513 rDNA construct) as a fixed effect and cohort as a random effect. In this model, the estimate for the genotype effect corresponds to an estimate of log(RR), and an approximate 95% confidence interval (CI) for RR can be obtained by exponentiating the limits of the 95% CI for the genotype effect. The null hypothesis is rejected if the estimated RR is greater than and the 95% CI for RR derived from the analysis does not include .
Model will include analysis of cohort effects in line with the study design.
Primary Objective part 2 will be achieved if the null hypothesis is rejected:
Oxitec Ltd CONFIDENTIAL Page 27 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
15.2.1 Null hypothesis: Progeny inheriting the #OX513 rDNA construct will have the same probability of mortality before reaching functional adulthood as local comparator larvae or will have the probability of mortality before reaching functional adulthood increased by less than or equal to relative to local comparator larvae without #OX513 rDNA construct. 15.2.2 Outcome: Null hypothesis will be rejected if the ratio of the probability of mortality of OX513A (hemizygous) larvae from field ovitrap collection before reaching functional adulthood in the absence of tetracycline over the probability of mortality of the local Ae.aegypti larvae before reaching functional adulthood is significantly greater than , as judged by the 95% confidence interval.
15.3 Secondary Objective: Sustained release of OX513A males results in a statistically significant suppression of the local populations of Ae.aegypti relative to untreated comparator area. Two trapping methods are used in this analysis; ovitraps and BG‐Sentinel adult traps to provide a measure of female Ae.aegypti abundance. Similar analysis will be used for both resulting in a Relative Ovitrap Index and Relative Adult Density. 15.3.1 Calculation of Relative Ovitrap Index: The test statistic will be the Relative Ovitrap Index where at a given time point of trap collection:
Where TA = treated area UCA = untreated comparator area
For a single time point of trap collection, a bootstrap confidence interval5 can be obtained for the relative ovitrap index by:
. bootstrapping (resampling from a binomial6) the ovitrap indices for the treated and untreated area,
5Manly, B.F.J. (2007). Randomization, Bootstrap and Monte Carlo Methods in Biology (Boca Raton, FL, USA, Chapman & Hall).
Oxitec Ltd CONFIDENTIAL Page 28 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
. taking the ratio of ovitrap indices for each bootstrap sample. The central estimate will be the observed relative ovitrap index, with the 95% bootstrap confidence interval being obtained from the 2.5th and 97.5th centiles of the distribution of bootstrap estimates on bootstrap samples.
Change in Relative Ovitrap Index will be derived from the mean of multiple ovitrap collections before suppression (minimum ) and a mean from a period, calculated as rolling average, from the suppression phase.
When the time period under analysis includes multiple ovitrap collections from T different traps, the
1
Where TA = treated area UCA = untreated comparator area
For multiple trap collections, a bootstrap confidence interval can be obtained for the relative ovitrap index by:
. bootstrapping (nonparametric resampling) of the T trap‐specific ovitrap indices for the treated and untreated area, . taking the mean for each bootstrap sample separately for the treated and untreated area, . taking the ratio of mean ovitrap indices for each bootstrap sample. The central estimate will be the observed relative ovitrap index, with the 95% bootstrap confidence interval being obtained from the 2.5th and 97.5th centiles of the distribution of bootstrap estimates on bootstrap samples.
6 When bootstrapping for a single time point, parametric binomial resampling is performed because the data are simply a positive or negative result for each trap.
Oxitec Ltd CONFIDENTIAL Page 29 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
The Relative Ovitrap Index prior to suppression phase (minimum ) will be compared with mean Relative Ovitrap Index following initiation of Suppression phase .
We will estimate the reduction in the Relative Ovitrap Index by calculating:
% 1 ∗ 100
We will obtain the bootstrap confidence interval for this by recalculating this % reduction for all 10,000 bootstrap samples, with the 95% bootstrap confidence interval being obtained from the 2.5th and 97.5th centiles of the distribution of bootstrap estimates of % reduction on bootstrap samples. We define suppression to be defined as a statistically reduced relative ovitrap index by at the 0.05 level of statistical significance compared to UCA.
15.3.1.1 Null Hypothesis: Sustained release of OX513A will not affect the Ae.aegypti population in TA or will result in suppression of the Ae.aegypti population in the TA by relative to untreated comparator areas. This hypothesis is valid for both ovitraps and BG‐Sentinel adult traps.
15.3.2 Calculation of Relative Adult Density: The test statistic will be the Relative Adult Density where at a given time point of trap collection:
For a single time point of trap collection, a bootstrap confidence interval can be obtained for the relative adult density by:
. bootstrapping (resampling with replacement) the numbers trapped for the treated and untreated area, . taking the ratio of average adult densities for each bootstrap sample.
Oxitec Ltd CONFIDENTIAL Page 30 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
The central estimate will be the observed relative adult density, with the 95% bootstrap confidence interval being obtained from the 2.5th and 97.5th centiles of the distribution of bootstrap estimates.
Change in Relative Adult Density will be derived from the mean of multiple BG‐Sentinel adult trap collections before suppression and a mean from a period, calculated as rolling average, from the suppression phase.
When the time period under analysis includes BG‐Sentinel adult trap collections from T different traps, the analysis will be calculated as:
1
For multiple trap collections, a bootstrap confidence interval can be obtained for the relative average adult density by:
. bootstrapping (resampling with replacement) of the T trap‐specific numbers trapped for the treated and untreated areas, . taking the average for each bootstrap sample separately for the treated and untreated areas, . taking the ratio of average adult densities for each bootstrap sample. The central estimate will be the observed relative adult density, with the 95% bootstrap confidence interval being obtained from the 2.5th and 97.5th centiles of the distribution of bootstrap estimates on bootstrap samples.
The Relative Adult Density prior to suppression phase will be compared with Relative Adult Density following initiation of Suppression phase .
We will estimate the reduction in the Relative Adult Density by calculating:
Oxitec Ltd CONFIDENTIAL Page 31 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
% 1 ∗ 100
We will obtain the bootstrap confidence interval for this by recalculating this % reduction for all bootstrap samples, with the 95% bootstrap confidence interval being obtained from the 2.5th and 97.5th centiles of the distribution of bootstrap estimates of % reduction on bootstrap samples. We define suppression to be defined as a statistically reduced relative average density by at the 0.05 level of statistical significance compared UCA.
16 SPECIFIC POTENTIAL RISKS RELATED TO THIS PROTOCOL INCLUDE 16.1 Tetracycline contamination affecting the laboratory analysis of the larvae recovered from ovitraps: This will be mitigated by using new equipment for tetracycline –free analyses and conducting rearing of field samples in a separate laboratory to the mass rearing unit. If tetracycline contamination is suspected (for example by a large deviation in survival rates), then the analysis will be repeated using new equipment.
16.2 Numbers of eggs collected from the ovitraps are too low for the analysis for Primary Objective part 2: This will be mitigated by extending collections until sufficient numbers are obtained, over multiple weeks.
16.3 If non Ae.aegypti species are found in ovitraps within the Site Area (TA and UCA): If this is the case the mating fraction is unaffected but the analysis of the Primary Objective part 2 may be affected. Should this occur mitigations will include any or all of the following: . the selection of further cohorts without non Ae.aegypti . increasing the trap catch to achieve target number of Ae.aegypti larvae . removing the non‐Ae.aegypti species from the trap . inclusion of an additional step for the testing of species in that objective.
16.4 Traps are missing from the locations on collection: Trap loss is accounted for in the protocol. If traps are missing from the locations, it is already stated in the protocol that traps may be replaced in different locations or redeployed. The monitoring systems have been designed to be robust to a proportion of traps failing and average recapture per trap can still be calculated.
Oxitec Ltd CONFIDENTIAL Page 32 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
16.5 Releases of insects may be prevented or stopped by community and/or NGO protest activities: If releases are prevented by protest activities this will break the sustained releases and the secondary objective to demonstrate population suppression may not be achieved. The time frame of the study may need to be extended to allow sufficient sustained releases of OX513A to suppress the local population of Ae.aegypti. If the stoppage leads to premature termination of the study a short final report will be submitted.
16.6 Adverse weather event may extend the suppression window: An adverse weather event (e.g. tropical storm, hurricane, flooding) may prevent releases taking place for an extended time period. If this is the case then the time frame of the study may need to be extended to allow sufficient sustained releases of OX513A to supress the local population of Ae.aegypti and the secondary objective to be achieved. If the length of the stoppage leads to premature termination of the study a short final report will be submitted.
16.7 Mosquito abatement measures: It is unethical not to continue with mosquito abatement activities during the course of the investigational use in the Site Area. Mosquito abatement measures used will be recorded. Data points may need to be removed if mosquito abatement measures are unequal between sites or adulticides have been used on the same day as releases of OX513A. This may increase the length of the data collection phase to reach the minimum timeframe.
16.8 Lack of adaptation of local Ae.aegypti to laboratory rearing: The locally caught Ae.aegypti need to be reared in a laboratory environment to assess the phenotype, whereas OX513A is already well adapted to laboratory rearing. The potential consequence is that the analysis may fail if the local Ae.aegypti do not rear as well as the OX513A in the laboratory. If this is found to be the case, this can be mitigated by modification of the rearing methodology to better suit the local Ae.aegypti, such as modifying the amount of diet. Modifications that may be required will be recorded in the relevant SOP.
16.9 Potential to release homozygous female OX513A: OX513A males are manually sorted from OX513A females prior to release. A QC process is in place where a batch for release is re‐sorted should % females in the males be exceeded (Appendix F and G). There is therefore the potential that % homozygous females could contribute to the fluorescence of the progeny determinedm fro ovitraps, if the analysis was based on fluorescence alone. The use of PCR to determine only the hemizygous progeny will therefore provide evidence of the ability of OX513A to mate with local Ae.aegypti females in a population without interference from homozygous female OX513A.
Oxitec Ltd CONFIDENTIAL Page 33 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
16.10 Potential for failure of the self‐limiting trait: If in the unlikely event that the phenotypic trait is not achieving the objectives of this protocol during the investigational period, it will be detected through the fluorescence analysis as no statistically significant difference in mortality between OX513A and wild‐type progeny and the trial will be stopped. If the study is stopped, additional mosquito control measures can continue to be applied such as the use of larvicides or adulticides.
Risks to human, animal health and the environment are assessed in an Environmental Assessment that has been prepared to accompany this protocol.
17 EMERGENCY MEASURES Emergency measures are described in SOP (Appendix H).
18 RETENTION OF DATA Raw data will be retained by the principal investigator for a period of at least 5 years.
19 PERSONNEL Study Director: Principal Investigator: Statistician: Other staff: Florida Keys Mosquito Control District staff and/or Oxitec staff Sponsor: Oxitec Ltd, 71 Innovation Drive, Milton Park, Abingdon, Oxfordshire, OX14 4RQ, UK Sponsor representative:
20 PROTOCOL DEVIATIONS It is possible, depending on conditions at the time, and the nature of the biological system that is being studied, that deviations may be required.
Deviations which result in a systematic change, e.g. in the SOPs or in the protocol, will be made by a protocol amendment.
Unforeseen circumstances, which have only a one‐time effect (e.g. different date of sample collection) will be reported only in the raw data and the final report.
Oxitec Ltd CONFIDENTIAL Page 34 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
21 APPENDICES
Appendix Title Appendix A Supplementary information to Protocol OX‐ USA‐1 Appendix B OX513A quality control SOP’s comprising of
OX513A Quality Control Protocol for Assessment of Penetrance and Doxycycline Sensitivity
OX513A Quality Control Protocol for Assessment of Mating Competitiveness
2.a OX513A Quality Control Protocol for Colony Genotyping
Appendix C SOP Mass Rearing Process for Release Appendix D: SOP Field Penetrance Assay
Appendix E New Genomic DNA extraction using Purelink Genomic DNA Kit by Invitrogen Appendix F SOP OX513A Sex Sorting of Pupae for Release
Appendix G SOP OX513A Sex Sorting Criteria for Release
Appendix H SOP Emergency Response Procedures
Oxitec Ltd CONFIDENTIAL Page 35 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd. Investigational field protocol Effective Date 2...
This page left intentionally blank.
Oxitec Ltd CONFIDENTIAL Page 36 of 36 This document contains confidential business information which is proprietary and the publication or disclosure of which would harm the legitimate business interests of Oxitec Ltd.
Aedes aegypti OX513A
Supplementary Information Supporting Protocol OX‐ USA‐1: Trial Protocol for the Evaluation of Efficacy
Oxitec Ltd, 71 Innovation Drive, Milton Park, Oxfordshire, UK OX14 4RQ
CONFIDENTIAL 1
SUPPLEMENTARY INFORMATION SUPPORTING PROTOCOL OX‐USA‐1: Evaluation of OX513A Aedes aegypti ability to transfer #OX513 rDNA construct by mating, resulting in the progeny having increased mortality.
Contents List of Abbreviations/Glossary of Terms ...... 3 Study objectives ...... 3 . Primary Objective part 1: To determine whether released OX513A mate with local females resulting in their progeny inheriting a copy of the #OX513 rDNA construct...... 4 . Primary Objective part 2: To determine whether progeny inheriting the #OX513 rDNA construct have the expected phenotype resulting in significantly increased mortality
before they reach functional adulthood...... 4 . Secondary Objective: To determine whether sustained release of OX513A males results in a statistically significant suppression with 95% confidence interval (CI)) of the local populations of Ae.aegypti relative to an untreated comparator area...... 4 Product claim (proposed) ...... 4 Introduction ...... 4 Primary Objectives ...... 4 Secondary Objective ...... 5 Preparation Phase ...... 9 Rangefinder Phase...... 9 Suppression Phase ...... 10 Mosquito Monitoring Systems ...... 10 Ovitraps ...... 11 Larval screening ...... 11 BG‐Sentinel adult traps ...... 12 Study Design ...... 13 REFERENCES ...... 14 APPENDIX A of Supplementary Information ...... 16 Methods ...... 19 Study area ...... 19 Community Engagement/regulatory ...... 20
CONFIDENTIAL 2
Mosquitoes ...... 21 Mass rearing of OX513A ...... 21 Eclosion and Release ...... 21 Monitoring ...... 22 Statistics ...... 22 Eclosion and Release ...... 35 Monitoring ...... 35 References ...... 36
List of Abbreviations/Glossary of Terms
Abbreviation/Term Description IgG Immunoglobulin G LSTMH Liverpool School of Tropical Medicine and Hygiene Mating fraction Proportion of local females that mate release OX513A vs. local male Ae.aegypti RIDL Release of Insects carrying a Dominant Lethal RIDL‐SIT Release of Insects carrying a Dominant Lethal‐ Sterile Insect Technique SIT Sterile Insect Technique Treatment Area Area under treatment with OX513A Untreated Comparator Comparator areas without OX513A treatment Areas WHO World Health Organization
Study objectives The study has two primary objectives and one secondary objective:
The primary objective of the study, which is in two parts, is to demonstrate released OX513A males mate with local Ae.aegypti females; the progeny of OX513A inherit the #0X513 rDNA construct, causing premature mortality before reaching functional adulthood1, thereby disrupting the reproductive cycle of local population.
The secondary objective is to demonstrate sustained releases of OX513A males can disrupt the reproductive cycle sufficiently to result in a suppression of local population in a Treatment Area (TA) relative to Untreated Comparator Area (UCA). This study therefore is pivotal towards providing data to support efficacy claims relating to the use of OX513A Ae.aegypti for vector
1 Functional Adult: Adults that are capable of flying and therefore assumed to be fully functional in ability seek hosts for blood meal and mate in order to reproduce. Adults that dye on emergence or are incapable of maintaining sustained flight, are not considered to be functional adults.
CONFIDENTIAL 3
For the system to be effective in context of a SIT suppression program, released males need to seek and mate with local females and transfer #OX513 rDNA construct to offspring. In essence this is equivalent to evaluating the ‘delivery’ system for the ‘active ingredient’ or ‘antigen’ for conventional insecticide or vaccine respectively.
In order to evaluate the “delivery system” we will recover progeny from local females in the field (in form of eggs) following release of OX513A males. Detection of progeny containing the #OX513A rDNA construct, and confirmed as hemizygous by PCR will demonstrate that the released OX513As male have successfully mated with local females and transferred the #OX513A rDNA construct to the progeny.
Primary Objective part 2: The progeny inheriting the #OX513 rDNA construct have the expected phenotype resulting in significantly increased mortality, whereby there is OX513A progeny dying before they reach functional adulthood.
Having ‘delivered’ #OX513 rDNA construct into the target, we need to evaluate if it performs as intended. Taking the insecticide analogy, this would be evaluating the lethality of active ingredient on the target insect. For a vaccine, this would equate to checking vaccine (antigen) elicited an immune response that conferred protection for future challenge by target pathogen. In the case of #OX513 rDNA construct, the intended action is to confer premature mortality to progeny of matings with OX513A‐ expressing insects before reaching functional adulthood, and thereby stop its potential to contribute to the next generation.
To evaluate this we intend to rear a subset of larvae derived from field collected eggs and assess mortality prior to functional adulthood. If #OX513 rDNA construct functions as intended, mortality in larvae containing the #OX513 rDNA construct would manifest in significantly higher premature mortality compared to local larvae with no #OX513 rDNA construct.
Secondary Objective Intended use of OX513A males is within the context of an SIT control program, where premature mortality of offspring induced by #OX513 rDNA construct is analogous to sterilising males with irradiation for conventional SIT. With sustained release of sterile males it is possible to interfere with the natural reproductive cycle of local Ae aegypti population, and lead to its suppression over time.
The rationale and methodology used to demonstrate suppression of local population will broadly follow those used in previous open field demonstrations using OX513A males, conducted in Grand Cayman, Cayman Islands (Harris et al., 2012) and Brazil (Carvalho et al, 2015).
Mosquito populations are characteristically dynamic, often showing large seasonal fluctuation. This is true of the proposed study area (Key Haven, Florida) with high Ae aegypti numbers between May to October, and substantially lower numbers associated with cooler dryer weather from November to April (Figure 1).
CONFIDENTIAL 5
CONFIDENTIAL 6
governing the rate of population reduction is the mating fraction of local females mating released OX513A over local males. The threshold mating fraction for local elimination in the absence of immigration can be estimated from models of mosquito population dynamics as 0.13‐0.57 (Dye, 1984; Phuc et al., 2007). Observations from previous field studies with OX513A show suppression of local population was possible at 0.12 and 0.43 mating fraction in Grand Cayman (Harris et al. 2012) and Brazil (Appendix A) respectively. For operational deployment, we therefore aim for release rate sufficient to achieve a mating fraction .
The ratio of OX513A : local males (over flooding ratio) resulting from sustained release is critical to the probability of local female Ae aegypti mating with an OX513A male, rather than local male counterpart. Therefore release rate ( equivalent to the “dose”) necessary to achieve a given mating fraction is directly proportional to the local mosquito population. The greater the over flooding rate achieved, the greater the mating fraction and likely impact on local population. It is therefore not possible to recommend a standard release rate (dose) per Hectare, as is often done in conventional pesticide application.
The inclusion of a fluorescent marker in the #OX513 rDNA construct provides a convenient way to assess parentage (OX513A or local males) from field collected eggs, and thus estimate the mating fraction attained from a given release rate. This is further described in larval screening below.
The first phase of releases, known as ‘Rangefinder’, is designed to calibrate known constant release rate against local population density in terms of mating fraction achieved. If initial release rate is insufficient to achieve a target mating fraction the release rate can be adapted accordingly for the main Suppression phase:
Based on the known release rate in Rangefinder phase (NR) and estimated mating fraction (PR)
observed it is possible to estimate the release rate (NC) required to achieve mating fraction or above, assuming all other variables remain the same:
1 1 Where:
NC = Release rate in Suppression phase
NR = Release rate in Rangefinder phase
PR = Proportion fluorescent (Mating Fraction) larvae observed in Rangefinder phase
PC = Target Mating Fraction (proportion fluorescent larvae) in Suppression phase (=0.5)
Given the time lag inherent in the mode of action (as described above), we would normally expect several generations before measurable impact on population density is detectable. The fluctuating nature of Aedes population dynamics, and high level of variability, means we would monitor for
CONFIDENTIAL 8 minimum 8 weeks before treatment, and compare with 4 week period (calculated as a rolling average) after suppression is achieved.
One of the benefits of SIT, over other control methods, is that as local population decreases, the efficacy of release increases as it becomes possible to achieve higher over‐flooding rate in relation to local population. This is contrary to conventional control methods where you tend to get diminishing returns from control effort with a reduced pest population, as targeting the remnant local population becomes increasingly difficult. In practice SIT is seldom used as a stand‐alone solution, rather it is deployed as part of an integrated pest management system in concert with conventional control methods.
The study will be conducted in the following phases:
Preparation Phase Preparation phase will include technology transfer and initial shipments of eggs from Oxitec Ltd. UK laboratories to the local mass rearing unit. Mass rearing will be initiated and rearing protocols adapted to conditions in local mass rearing unit, if necessary. Monitoring (of the local Ae. aegypti population) will be initiated giving relative measure of population density within and between treated and untreated comparator control area. This will provide a basis to against which subsequent change in treatment area Ae aegypti population can be assessed in relation to untreated comparator area in order to verify the secondary objective of demonstrating population suppression. Monitoring will consist of ovitrap and BG‐ sentinel adult trap survey to assess relative abundance of local Ae aegypti.
Rangefinder Phase Rangefinder phase will be used to establish the initial release rate needed for suppression, specific to local Ae aegypti population density in the treatment area. Limited releases will be initiated for a period of 6‐8 weeks. Ae aegypti release rates are calculated in terms of males/person/week. For the treatment area, numbers released will be a factor of human population density and estimated Ae aegypti infestation level:
HUMAN POPULATION DENSITY ‐ Estimate human population density in the overall treatment area.
INITIAL INFESTATION LEVEL – Estimate Ae. aegypti infestation level based on available information including, historical Aedes surveillance data, seasonality, epidemiology records, existing mosquito abatement and qualitative factors such as housing type and proliferation of breeding sites. Assessments will be subjective, based on the judgement of the Study Director. Ae. aegypti INITIAL INFESTATION LEVEL will be categorised as VERY LOW, LOW, MEDIUM, or HIGH (as described in Table 1).
CONFIDENTIAL 9
Table 1 Guideline minimum Initial Release Rate (OX513A males/person/week) according to estimated infestation level.
Initial Ovitrap Minimum infestation Index Release Rate per person in level of treatment area Ae.aegypti HIGH
MEDIUM
LOW VERY LOW
The critical factor here is the release of a known and constant rate of OX513A. The mating fraction estimated, from the known release rate, will be used to calculate release rates required to achieve the target mating fraction for population suppression with OX513A. A subset of field collected eggs from ovitraps in rangefinder and/or suppression phases will be used to evaluate the two primary objectives of the study.
Suppression Phase Release numbers will be proportional to the size of the Ae aegypti population in the field as determined by the Rangefinder phase. Releases will be conducted in a grid in order to ensure even and consistent coverage of the treatment area. Releases will occur at regular intervals, up to 3 times/week. On‐going monitoring from ovitraps will assess the relative impact of the release on local populations of Ae aegypti compared to untreated control sites2.
Release rates will be revised in an adaptive management system based on sustaining a mating fraction . In the anticipated event of Ae aegypti populations decreasing, release numbers may be revised downward correspondingly. Alternatively, rapid increase in local populations associated with seasonality may require sustained or even an increase in the release rate to maintain mating fraction in desired range (≥0.5) to supress the local Ae aegypti population.
This is linked to the secondary objective of the trial: Sustained release of OX513A males results in a statistically significant suppression of the local populations of Ae aegypti relative to untreated comparator areas.
Mosquito Monitoring Systems The monitoring system used is based on well recognised methodologies for mosquito monitoring. Methods have been tested extensively and used in similar studies to evaluate field efficacy of sustained release of OX513A males for the control of Ae aegypti populations (Harris et al., 2012)(see Appendix A).
2 Untreated comparator site means without treatment with OX513A. Other mosquito control treatments may continue to be used.
CONFIDENTIAL 10
Ovitraps Ovitraps are a widely used method for measuring the presence and abundance of container breeding mosquitoes, principally Ae aegypti and Ae albopictus (Silver, 2008). There are a number of variations on the same theme but the principle is the same. In essence, ovitraps induce container breeding mosquitoes to lay eggs in ovitraps for subsequent identification. They consist of a simple water container (~300 ml) with dark sides, constituting an attractive oviposition site (Figure 2). A partially submerged ovipositon substrate (eg hardboard paddle, strip of germination paper) is placed within the ovitrap as a suitable oviposition substrate. Aedes mosquitoes do not lay eggs directly in water, rather on a substrate that is subject to intermittent flooding such as plant material, soil on the edge of a pond or sides of man‐made containers. Ovitraps provide an indirect measure of female abundance allowing assessment of local population, without interference from released OX513A males. Ovitraps will be used to monitor changes in relative abundance of the field population of Ae aegypti, and collect eggs for the evaluation of matings between released OX513A males and local females.
Figure 2: Ovitrap
Larval screening Larval screening will take place for the duration of the trial after releases start. Eggs collected from ovitraps will be matured, by allowing them to dry at room temperature (range 20‐30°C) for a minimum 2 days (with upper limit of 14 days). Larvae can then be visually screened for the presence of genetic material from OX513A. The OX513A strain has a fluorescent marker gene (Phuc et al., 2007), which is visible at wavelength excitation of 520‐550 nm, emission 580+ nm under a fluorescence microscope (e.g. Leica MZ10F) in a darkened room. Some training of staff is required to optimise the visualisation.
CONFIDENTIAL 11
A subset of field collected eggs from ovitraps in rangefinder and/or suppression phases will be used to evaluate the two primary objectives. The presence of #OX513 rDNA in fluorescent larvae being confirmed by PCR to satisfy Primary Objective part 1 that OX513A males can mate and transfer #OX513 rDNA construct to progeny in the field. The Primary Objective part 2 to assess expression of self‐limiting trait will be assessed by rearing larvae to adult and demonstrating higher mortality in larvae containing #OX513 rDNA construct, relative to larvae without #OX513A rDNA construct.
BG‐Sentinel adult traps BG‐Sentinel adult trap (www.biogents.com) was developed specifically targeting Aedes mosquitoes. It utilises a combination of factors to trap live adults. The trap incorporates a fan generating a counter flow air stream that produces a plume of kairomones (volatiles associated with host) attracting mosquitoes (figure 3). The trap also uses visual attractant cues; with a black collection tube contrasting against white casing housing the trap. As mosquitoes get close they are caught in the rapid down draft airflow generated by the fan and get trapped in a net bag within the centre of the trap. These traps have been extensively field tested in direct comparison with other live adult sampling methods including the CDC backpack aspirator, human bait landing catch, and Fay‐Prince traps (Krockel et al., 2006; Maciel‐De‐ Freitas et al., 2007; Maciel‐De‐Freitas et al., 2006; Williams et al., 2006; Williams et al., 2007). Traps collect both males and female in comparable numbers and are also effective at sampling Ae albopictus (Ritchie et al., 2006). In most direct comparison studies, BG‐Sentinel adult traps generally outperform other types of traps, and capture comparable or significantly more Ae aegypti over 24 h, compared to a standard CDC backpack aspirator household survey (15‐30 min).
Unlike aspiration surveys, BG‐Sentinel trap catches will be independent of operator skill; are less intrusive that some other sampling methods; and less labour intensive. The negative is that they are relatively expensive ($100‐300) and need to be connected to mains power or heavy‐duty batteries (small car/motorbike) to operate the fan.
Figure 3. BG‐Sentinel adult trap. A) Functional Diagram ‐ longitudinal section: CB, catch bag; T, black tube; F, Fan; BGL, BG‐Lure. B) Photo of trap. Arrows indicate direction of airflow.
A) B)
CONFIDENTIAL 12
Study Design Due to nature of field site and scale of proposed study, randomised replicated block/clusters are not feasible. Selection of treatment and untreated comparator areas are matched as close as possible in terms of habitat, mosquito population and geographic proximity and consequently climate. However, Ae aegypti are highly variable both temporally and spatially, and finding paired treatment and untreated comparator sites with identical population density and dynamic is challenging. We therefore propose analysis of outcome of sustained OX513A release in terms of impact on population in treatment area relative to an untreated comparator area. Relative measures allow direct comparison between treatment and untreated comparator area in the context of overarching seasonal factors and other abiotic effects. This provides a more robust approach to assessing impact of OX513A releases compared to assessing change before and after within a treated site.
CONFIDENTIAL 13
REFERENCES Alphey, L., Benedict, M., Bellini, R., Clark, G., Dame, D.A., Service, M.W., and Dobson, S.L. (2010). Sterile‐Insect Methods for Control of Mosquito‐Borne Diseases: An Analysis. VECTOR‐BORNE AND ZOONOTIC DISEASES 10.
Carvalho, D. O., et al. (2015). "Suppression of a Field Population of Aedes aegypti in Brazil by Sustained Release of Transgenic Male Mosquitoes." PLoS Negl Trop Dis 9(7): e0003864.
Dye, C. (1984). Models for the population dynamics of the yellow fever mosquitoe, Aedes aegypti. Journal of Animal Ecology 53, 247‐268.
Focks, D., Brenner, R., Hayes, J., and E, D. (2000). Transmission thresholds for dengue in terms of Aedes aegypti pupae per person with discussion of their utility in source reduction efforts. Am J Trop Med Hyg 11‐18.
Harris, A.F., McKemey, A.R., Nimmo, D., Curtis, Z., Black, I., Morgan, S.A., Oviedo, M.N., Lacroix, R., Naish, N., Morrison, N.I., et al. (2012). Successful suppression of a field mosquito population by sustained release of engineered male mosquitoes. Nat Biotech 30, 828‐830.
Krockel, U., Rose, A., Eiras, A., and Geier, M. (2006). New tools for surveillance of adult yellow fever mosquitoes: Comparison of trap catches with human landing rates in an urban environment. Journal of the American Mosquito Control Association 22, 229‐238.
Maciel‐De‐Freitas, R., Codeco, C.T., and Lourenco‐De‐Oliveira, R. (2007). Body size‐associated survival and dispersal rates of Aedes aegypti in Rio de Janeiro. Medical and Veterinary Entomology 21, 284‐292.
Maciel‐De‐Freitas, R., Eiras, Á.E., and Lourenço‐de‐Oliveira, R. (2006). Field evaluation of effectiveness of the BG‐Sentinel, a new trap for capturing adult Aedes aegypti (Diptera: Culicidae). Mem Inst Oswaldo Cruz 101, 321‐325.
Phuc, H., Andreasen, M., Burton, R., Vass, C., Epton, M., Pape, G., Fu, G., Condon, K., Scaife, S., Donnelly, C., et al. (2007). Late‐acting dominant lethal genetic systems and mosquito control. BMC Biology 5, 11.
Ritchie, S.A., Moore, P., Carruthers, M., Williams, C., Montgomery, B., Foley, P., Ahboo, S., Van Den Hurk, A.F., Lindsay, M.D., Cooper, B., et al. (2006). Discovery of a Widespread Infestation of Aedes albopictus in the Torres Strait, Australia Journal of the American Mosquito Control Association 22, 358– 365.
Silver, J.B. (2008). Mosquito Ecology ‐ Field Sampling Methods, Third edn.
Williams, C.R., Long, S.A., Russell, R.C., and Ritchie, S.A. (2006). Field efficacy of the BG‐Sentinel compared with the CDC Backpack Aspirator and CO2‐baited EVS trap for collection of adult Aedes aegypti in Cairns, Queensland, Australia. Journal of the American Mosquito Control Association 22: , 296‐300.
CONFIDENTIAL 14
Williams, C.R., Long, S.A., Webb, C.E., Bitzhenner, M., Geier, M., Russell, R.C., and Ritchie, S.A. (2007). Aedes aegypti Population Sampling Using BG‐Sentinel Traps in North Queensland Australia: Statistical Considerations for Trap Deployment and Sampling Strategy. Journal of Medical Entomology 44, 345‐350.
CONFIDENTIAL 15
APPENDIX A of Supplementary Information
Suppression of a field population of Aedes aegypti in Brazil by sustained release of transgenic male mosquitoes
Danilo O. Carvalhoa,b,1, Andrew R. McKemeya,1,*, Luiza Garzierac, Renaud Lacroixa, Christl A. Donnellyd,
Luke Alpheya, Aldo Malavasic and Margareth L. Capurrob,e aOxitec Ltd, 71 Innovation Drive, Milton Park, Abingdon, Oxfordshire, UK bDepartamento de Parasitologia, Instituto de Ciências Biomédicas, Universidade de São Paulo, BR cMoscamed Brasil, Juazeiro, Bahia, BR dMedical Research Council Centre for Outbreak Analysis and Modelling, Department of Infectious
Disease Epidemiology, Faculty of Medicine, Imperial College London, St Mary's Campus, London, UK eInstituto Nacional de Ciência e Tecnologia em Entomologia Molecular (INCT‐EM), BR
1 Contributed equally
* Corresponding author [email protected] [email protected] [email protected] [email protected] [email protected] [email protected]
CONFIDENTIAL 16 [email protected] [email protected]
Abstract
The increasing burden of dengue and the relative failure of traditional vector control programs highlight the need to develop new control methods. RIDL‐SIT is one such promising method and has already reached the stage of field evaluation. Sustained releases of RIDL OX513A Aedes aegypti males led to 80% suppression of a target wild Ae. aegypti population in the Cayman Islands in 2010. Here we describe sustained series of field releases of RIDL Ae. aegypti males in a suburb of Juazeiro, Bahia, Brazil. This study spanned over a year and reduced the local Ae. aegypti population by 95% (95% CI: 92.2%‐97.5%) based on adult trap data and 81% (95% CI: 74.9‐85.2%) based on ovitrap indices compared to the adjacent no‐release control area. The mating competitiveness of the released males (0.031; 95% CI: 0.025‐0.036) was similar to that estimated in the Cayman trials (0.059), indicating that environmental and target‐strain differences had little impact on the mating success of the OX513A males. We conclude that sustained release of OX513A RIDL males may be an effective and widely useful method for suppression the key dengue vector Ae. aegypti. The observed level of suppression would likely be sufficient to prevent dengue epidemics in most or all local transmission settings.
Keywords: Dengue control; RIDL; OX513A; Brazil; genetic control; Aedes aegypti
Author Summary
Dengue is a major mosquito‐borne disease, increasing in prevalence and severity; there are no specific drugs or licensed vaccine. It is primarily transmitted by one mosquito species, Aedes aegypti. We released transgenic ‘sterile’ male mosquitoes in Itaberaba, a suburb of Juazeiro, a Brazilian city. Sustained release of these males, whose offspring typically die before adulthood as a consequence of the transgenic modification, strongly suppressed the target wild population – by 80‐95% according to different measures. These data are consistent with previous releases in the Cayman Islands, suggesting that differences between the two locations, including the environment or wild mosquito strain, made little difference. Mathematical models suggest that this degree of suppression would be highly effective in preventing epidemic dengue.
CONFIDENTIAL 17
Introduction
Dengue is second only to malaria as most important mosquito‐borne disease. Unlike malaria and other major infectious diseases, dengue is increasing in incidence and severity, currently inflicting 50‐390 million or more cases per year worldwide (Bhatt et al., 2013; WHO‐TDR, 2009) . Dengue is widespread in tropical and sub‐tropical areas and is primarily associated with its principal vector Aedes aegypti (L.).
Dengue was reintroduced in Brazil in 1981 (Boa Vista, State of Roraima), after being almost entirely absent for at least 20 years following DDT‐based vector control. Brazil now has serotypes 1‐3 circulating throughout the country; in addition serotype 4 was recently detected in several states (de Souza et al., 2011). In an analysis of dengue in the Americas in 2000‐2007, Brazil was found to have the highest number of cases and economic burden (Shepard et al., 2011); more recently (Bhatt et al., 2013) estimated 16 million total infections annually. While this in part reflects the size of the Brazilian population, Wilder‐Smith et al. (Luz et al., 2009) concluded that the dengue burden is at least as high as the burden of other major infectious diseases that afflict the Brazilian population, including malaria. A cross‐sectional seroepidemiologic survey conducted in Recife, state of Pernambuco, Brazil, in 2006 found overall dengue virus IgG prevalence to be 80% indicating that the large majority of inhabitants have been infected at least once (Rodriguez‐Barraquer et al., 2011); these authors estimated that 5.2% of susceptible individuals become infected each year by each serotype and that this had increased sharply over the previous 20 years.
There are no specific drugs or licensed vaccine for dengue, so efforts to reduce transmission depend entirely on vector control (WHO‐TDR, 2009). However, even the most highly‐resourced and well‐ implemented programmes, such as in Singapore, have not been able to prevent epidemic dengue using current methods (Lee et al., 2012; Ooi et al., 2006; Wilder‐Smith et al., 2004). Furthermore, existing control tools are threatened by actual or potential spread of resistance in the vector population. Therefore there is an urgent need to develop new methods. The use of transgenic vectors may provide a set of new methods for reducing the density or vectorial capacity of vector populations. Here we describe a field evaluation of one prominent transgenic‐vector strategy, the use of male mosquitoes carrying a lethal or autocidal transgene in a sterile‐male‐release system.
The Sterile Insect Technique (SIT) is a genetic control system based on the release of large numbers of radiation‐sterilised insects. These mate with wild insects of the same species and thereby reduce the reproductive potential of the wild pest population, as they produce no or fewer viable offspring due to
CONFIDENTIAL 18 the radiation‐induced presence of lethal mutations in their gametes (Dyck et al., 2005; Knipling, 1955). Though successfully used against several agricultural pests, trials against mosquitoes have met with less success (Benedict and Robinson, 2003; Klassen and Curtis, 2005). This is in part due to the somatic damage, and associated performance reduction in the sterile insects, which inevitably accompanies radiation‐sterilisation. Interestingly, one successful example of SIT in mosquitoes used a chemosterilant in place of radiation (Lowe et al., 1980). Modern genetics can potentially overcome this problem, for example by using an engineered repressible dominant lethal gene in place of radiation (RIDL, Thomas et al., 2000). Operationally, the system would look very similar to SIT, and would share the clean, species‐ specific characteristics, and similarly benefit from the female‐seeking ability of the released males. However the insects would not be irradiated, rather they would be homozygous for a transgene which, when transmitted to an embryo via the sperm, would lead to death of the zygote at some stage in development (Alphey et al., 2010; Catteruccia et al., 2009). As well as avoiding the need for radiation, by adjusting the time of death one can improve efficiency against target populations with significant density‐dependence (Atkinson et al., 2007; Phuc et al., 2007). Simulation modelling suggests that such a method would potentially be effective and economical against Ae. aegypti (Alphey et al., 2011; Atkinson et al., 2007).
After extensive laboratory development and testing, field testing of engineered insects has begun, with encouraging results. In particular, in the Cayman Islands a RIDL strain of Ae. aegypti, OX513A, was shown to be able to compete successfully for wild mates, furthermore sustained release of OX513A males suppressed a wild population of Ae. aegypti (Harris et al., 2012; Harris et al., 2011). We tested whether this same strain and strategy could also be effective in Brazil. Within the overall project objective of evaluating RIDL technology in Brazil, we had three core technical activities. These were (i) to transfer the technology to Brazil, including adaptation and optimisation for local conditions; (ii) to assess the field performance in terms of mating competitiveness of RIDL males in Brazil; and (iii) to test the ability of RIDL males to suppress a wild Ae. aegypti population in this environment. The fourth core activity, which will be described in detail elsewhere, related to community engagement and regulatory activities.
Methods
Study area The study was conducted in the Itaberaba suburb of the city of Juazeiro, Bahia in the semi‐arid North East of Brazil (latitude – 9.450, longitude ‐ 40.481), both treated and control sites were in the same
CONFIDENTIAL 19 suburb and, consequently, had similar characteristics. The site consisted predominately of housing of relatively low social economic status and was identified by local public health officials as having high dengue incidence. A dependence on stored water (due to irregular services for piped water) and high human densities provided ideal habitats for Ae. aegypti and thus the area supported a relatively high and stable year‐round population that is atypical of fluctuating seasonal populations that are broadly prevalent in the region. This provides a highly challenging environment, terms of mosquito population, in which to evaluate OX513A technology. Baseline monitoring was initiated in July 2010 using ovitraps which revealed the presence of Ae. aegypti and absence of Ae. albopictus in the whole of Itaberaba.
Human population was estimated at 1810 (165 people ha‐1) in the treated area Juazeiro has a semi‐arid climate with average annual precipitation of 536 mm falling mostly in warmer summer months (November‐April). Throughout the study, conventional local mosquito control was deployed as normal and public heath agents followed standard procedures. Teams of public health agents typically visited homes between 4 and 6 times per year, where they destroyed some breeding sites and treated others with the organophosphate larvicide, temephos. The same team of public health agents were responsible for the whole of Itaberaba suburb, ensuring that underlying conventional mosquito control was similar between the treated and untreated areas of this study.
Community Engagement/regulatory Consent and support came from national and regional administration and local community leaders. Prior to establishment of the transgenic OX513A line in the mass rearing facility and subsequent open releases, regulatory approvals were obtained from the appropriate Brazilian national regulatory body, Brazilian National Biosafety Technical Commission (CTNBio), for the import permit from the UK to University of São Paulo (Diário Oficial da União (DOU): Extrato de Parecer (EP) 2.031/2009) and for the containment facility for rearing the strain at Biofábrica Moscamed Brasil (DOU: EP 2.577/2010). Approval for releasing OX513A males in the environment was granted in 2010 by CTNBio for five sites, including Itaberaba, around Juazeiro, Bahia (DOU: EP 2.765/2010).
From its inception the project sought to adopt full transparency with a vigorous and proactive community engagement campaign. Implementation included communication via local media (radio, TV and press), community meetings, printed information (posters and leaflets), school presentations, carnival parades, use of small vans with loudspeakers and social media (websites and blogs). Dedicated door‐to‐door campaigns and ongoing contact with field technicians working in the community provided face‐to‐face interaction on an individual basis, allowing specific questions to be addressed and for direct
CONFIDENTIAL 20 feedback and concerns to be aired. Full description of the community engagement will be reported elsewhere.
Mosquitoes Transgenic Ae. aegypti with the OX513A insertion were used during this study (Phuc et al., 2007). The sensitivity of the OX513A strain to chemical insecticide has been evaluated independent (LSTMH, following WHO protocols). The strain was found to be susceptible to discriminating doses of the insecticide representing all classes commonly used (temephos, permethrin, deltamethrin and malathion,) with the exception of bendiocarb which was found to be non‐discriminating for Ae. aegypti. This was the same strain previously used for field evaluation in the Cayman Islands (Harris et al., 2012; Harris et al., 2011). The breeding line was originally imported by Oxitec Ltd. to the University of São Paulo where it underwent laboratory evaluations against Brazilian Ae. aegypti lines before being transferred to Biofábrica Moscamed Brasil (www.Moscamed.com), Juazeiro City.
Mass rearing of OX513A Subsequent to obtaining regulatory approval for field release, production of male mosquitoes was conducted at the Moscamed facility in an 84m2 laboratory specifically adapted and approved for the purpose. Mass rearing insectaries were maintained at 27°C (+/‐ 2), 70‐90% relative humidity and a 12 hr day/night cycle. A colony of homozygous OX513A was established producing eggs to supply male mosquito production for release. Mosquitoes destined for release were reared to pupae where they were mechanically sorted to remove females (Ansari et al., 1977; Focks, 1980). For quality control a minimum of 1500 male pupae from every release batch were individually checked using a microscope to ensure < 1% female contamination. Residual female presence was 0.02% (95% bootstrap CI: 0.016%‐ 0.031%), equivalent to 1 female for every 4,300 males. Weekly quality control checks were made of the transgenic phenotype i.e. expression of the fluorescent marker and lethality in the absence of tetracycline. Detailed methods for production of male pupae followed (Carvalho et al., 2013).
Eclosion and Release Male pupae were aliquoted into release devices (RD) where they eclosed to adults over 24‐48 hr before release (for details, see SI). Mosquitoes were dispersed in field site by opening RDs at the rear of a vehicle moving slowly throughout the release area. Releases occurred three times week. In the initial rangefinder phase we released a total 185,000 over a 6‐week period starting on 7 May 2011. Thereafter rearing and distribution systems were optimised and sterile male production capacity increased. Release rates increased in line with production allowing extensive estimation of mating competitiveness and
CONFIDENTIAL 21 eventually suppression. Following suppression, releases were maintained at a lower level (ca. 5 times lower) designed to counter resurgence of population (Figure 3).
Monitoring Ovitraps were checked and replaced weekly. Ovitrap index was calculated as number of egg‐positive traps/total number of traps recovered. Direct monitoring of the adult population was conducted initially by aspiration surveys and later with BG‐Sentinel traps (Biogents, Regensburg, Germany). For details see SI.
Statistics Statistical analyses were performed using R freeware (R Core Team, Vienna, Austria). Population size was estimated using Petersen‐Lincoln test (Ito and Yamamura, 2005) as an estimate of the standing crop of released males. The number of pupae per person was determined from the wild mosquito population estimates following the calculations described by Focks et al. (2000). Confidence intervals (95%) for survival and population estimates were calculated by bootstrap (10,000 repeats).
Ethics Statement
All experiments with animal blood were carried out in accordance with the guidelines of the Ethical Principles for Experiment on Animals adopted by Sociedade Brasileira de Ciência de Animais de laboratório (SBCAL) and approved by the Institutional Ethics Review Committee (Comissão de Ética no Uso de Animais – CEUA)‐Universidade de São Paulo, protocol #022/11.
The community engagement protocol was approved by the Institutional Review Board in Human Research (Comissão de Ética em Pesquisa com Seres Humanos do Instituto de Ciências Biomédicas/USP) and Comissão Nacional de Ética em Pesquisa – CONEP, protocol #1115.
Results
Mating competitiveness of the released adult sterile males is crucial to the success of a sterile‐male control program. We investigated the mating competitiveness of OX513A males by releasing approximately 2,800 ha‐1 week‐1 for six weeks in an 11 ha region of Itaberaba (Figure 1, area A and B). Part of the rationale for this design, which we call ‘rangefinder’, was to provide an accurate estimation of wild population, which is otherwise very difficult to ascertain. This, together with mating competitiveness, enables estimation of likely release numbers required to achieve a predetermined
CONFIDENTIAL 22 ratio of released to wild population males in order to achieve suppression. Consequently, we released at a known constant rate.
We then trapped adults and eggs, here by aspiration and ovitraps respectively. We estimated a ratio of 3.7:1 (95% bootstrap CI: 3.19‐4.41) OX513A males per wild male was achieved, based on deviation in sex ratio from the 0.46 males per female found in untreated areas.
By hatching the eggs from ovitraps and scoring the resulting larvae for fluorescence, we could identify which had an OX513A father – and had therefore inherited a copy of the OX513A transgene, which carries a fluorescent marker – and which had not (Figure 2). By comparing the ratio of OX513A:wild males with the fluorescent:non‐fluorescent ratio in hatched eggs (‘fluorescence ratio’) we were able to estimate the mating competitiveness of the released OX513A males. As expected (Harris et al., 2011), following releases there was a lag period before fluorescence was detected and then an increase before stabilising 3‐4 weeks later. We interpret the lag period as representing the time during which the OX513A male population has accumulated to a steady state (new introductions matched by deaths), and similarly for females emerging late enough to have been exposed to these OX513A males as virgins. However this situation is itself transient as death of OX513A heterozygous offspring will in time start to have an impact on the number of wild mosquitoes, depending on the numbers of RIDL males released and their mating competitiveness in relation to wild males. Females additionally have to blood feed and find oviposition sites. We therefore expected a lag phase comprising a time delay of approximately one week between release of OX513A males and the appearance in ovitraps of eggs that they had fathered and a further delay as the population of females mated prior to this point declined to negligible levels. For the same reason, such eggs were expected for a short period after releases ceased. The data are consistent with these hypotheses (Figure 2). Accounting for a time lag before equilibration of released males and emerging females, we assessed for a 6‐week period (19 May‐29 June 2011). In total we screened 9,252 larvae of which 943 (10.2% (95% exact binomial CI: 9.6%‐10.8%)) were fluorescent. Taking this value together with the male release ratio (3.7:1) implies that the released males had a relative competitiveness of 0.031 (95% bootstrap CI: 0.025‐0.036). This is an underestimate due to the potential emigration of OX513A‐mated females and immigration of pre‐mated wild females.
We next developed rearing and distribution systems in Brazil, building on Oxitec’s prior experience from field release in other countries (Harris et al., 2012; Harris et al., 2011; Lacroix et al., 2012). This involved significant modification and optimization of methods to make them appropriate for the Brazilian environment, including local sourcing of material (Carvalho et al., 2013). Adult male production
CONFIDENTIAL 23 correspondingly increased from approximately 30,000 per week during the rangefinder to 540,000 per week in early 2012 (Carvalho et al., 2013).
In addition to refining rearing methods and associated assessment of the field performance of the released mosquitoes, population suppression was a key endpoint of the release program as this would validate the technology in a Brazilian setting. Sterile‐male methods such as the RIDL control strategy will successfully suppress the target population if a sufficient proportion of females mate sterile males. This threshold mating fraction required for local elimination in the absence of immigration can be estimated from models of mosquito population dynamics (Dye, 1984) as 13‐57% (Harris et al., 2011; Phuc et al., 2007). Field experience in Grand Cayman suggested that in that location the threshold mating fraction with OX513A RIDL males was towards the lower end of this range, as significant suppression was observed at an estimated mating fraction of 12% (Harris et al., 2012). While this does not show that the critical mating fraction was less than 12%, as release rates somewhat below the threshold level may still give significant suppression, it does suggest that the threshold is not at the upper end of the model estimates. We therefore aimed to achieve a mating fraction of 50%, reasoning that even if the population dynamics were somewhat different than in Grand Cayman this would likely be sufficient to achieve suppression. From the rangefinder study we predicted that we would have needed a nine‐fold increase in release rate, from 2,800 to 25,000 ha‐1 week‐1, to achieve a target 50% mating fraction, assuming mating competitiveness and wild populations remained the same.
Up to 11th February 2012 we released into areas A and B (Figure 1), comprising 11 ha in total. However, despite improvements in rearing over the period, in this highly infested area we were unable to produce enough OX513A males with the available resources to consistently maintain a mating fraction of 50%, as judged by the percentage of fluorescent larvae. We therefore reduced the release area to an area of 5.5 ha (A, Figure 1). As expected, the fluorescence ratio increased correspondingly.
Actual release numbers, observed mating fraction (% fluorescent larvae), and ovitrap index – a measure of population density – are presented in Figure 3B‐D. The ovitrap index in the untreated area remained relatively stable, showing little seasonal variation. Evaluation of impact of any treatment should be assessed in relation to untreated control areas thereby controlling for fluctuation resulting from local environmental factors such as rainfall and temperature. Ovitrap indices for treated areas A and B are correspondingly presented as relative ovitrap index (ovitrap index in treated area divided by ovitrap index in control area, Figure 3C). In 2011 the ovitrap indices in the treated areas were approximately double that in the untreated area, with a mean relative ovitrap index of 1.91 (95% bootstrap CI: 1.47‐
CONFIDENTIAL 24
2.49) and 2.29 (95% bootstrap CI: 1.89‐2.80) for A and B respectively. As treatment started to take effect, ovitrap indices declined in the treated areas relative to the untreated area, with average relative ovitrap indexes of 0.37 (95% bootstrap Cl: 0.26‐0.51) and 0.50 (95% bootstrap CI: 0.38‐0.62) for area A and B respectively from June 2012 onwards. This represented significant reduction (One‐way ANOVA: df=1, P<0.001) in relative ovitrap indices of 81% (95% bootstrap CI: 74.9%‐85.2%) and 78% (95% bootstrap CI: 73.7%‐82.1%) for areas A and B respectively. Mating competitiveness is a key performance measure for released males; estimates of net field mating competitiveness of the released males ranged between 0.0004‐0.047 (S2). Mean competitiveness as estimated by this method declined substantially with suppression of the wild population (first 5 estimates before suppression mean 0.030, 95% bootstrap CI: 0.020‐0.040; last 5 estimates mean 0.008, 95% bootstrap CI: 0.002‐0.016). These estimates may be influenced by immigration of pre‐mated females; assuming this is constant, as the wild population is suppressed the proportion of positive ovitraps resulting from immigrating pre‐mated females increases. This would result in a reduced estimate of mating competitiveness. The pre‐ suppression mean of 0.030 therefore represents the best estimate of the overall mean mating competitiveness for OX513A in this study. This is consistent with the estimate from the rangefinder (0.031; 95% bootstrap CI: 0.025‐0.036). Although the exact turning point is difficult to pinpoint, it is clear that suppression of the target population, relative to the untreated control area, began at the same time or even just before reducing the area treated (13 Feb 2012). This, combined with ongoing immigration of sterile males from the adjacent Area A, presumably explains that suppression was seen also in Area B. Sterile males do not immediately reduce the target population; rather they lead to mortality in the next generation. Therefore, there is a delay of approximately one generation between the release of sterile males and any consequent effect on population size. Mean fluorescence ratio over the prior 6‐week period (28 December 2011‐31 January 2012) was 43% (95% bootstrap CI: 34.8‐52.1%) and corresponded with a mean release rate of 28,644 ha‐1 week‐1 (95% bootstrap CI: 24,929‐32,102). This corresponds well with the target release rates predicted from the rangefinder. Ovitraps provide only an indirect measure of the adult population density, reduction of which is the key target for vector control purposes. Though we consider that changes in ovitrap metrics provide good indications of changes in adult density within an area, they provide a poor guide to absolute number (Focks et al., 2000). To provide an independent measure of the adult population we used mark‐release‐recapture methods to estimate the standing crop of wild Ae. aegypti adults in the trial area (Silver, 2008a), exploiting the periodic releases of OX513A males. Adult collection was initially based on aspiration surveys; later on BG Sentinel traps, for which calculations were carried out for trap collections spanning consecutive periods of approximately four
CONFIDENTIAL 25 weeks. For each period, the average OX513A male standing crop was calculated from numbers released and an estimated daily survival probability (DSP) of 0.49, this DSP estimate being derived from a series of 14 mark‐release‐recapture studies using dye‐marked cohorts. The wild population density was then estimated based on the relative recapture rates of OX513A and wild males.
These experiments provide clear evidence of a reduction in the standing crop of adults (Figure 3D). We observed a 95% (95% bootstrap CI: 92.2%‐97.5%) reduction in the estimated Ae. aegypti adult standing crop from an average of 418 ha‐1 (95% bootstrap CI: 307‐532), prior to January 2012, to 20 ha‐1 (95% bootstrap CI: 10‐29) in area A from May 2012 onwards. As discussed above, this 95% reduction represents an independent and likely more accurate, assessment of the impact on absolute adult population than ovitrap index, and is the most pertinent measure in relation to impact of intervention with regard to reducing the risk of disease transmission.
Discussion
In this study we demonstrate effective control of a wild population of Ae aegypti by sustained releases of RIDL OX513A male Aedes aegypti. We diminished Ae. aegypti population by 95% (95% CI: 92.2%‐ 97.5%) based on adult trap data and 81% (95% CI: 74.9‐85.2%) based on ovitrap indices compared to the adjacent no‐release control area. We estimated mating competitiveness of the released males to be 0.030 (95% bootstrap CI: 0.020‐0.040).
These results are similar to the 82% (95% bootstrap CI:69.7‐90.0%) suppression achieved by release of OX513A males in a similar study in Grand Cayman (Harris et al., 2012). In both cases, the degree of suppression attainable was expected to be limited by immigration of wild mosquitoes from adjacent untreated areas. Aedes aegypti dispersal is relatively short, with most published mean dispersal distances falling between 30‐100 m as reviewed by Silver et al. (Silver, 2008b). Effects of immigration would therefore likely be limited to a relatively small boundary zone in a larger program, or not even that if the whole of an isolated population were treated. Suppression may also be limited by the persistence of viable eggs laid at an earlier period. These may hatch over a period of months after deposition, depending on environmental conditions. The gradual reduction in the target population from April 2012 onwards may relate to gradual depletion of this egg bank. As for spatial migration, this ‘temporal migration’ would be of very limited consequence for a larger, longer operational control program. The 0.030 mating competiveness observed was also consistent with the estimate of 0.059 (95% bootstrap CI: 0.011 – 0.21) from the Grand Cayman study (Harris et al., 2012). This suggests that
CONFIDENTIAL 26 differences between these two wild populations, or in environmental parameters between the sites, or in experimental procedures such as rearing and distribution, had little effect on the ability of OX513A to win mates. Mating competitiveness as measured by this approach includes any effect of the transgene on the released males, the effect of artificial rearing, handling and distribution, and the effect of migration both of pre‐mated females into the area and of released males and mated females out of the area. Relatively few estimates of mating competitiveness under open‐field conditions have been published, despite the long history of sterile‐male methods. In large‐scale, successful SIT programmes, field competitiveness of sterile males was estimated at 0.1 for New World screwworm (Cochliomya hominivorax) (Mayer et al., 1998; Vreysen, 2005) and <0.01 for Mediterranean fruit fly (Ceratitis capitata) (Rendón et al., 2004; Shelly et al., 2007). Mean fluorescence ratio corresponding with period when suppression was observes was estimated to be 43% (95% bootstrap CI: 34.8‐52.1%). This falls within model predictions of 13‐57% (19) as did the Grand Cayman suppression (12%); variations in that threshold may indicate somewhat different population dynamics, as might be expected between two very different areas.
Conventional sterile insect based programs generally achieve greater efficiency by initiating releases in conjunction with reduced mosquito population, ether by utilising naturally occurring seasonality or temporary knockdown with alternate control such as insecticide application. In this study site there was consistently high mosquito infestation (figure 2) and no targeted controls was used other than that routinely deployed by vector control teams. Despite this we were able to demonstrate efficacy in challenging conditions, with scope for substantial improved efficiency areas with lower mosquito populations and by incorporation with an integrated vector management program.
This study comprised exclusively entomological endpoints. What would be the impact of such striking reduction of vector population density on dengue transmission? Focks et al. estimated a disease transmission threshold in relation to pupae per person‐1 (as a proxy for adult mosquito population), ambient temperature and herd immunity (Focks et al., 2000; Focks et al., 2006). For a mean temperature of 28°C Focks et al. calculated an epidemic transmission threshold of 0.42, 0.61 or 1.27 pupae per person for initial seroprevalence of 0%, 33% and 67%, respectively. The average temperature during peak dengue transmission season (January‐July) was 27.7°C at the Juazeiro field site, and we can assume a moderate to high seroprevelance given the historical high level of dengue incidence reported in the site by residents and public health workers. Using calculations and assumptions given in Focks et al. (2000), we estimate that average pupae person‐1 decreased in our treated area from 0.7 pre‐
CONFIDENTIAL 27 treatment to 0.04 post‐treatment, which in their model would be sufficient to prevent epidemic transmission under these conditions, or indeed under the most adverse conditions modelled for a naive population with 0% seroprevalence. The long‐term goal for vector control should be to suppress below the transmission threshold even given low herd immunity (Ooi et al., 2006) – our data indicate that release of OX513A RIDL males is able to achieve this goal.
Acknowledgements
We are very grateful to Juazeiro municipal government and vector control authorities, and especially to the people of Itaberaba, without whose constant support, participation and encouragement the work described here would not have been possible. We thank Josué Young from Gorgas Institute, Pannama and staff at Moscamed and Oxitec for their assistance especially Jair F. Virginio, José Carlos Valença, Gildeane Silva, Gessilane dos Santos, Fabio Gonçalves, John Paul Oliveira, Derric Nimmo, Jessica Stevenson, and Neil Naish. For regulatory support we thank Camilla Beech and for statistical analysis Pete Winskill.
References
Alphey, L., Benedict, M.Q., Bellini, R., Clark, G.G., Dame, D.A., Service, M.W., and Dobson, S.L. (2010). Sterile‐insect methods for control of mosquito‐borne diseases: an analysis. Vector Borne Zoonotic Dis 10, 295‐311. Alphey, N., Alphey, L., and Bonsall, M.B. (2011). A Model Framework to Estimate Impact and Cost of Genetics‐Based Sterile Insect Methods for Dengue Vector Control. PLoS ONE 6, e25384. Ansari, M., Singh, K., Brooks, G., Malhotra, P., and Vaidyanathan, V. (1977). The development of procedures and techniques for mass rearing of Aedes aegypti. Indian J Med Res 65, (Suppl) 91‐99. Atkinson, M.P., Su, Z., Alphey, N., Alphey, L.S., Coleman, P.G., and Wein, L.M. (2007). Analyzing the control of mosquito‐borne diseases by a dominant lethal genetic system. Proc Natl Acad Sci U S A 104, 9540‐9545. Benedict, M.Q., and Robinson, A.S. (2003). The first releases of transgenic mosquitoes: an argument for the sterile insect technique. Trends Parasitol 19, 349‐355. Bhatt, S., Gething, P.W., Brady, O.J., Messina, J.P., Farlow, A.W., Moyes, C.L., Drake, J.M., Brownstein, J.S., Hoen, A.G., Sankoh, O., et al. (2013). The global distribution and burden of dengue. Nature 496, 504‐507. Carvalho, D.O., Wilke, A.B.B., Nimmo, D.D., Naish, N., McKemey, A.R., Gray, P., Marrelli, M.T., Virginio, J.F., Alphey, L., and Capurro, M.L. (2013). Mass production of RIDL® Aedes aegypti for field releases in Brazil. JOVE in press. Catteruccia, F., Crisanti, A., and Wimmer, E. (2009). Transgenic technologies to induce sterility. Malaria Journal 8, S7.
CONFIDENTIAL 28 de Souza, R.P., Rocco, I.M., Maeda, A.Y., Spenassatto, C., Bisordi, I., Suzuki, A., Silveira, V.R., Silva, S.J.S., Azevedo, R.M., Tolentino, F.M., et al. (2011). Dengue Virus Type 4 Phylogenetics in Brazil 2011: Looking beyond the Veil. PLoS Negl Trop Dis 5, e1439. Dyck, V.A., Hendrichs, J., and Robinson, A.S. (2005). Sterile insect technique: principles and practice in area‐wide integrated pest management (Netherlands, Springer). Dye, C. (1984). Models for the population dynamics of the yellow fever mosquito, Aedes aegypti. Journal of Animal Ecology 53, 247‐268. Focks, D., Brenner, R., Hayes, J., and E, D. (2000). Transmission thresholds for dengue in terms of Aedes aegypti pupae per person with discussion of their utility in source reduction efforts. Am J Trop Med Hyg, 11‐18. Focks, D.A. (1980). An improved separator for separating the developmental stages, sexes and species of mosquitoes. Mosq News 19, 144‐147. Focks, D.A., Alexander, N., and Villegas, E. (2006). Multicountry study of Aedes aegypti pupal productivity survey methodology : findings and recommendations (Geneva, World Health Organization), pp. 48. Harris, A.F., McKemey, A.R., Nimmo, D., Curtis, Z., Black, I., Morgan, S.A., Oviedo, M.N., Lacroix, R., Naish, N., Morrison, N.I., et al. (2012). Successful suppression of a field mosquito population by sustained release of engineered male mosquitoes. Nat Biotech 30, 828‐830. Harris, A.F., Nimmo, D., McKemey, A.R., Kelly, N., Scaife, S., Donnelly, C.A., Beech, C., Petrie, W.D., and Alphey, L. (2011). Field performance of engineered male mosquitoes. Nat Biotechnol 29, 1034‐1037. Ito, Y., and Yamamura, K. (2005). Role of population and behavioural ecology in the Sterile Insect Technique. In Sterile Insect Technique; Principles and Practice in Area‐Wide Integrated Pest Management, V.A. Dyck, J. Hendrichs, and A.S. Robinson, eds. (Vienna, Springer), pp. 177‐208. Klassen, W., and Curtis, C.F. (2005). History of the sterile insect technique. In Sterile Insect Technique Principles and practice in area‐wide integrated pest management, V.A. Dyck, J. Hendrichs, and A.S. Robinson, eds. (The Netherlands, Springer), pp. 3‐36. Knipling, E. (1955). Possibilities of insect control or eradication through use of sexually sterile males. J Econ Entomol 48, 459‐462. Lacroix, R., McKemey, A.R., Norzahira, R., Lim, K.W., Wong, H.M., Teoh, G.N., Siti Rahidah, A.A., Sawaluddin, S., Selvi, S., Oreenaiza, N., et al. (2012). Open Field Release of Genetically Engineered Sterile Male Aedes aegypti in Malaysia. PLoS ONE 7, e42771. Lee, K.‐S., Lo, S., Tan, S.S.‐Y., Chua, R., Tan, L.‐K., Xu, H., and Ng, L.‐C. (2012). Dengue virus surveillance in Singapore reveals high viral diversity through multiple introductions and in situ evolution. Infection, Genetics and Evolution 12, 77‐85. Lowe, R.E., Bailey, D.L., Dame, D.A., Savage, K.E., and Kaiser, P.E. (1980). Efficiency of techniques for the mass release of sterile male Anopheles albimanus Wiedemann in El Salvador. American Journal of Tropical Medicine and Hygiene 29, 695‐703. Luz, P.M., Grinsztejn, B., and Galvani, A.P. (2009). Disability adjusted life years lost to dengue in Brazil. Tropical Medicine & International Health 14, 237‐246. Mayer, D.G., Atzeni, M.G., Stuart, M.A., Anaman, K.A., and Butler, D.G. (1998). Mating competitiveness of irradiated flies for screwworm fly eradication campaigns. Preventive Vet Med 36, 1‐9. Ooi, E., Goh, K., and Gubler, D. (2006). Dengue prevention and 35 years of vector control in Singapore. Emerging Infectious Diseases 12, 887‐893. Phuc, H.K., Andreasen, M.H., Burton, R.S., Vass, C., Epton, M.J., Pape, G., Fu, G., Condon, K.C., Scaife, S., Donnelly, C.A., et al. (2007). Late‐acting dominant lethal genetic systems and mosquito control. BMC Biol 5, 11.
CONFIDENTIAL 29
Rendón, P., McInnis, D., Lance, D., and Stewart, J. (2004). Medfly (Diptera:Tephritidae) genetic sexing: large‐scale field comparison of males‐only and bisexual sterile fly releases in Guatemala. J Econ Entomol 97, 1547‐1553. Rodriguez‐Barraquer, I., Cordeiro, M.T., Braga, C., de Souza, W.V., Marques, E.T., and Cummings, D.A.T. (2011). From Re‐Emergence to Hyperendemicity: The Natural History of the Dengue Epidemic in Brazil. PLoS Negl Trop Dis 5, e935. Shelly, T.E., McInnis, D.O., Rodd, C., Edu, J., and Pahio, E. (2007). Sterile Insect Technique and Mediterranean Fruit Fly (Diptera: Tephritidae): Assessing the Utility of Aromatherapy in a Hawaiian Coffee Field. J Econ Entomol 100, 273‐282. Shepard, D.S., Coudeville, L., Halasa, Y.A., Zambrano, B., and Dayan, G.H. (2011). Economic Impact of Dengue Illness in the Americas. The American Journal of Tropical Medicine and Hygiene 84, 200‐207. Silver, J.B. (2008a). Estimating the size of the Adult Population. In Mosquito Ecology, J.B. Silver, ed., pp. 1273‐1376. Silver, J.B. (2008b). Measuring Adult Dispersal. In Mosquito Ecology, J.B. Silver, ed., pp. 1377‐1424. Thomas, D.D., Donnelly, C.A., Wood, R.J., and Alphey, L.S. (2000). Insect population control using a dominant, repressible, lethal genetic system. Science 287, 2474‐2476. Vreysen, M.J.B. (2005). Monitoring sterile and wild insects in area‐wide integrated pest management programmes. In Sterile Insect Technique Principles and practice in area‐wide integrated pest management, V.A. Dyck, J. Hendrichs, and A.S. Robinson, eds. (the Netherlands, Springer), pp. 325‐361. WHO‐TDR (2009). Dengue: guidelines for diagnosis, treatment, prevention and control (Geneva, WHO). Wilder‐Smith, A., Foo, W., Earnest, A., Sremulanathan, S., and Paton, N.I. (2004). Seroepidemiology of dengue in the adult population of Singapore. Tropical Medicine & International Health 9, 305‐308.
CONFIDENTIAL 30
Figure 1. Itaberaba study area. Untreated control area and treatment areas (A and B). Ovitrap distribution is shown for the period 21/11/2011 – 19/9/2012; open circles = 1 trap house‐1, solid circles = 2 traps house‐1. Adult BG Sentinel trap distribution for the period 10/7‐25/9/2012 is also shown (open diamonds). Control area = 43.0 Ha, Area A= 5.5 ha, Area B= 5.5 ha. Satellite Image ©CNES 2013, distribution Astrium Services/ Spot Image.
CONFIDENTIAL 31
Figure 3. Field data. (A) Weather Data. Temperature and weekly rainfall. (B) Releases and fluorescence. Weekly numbers of adult males released per hectare and percentage of larvae recovered from ovitraps in treated areas with the OX513A transgene as detected by fluorescence. (C) Relative ovitrap index. Relative Ovitrap Index (Treated/Control) in treated areas A and B. The horizontal line represents an equal value of ovitrap index in treated areas and control area. (D) Adult density in treated area and Ovitrap index in control area. Estimated adult wild population of Ae. aegypti per hectares (error bars = 95% CI) in treated area and ovitrap index in control area. Overall. There is a clear decrease in relative ovitrap index and estimated wild male population from March 2012 to September 2012 following the increase in fluorescence induced by the increased releases while the ovitrap index in the control area remains relatively stable.
CONFIDENTIAL 33
CONFIDENTIAL 34
Supporting Information (submitted in separate file)
Materials and Methods Eclosion and Release Male pupae were aliquoted into combined pupal eclosion/adult release devices (RD). Each RD consisted of a 1.8 litre (14cm high x 13 cm diameter) clear plastic cylindrical container (Produtos Prafesta®, Brazil) with a large hole in the lid that was covered with fine mesh to allow for air exchange and sugar feeding provided by cotton wool soaked in 10% sucrose solution. For the 16 releases of the ‘rangefinder study’ ~500 pupae were used per RD, thereafter ~1000 pupae were added to each RD. Once adults had eclosed, any remaining water was drained from pots through a slit in the side of the container before release, which took place approximately 48‐72 hrs after pupation. Releases occurred 3 times per week. RD’s were transported in large insulated boxes by truck to the release site and opened according to a scheduled release plan. Total numbers of adults released were estimated by subtracting the numbers of dead adults and pupae remaining in RD’s after release, from the initial numbers of pupae added.
Monitoring A grid of ovitraps spanning treated and untreated control areas was used throughout the study to provide an indirect measure of adult abundance (Figure 1). Numbers of traps increased in accordance with the increasing study scale (S1). From 11 May 2011 onward, larvae were scored for the presence of the OX513A gene based on the characteristic red fluorescence phenotype due to the DsRed2 marker (1, 2) (Clontech Laboratories Inc.) using a Leica MZ10F epi‐fluorescence microscope (Leica, Wetzlar, Germany). Throughout the release period, non‐fluorescent larvae were reared to adults as an additional check for the presence of Ae. albopictus; none were detected. Ovitrap index (number of egg‐positive traps/total number of traps recovered) based on the number of egg positive traps, rather than larvae identified as Ae. aegypti, was used as a representative statistic. This avoids the possibility of inconsistencies in calculated ovitrap index due to variation in egg hatch. Ovitraps are designed to mimic natural oviposition sites (3) and consisted of black plastic pots (11.5 x 11 cm) three‐quarters filled with clean water with a fibre board paddle (12 x 3cm) protruding above the water line to provide an oviposition substrate. Traps were checked and replaced weekly.
Direct monitoring of the adult population was conducted initially by aspiration surveys and later with BG‐Sentinel traps (Biogents, Regensburg, Germany). Aspiration surveys were conducted using locally made hand‐held battery powered aspirators (similar to InsectaZooka www.BioQuip.com). After obtaining consent from the respective property owner, each building was sampled for a set period of 15
CONFIDENTIAL 35 minutes. Aspiration surveys are very labour intensive and intrusive to the local population and were therefore only conducted at periods coinciding with releases of marked cohorts of mosquitoes. For these reasons BG‐Sentinels were later introduced as an alternative adult sampling method. From March 2012 BG‐Sentinel traps were installed permanently providing continuous adult monitoring in the treated area. Traps were serviced daily in the course of some mark‐release‐recapture experiments, otherwise weekly. As trap catches are cumulative over time, it is possible to use the mean catch per day statistic for comparative purposes where different service intervals were used.
References 1. Lukyanov KA, et al. (2000) Natural animal coloration can be determined by a nonfluorescent green fluorescent protein homolog. J Biol Chem 275(34):25879‐25882. 2. Matz M, et al. (1999) Fluorescent proteins from nonbioluminescent Anthozoa species. Nat Biotechnol 17(10):969‐973. 3. Silver JB (2008) Mosquito Ecology ‐ Field Sampling Methods Third Ed p 1477.
Table 1 (S1). Ovitrap deployment plan in A, B and Control area
Trapping Area Period A B Control 22/02/2010 ‐ Houses 7 5 9 14/02/2011 Traps 7 5 9 21/02/2011‐ Houses 7 5 28 11/05/2011 Traps 7 5 28 18/05/2011 ‐ Houses 30 30 40 13/07/2011 Traps 60 60 40 20/07/2011 ‐ Houses 30 30 69 14/12/2011 Traps 60 60 69 21/12/2011 ‐ Houses 30 30 69 19/09/2012 Traps 60 60 100
CONFIDENTIAL 36
Table 2 (S2). Mating competitiveness and wild population estimation. Details of the calculations of mating competitiveness and wild population of Ae. aegypti over the release period. φ Sex ratio was different for aspiration (0.45) and for BG‐Sentinel traps (0.69).
CONFIDENTIAL 37
Date 31/05/2011 29/07/2011 31/08/2011 23/09/2011 11/11/2011 21/01/2012 14/04/2012 14/05/2012 11/06/2012 13/07/2012 12/08/2012 08/09/2012
Study Period 1 2 3 4 5 6 7 8 9 10 11 12
Suppression Before Before Before Before Before After After After After After
A +B A +B A +B A +B A +B A +B A A A A A A Release area (ha)
11ha 11ha 11ha 11ha 11ha 11ha 5.5ha 5.5ha 5.5ha 5.5ha 5.5ha 5.5ha
Adult Trapping Method: AS = AS AS AS AS BG BG BG BG BG BG BG BG Aspiration, BG = BG Sentinel
Total male 893 918 171 258 2363 6020 11368 10402 11506 7653 1889 5280
Total Female 417 85 55 37 268 174 42 22 5 66 22 63
2301.2 Sex Ratio / 2.14 10.80 3.11 6.97 8.82 34.60 270.67 472.82 0 115.95 85.86 83.81
Sex Ratioφ (M/F) in untreated area 0.45 0.45 0.45 0.45 0.69 0.69 0.69 0.69 0.69 0.69 0.69 0.69
Estimated # wild * male 189 39 25 17 185 120 29 15 3 46 15 43
Estimated # OX513A ‐ 704 879 146 241 2178 5900 11339 10387 11503 7607 1874 5237
3335.7 Over‐flooding ratio O / 3.72 22.80 5.85 14.37 11.78 49.17 391.47 684.59 4 167.13 123.50 120.52
OX513A larvae 943 1533 719 874 2314 2611 400 566 76 143 80 123
Wild larvae 8309 5283 3354 2432 4171 4169 398 135 60 142 103 45
Proportion OX513A P larvae / 0.10 0.22 0.18 0.26 0.36 0.39 0.50 0.81 0.56 0.50 0.44 0.73
C
P* / 1‐
P 0.031 0.013 0.037 0.025 0.047 0.013 0.003 0.006 0.0004 0.006 0.006 0.023 Mating Competitiveness Bootstrap 95%CI lower
limit 0.0254 0.0089 0.0223 0.0138 0.0399 0.0104 0.0016 0.0031 0 0.0039 0.0031 0.0139
Bootstrap 95%CI upper
limit 0.0361 0.0174 0.0546 0.0391 0.0549 0.0152 0.0036 0.0097 0.0008 0.0085 0.0104 0.0352
365 2611 494 2714 2557 5966 10421 6925 6604 2360 625 1081 Mean OX513A male
CONFIDENTIAL 38
standing crop/ha
Male wild standing /O crop/ha 98 114 84 189 217 121 27 10 2 14 5 9
Female wild standing / crop/ha 216 252 186 416 315 176 39 15 3 20 7 13
315 367 270 605 531 297 65 25 5 35 12 22 Wild adult Bootstrap 95%CI lower Ae aegypti /ha limit 265 263 163 351 455 244 43 13 0 25 7 15
Bootstrap 5%CI
upper limit 367 502 407 933 614 351 90 38 10 46 19 28
CONFIDENTIAL 39
3A Quality Control Protocol for Assessment of Mating Co...
SECTION 5. ORGANISATION, STANDARDS AND ADMINISTRATION
1 PURPOSE This procedure will serve as the standard method for assessing the mating competitiveness of the Aedes aegypti OX513A strain as a part of routine quality control testing. It will be used to analyse the ability of OX513A homozygous males to compete for mates with a WT strains reared under the same conditions and of similar size. The purpose of this procedure is, by comparing data from this assay to a baseline characterisation, to assess any changes in mating competitiveness of the OX513A strain that would warrant further investigation. 2 SCOPE Quality control of Aedes aegypti OX513A strain throughout Oxitec’s operations worldwide. 3 ABBREVIATIONS AND RESPONSIBILITIES QC Quality Control N/A QCO Quality Control Operator Responsible for all stages of this procedure. QCL Quality Control Leader Responsible for verifying satisfactory completion of the process and approving of associated report. WT Wild Type N/A 4 EQUIPMENT AND MATERIALS ‐ Latin Wild‐type (WT) and OX513A Aedes aegypti strains ‐ Mosquito consumables including hatch pots, plastic transfer pipettes, ground mosquito diet, universals, sugar solution (SOP OX513A Adult Management), doxycycline hyclate, defibrinated blood, seed germination cages/paper, Parafilm™, glass Pasteur pipettes, masking tape, 100 mL weigh boats, petri dishes, 7 mL weigh boats, Fry fish food. ‐ Reusable equipment including vacuum hatcher, 15 cm X 15 cm X 15 cm cages, 30 cm X 30 cm X 30 cm cages, plate feeders, egg collection pots, microscope, with camera attachment and graticule. ‐ MilliQ Water System 5 RELATED DOCUMENTS OX513A Adult Management QC Sample Receipt, Processing, Scheduling and Reporting OX513A Insectary Conditions Specification OX513A Mating Competitiveness Data Recording Sheet Standard Rearing Protocol for Aedes aegypti Mosquitoes 6 REFERENCES Baseline data in Internal Research Report PH‐2014‐3
Harris, A. F., A. R. McKemey, et al. (2012). "Successful suppression of a field mosquito population by sustained release of engineered male mosquitoes." Nat Biotechnol 30(9): 828‐830.
Harris, A. F., D. Nimmo, et al. (2011). "Field performance of engineered male mosquitoes." Nat Biotechnol 29(11): 1034‐1037.
Khongtak, P., S. Vasan, et al. (2009). Mating competitiveness in lab cages under total containment using RIDL OX513A genetically sterile Aedes aegypti mosquitoes for research purposes. I. Report, World Health Organisation Collaborating Centre for Diagnostic Reference, Training and investigation of Emerging Infectious Diseases, Armed Forces Research Institue of Medical Sciences (AFRIMS), Bangkok, Thailand.
Lee, H. L., S. Vasan, et al. (2013). "Mating compatibility and competitiveness of transgenic and wild type Aedes aegypti (L.) under contained semi‐field conditions." Transgenic Res 22(1): 47‐57.
Phuc, H. K., M. H. Andreasen, et al. (2007). "Late‐acting dominant lethal genetic systems and mosquito control." BMC Biol 5: 11.
Oxitec Ltd CONFIDENTIAL Page 12 of 13
OX513A Field Penetrance Assay Effective Date 12MAR15
SECTION 1. INSTRUCTIONS AND RECORD UNIT: COHORT:
HEALTH AND SAFETY PRECAUTIONS: Observe all relevant Health and Safety precautions related to the workplace and its operations. - Wear Eye Protection when working with water used for insect‐rearing. - Read relevant MSDSs before handling chemicals/materials. - Take care not to slip if floors become wet. - Observe the guidance in HS/G/00048, Working in High Temperature Environments ENVIRONMENTAL PROTECTION: strictly observe containment precautions when working with ox513a.
1 EQUIPMENT AND PREPARATION
1.1 UTO Confirm that the eggs have been hatched in compliance with: - , OX513A Hatching Larvae from Ovitrapped Eggs
1.2 UTO For the following items, use brand new equipment for each batch - Paintbrushes (1 per operator) - Rearing pots - Eclosion containers
1.3 UTO Prepare the water level measure stick - Measure 4 x 40 mL and pour in 4 rearing pots o Use the stick in the first rearing pot and mark the water level with permanent pen (100%) o Check the level is accurate in the 3 other rearing pots o Repeat until the level is accurate in all rearing pots - Measure 4 x 30 mL and pour in 4 rearing pots o Use the stick in the first rearing pot and mark the water level with permanent pen (75% level) o Check the level is accurate in the 3 other rearing pots o Repeat until the level is accurate in all rearing pots
1.4 UTO Prepare 20 black rearing pots - Fill with mL of distilled water. - Label rearing pot with description (10 for OX513A heterozygotes and 10 for wild) and screening date.
OX513A penetrance study – Rearing pot Larvae phenotype: nnnnnn Cohort: nnnnnn Screening date: dd mmm yyyy Pot #: nnnnnn
1.5 UTO Confirm that the processes described above have been conducted
Oxitec Ltd CONFIDENTIAL Page 1 of 8 Printed: 15 Dec 2016 OX513A Field Penetrance Assay Effective Date 12MAR15
SECTION 1. INSTRUCTIONS AND RECORD UNIT: COHORT:
2 SORTING AND PROCESSING SCREENED LARVAE
2.1 UTO For each hatching container with larvae, label a weigh boat with the trap identifier.
2.2 UTO Pipette individual live larvae in drops of water in a grid pattern on corresponding weigh boat with a pasteur pipette.
2.3 UTO Screen larvae from each traps for fluorescence using a fluorescent microscope. Using the paintbrush, gently transfer each larvae to the corresponding rearing pot while screening: - Fluorescent larvae in OX513A rearing pots per rearing pot) - Non‐fluorescent larvae in Wild rearing pots per rearing pot) Set up as many OX513A rearing pots as possible and set up an identical amount of Wild rearing pots. If there are not enough wild larvae, use traps from the control site and follow process from 2.1.
2.4 UTO If the last OX513A rearing pot does not have enough larvae - Transfer the fluorescent larvae to labelled sample tubes. - Transfer remaining wild larvae to labelled container for species screening and follow , OX513A Species Screening of OviTrapped Larvae.
Comments (if required):
2.5 UTO Place all rearing pots in the incubator (27˚C ± 4˚C).
2.6 UTO Record for each rearing pot the cohort, pot number and genotype of the larvae (OX513A or wild) in , Penetrance sheet.
2.7 UTO Record number of rearing pots setup. (a)
2.8 UTO From the most recent preceding cohort, report the total (b) number of rearing pots already setup (2.9, (c)).
2.9 UTO Total rearing pots setup: (c) ܿൌܾܽ
2.10 UTL Are there enough rearing pots setup to complete the Yes/No study? sign and date - Yes: (c) > rearing pots (no more bacth setup) - No: (c) < rearing pots (set more next week)
Oxitec Ltd CONFIDENTIAL Page 2 of 8 Printed: 15 Dec 2016
OX513A Field Penetrance Assay Effective Date 12MAR15
SECTION 1. INSTRUCTIONS AND RECORD UNIT: COHORT:
4 REARING POTS EXAMINATION
4.1 UTO Examine rearing pots daily.
4.2 UTO Check all rearing pots for water level using measure stick: - if any one rearing pot is below %, refill all rearing pots to 100% with Distilled water (report in comments below).
Comments (if required):
4.3 UTO Dead individuals: - Collect each dead individual (larvae or pupae) separately into a 1.5ml microcentrifuge tube containing 0.5 ml of 70% ethanol. - Label each tube with the cohort description, date and mortality code (Table 2). - Store at ≤ ‐15°C. - Record the numbers of dead larvae and dead pupae for each cohort in Penetrance sheet.
4.4 Live individuals: - Transfer all live pupae from each rearing pot to the eclosion container (weigh boat or similar) assigned to that rearing pot. - Place and maintain the weigh boat into the adult cage assigned to that cohort. o Label each cage with the cohort description, setup date, pot number and completion date (14 days after setup date)
OX513A penetrance study – Adult cage Larvae phenotype: nnnnnn Cohort: nnnnn Set up date: dd mmm yyyy Pot #: nnnnn Completion date: dd mmm yyyy
- Record number of live pupae picked up each day for each cohort in , Penetrance sheet.
Comments (if required):
Oxitec Ltd CONFIDENTIAL Page 4 of 8 Printed: 15 Dec 2016 OX513A Field Penetrance Assay Effective Date 12MAR15
SECTION 1. INSTRUCTIONS AND RECORD UNIT: COHORT:
5 ECLOSION POTS EXAMINATION
5.1 UTO Examine eclosion pots daily.
5.2 UTO Dead individuals - Collect each dead individual (pupae or adults on water) separately into a microcentrifuge tube containing 0.5 ml 70% ethanol. - Label each tube with the cohort description, date and mortality code (Table 2). - Store at ≤ ‐15°C. - Record the numbers of dead pupae and dead adults on water for each cohort in , Penetrance sheet.
Comments (if required):
6 ADULT CAGES EXAMINATION
6.1 UTO Examine adult cages daily.
6.2 UTO Dead individuals - Collect each dead adult in the adult cage separately into a microcentrifuge tube containing 0.5 ml 70% ethanol. - Label each tube with the cohort description, date and mortality code (Table 2). - Store at ≤ ‐15°C. - Record the numbers of dead adults in the cage for each cohort in , Penetrance sheet.
Comments (if required):
Oxitec Ltd CONFIDENTIAL Page 5 of 8 Printed: 15 Dec 2016 OX513A Field Penetrance Assay Effective Date 12MAR15
SECTION 1. INSTRUCTIONS AND RECORD UNIT: COHORT:
7 TERMINATION OF THE STUDY
7.1 UTO Rearing pots - Collect remaining dead/live larvae/pupae individually in microcentrifuge tubes containing 70% ethanol. - Label each tube with the cohort description, date and mortality code (Table 2). - Store at ≤ ‐15°C. - Record the numbers of dead/live pupae and live larvae for each cohort in , Penetrance sheet.
7.2 UTO Eclosion containers - Collect each dead/live individual (pupae or adults on water) separately into a microcentrifuge tube containing 0.5 ml 70% ethanol. - Label each tube with the cohort description, date and mortality code (Table 2). - Store at ≤ ‐15°C. - Record the numbers of dead/live pupae and dead adults on water for each cohort in , Penetrance sheet.
7.3 Adult Cages - Collect each dead adults on the bottom of the cage separately in microcentrifuge tubes containing 70% ethanol. - Label each tube with the cohort description, date and mortality code (Table 2). - Store at ≤ ‐15°C. - Record the numbers of dead adults on the bottom of the cage, - Record and collect non‐functional adults (shake the cage to make adults fly, adults falling on the bottom of the cage and unable to maintain flight are considered non‐functional) for each cohort in , Penetrance sheet. - Freeze cage at ≤ ‐15°C for > 12 hours. - Collect remaining adults (functional adults) individually in microcentrifuge tubes containing 70% ethanol. - Label each tube with the cohort description, date and mortality code (Table 2). - Store at ≤ ‐15°C. - Record the numbers of functional adults for each cohort in , Penetrance sheet.
Comments (if required):
Oxitec Ltd CONFIDENTIAL Page 6 of 8 Printed: 15 Dec 2016 OX513A Field Penetrance Assay Effective Date 12MAR15
SECTION 1. INSTRUCTIONS AND RECORD UNIT: COHORT:
8 MORTALITY CATEGORIES AND CODES
Table 2. Recording categories and codes for mortality from larval rearing/eclosion pots and adult cages
Category Code Description Dead larvae L Larvae that die Dead Pupae P Pupae that die in rearing pot and eclosion container, including those partially eclosed Dead adults on water W Fully eclosed adults that have died on the surface of the water in the eclosion container. Dead adults in cage A Fully eclosed adults that have left eclosion container and subsequently died. Live larvae at end of study (D14 K Live larvae at end of study (D14 after first pupation) after first pupation) Live Pupae at end of study (D14 H Live Pupae at end of study (D14 after first pupation) after first pupation) Non‐functional adults N Fully eclosed live adults that are unable to maintain flight Functional adults F Fully eclosed live adults able to maintain flight
9 PROCESS COMPLETION
9.1 UTO Confirm that the process has been performed in compliance with this procedure and that the record is complete and accurate.
9.2 UTO Send the samples to UK following , OX513A Policy on storage and disposal of insect samples. Proceed to OX513A Quality Control Protocol for Colony Genotyping‐QMS for genotyping of samples.
Comments (if required):
9.3 UTL Verify that the process has been performed in compliance with this procedure and that the record is complete and accurate.
Oxitec Ltd CONFIDENTIAL Page 7 of 8 Printed: 15 Dec 2016
Oxitec Ltd Genomic DNA extractions using Purelink Genomic DNA Kit from Invitrogen
Document No. Revision Date Effective Copy Number New MASTER Owner Sign on MASTER: Approver Sign on MASTER:
Issued to Technical Register
1.0 Purpose To describe the procedure for extracting genomic DNA from insects. Multiple insects can be processed on one column if required. 2.0 Scope This procedure applies to extracting genomic DNA using the Invitrogen Purelink Genomic DNA purification Kit only. 3.0 References / Associated Documents Invitrogen Purelink Genomic DNA purification Kit manual
4.0 Responsibilities This procedure will be carried out by trained staff only. The owner of this document is responsible for maintaining and revising this document. 5.0 Equipment Invitrogen Purelink Genomic DNA purification Kit, sterile pestles, microfuge tubes, pipettes, pipette tips, nitrile gloves, water bath, microfuge, Virkon. 6.0 Precautions Buffer Lysis / Binding solution contains Guanidine hydrochloride which is Harmful if swallowed and Irritating to the eyes and skin wear gloves and safety glasses.
Make sure there is no precipitate visible in PureLink™ Genomic Digestion Buffer or PureLink™ Genomic Lysis/Binding Buffer. If any precipitate is visible in the buffers, warm the buffers at 37°C for 3-5 minutes and mix well to dissolve the precipitate before use. 7.0 Instructions •
7.1 Add % ethanol to PureLink™ Genomic Wash Buffer 1 and PureLink™ Genomic Wash Buffer 2 according to Instructions on each label. Mix well. Mark on the labels that ethanol is added. Store both wash buffers with ethanol at room temperature.
7.2 Set a water bath or heat block at °C.
7.3 Place insect into microfuge tube. Add 180 μl PureLink™ Genomic Digestion Buffer and 20 μl Proteinase K to the tube. break the insect up with a sterile pestle. After use, put the pestles in a beaker of Virkon for at least hours before washing and autoclaving. Ensure the tissue is completely immersed in the buffer mix.
7.4 Incubate at C with occasional vortexing until lysis is complete ( hours). You may perform overnight digestion.
7.5 To remove any particulate materials, centrifuge the lysate at maximum speed for minutes at room temperature. Transfer supernatant to a new microcentrifuge tube. Page 1 of 3 15/12/2016 Genomic DNA extractions using Purelink Genomic DNA Kit fro... Oxitec Ltd Genomic DNA extractions using Purelink Genomic DNA Kit from Invitrogen
7.6 Add μl RNase A to lysate, mix well by brief vortexing, and incubate at room temperature for minutes.
7.7 Add μl PureLink™ Genomic Lysis/Binding Buffer and mix well by vortexing to yield a homogenous solution.
7.8 Add μl % ethanol to the lysate. Mix well by vortexing to yield a homogenous solution. If processing a large number of samples the Lysis/binding buffer and 100% Ethanol can be mixed before adding.
7.9 Remove a PureLink™ Spin Column in a Collection Tube from the kit. Add the lysate ( μl) prepared with PureLink™ Genomic Lysis/Binding Buffer and ethanol to the spin column.
7.10 Centrifuge the column at × g for minute at room temperature. Discard the collection tube and place the spin column into a clean PureLink™ Collection Tube supplied with the kit.
7.11 Add μl Wash Buffer 1 prepared with ethanol to the column. Centrifuge column at × g for minute at room temperature. Discard the collection tube and place the spin column into a clean PureLink™ collection tube supplied with the kit.
7.12 Add μl Wash Buffer 2 prepared with ethanol to the column. Centrifuge the column at maximum speed for minutes at room temperature. Discard flow through and respin for a further minute at x g.
7.13 Place the spin column in a sterile 1.5-ml microcentrifuge tube. Add μl of PureLink™ Genomic Elution Buffer to the column. Choose the suitable elution volume for your needs.
7.14 Incubate at room temperature for minute. Centrifuge the column at maximum speed for minute at room temperature.
7.15 To recover more DNA, perform a second elution step using the same elution buffer volume as first elution. Centrifuge the column at maximum speed for minutes at room temperature.
7.16 Remove and discard the column. Use DNA for the desired downstream application or store the purified DNA at ºC (short-term) or °C (long-term).
7.17 Record all details in lab book.
8.0 Records / Attachments Lab book of person carrying out the procedure.
9.0 Related Documents
10.0 Revision History Page Nature of Revision
Page 2 of 3 15/12/2016 Genomic DNA extractions using Purelink Genomic DNA Kit fro... Oxitec Ltd Genomic DNA extractions using Purelink Genomic DNA Kit from Invitrogen
Training record Each person should sign and date that they have read the SOP attached.
Name Signature Date signed
Page 3 of 3 15/12/2016 Genomic DNA extractions using Purelink Genomic DNA Kit fro... OX513A Sex Sorting of Pupae for Release Effective Date 28JA...
SECTION 1. INSTRUCTIONS AND RECORD UNIT: BATCH:
HEALTH AND SAFETY PRECAUTIONS: Observe all relevant Health and Safety precautions related to the workplace and its operations. - Wear Eye Protection when working with water used for insect‐rearing. - Take care not to slip if floors become wet. - Observe the guidance in , Working in High Temperature Environments ENVIRONMENTAL PROTECTION: Strictly observe containment requirements when working with OX513A.
1 SETTING UP THE WIRE SORTER
1.1 UTO Confirm that the pupae have been sorted from larvae in compliance with: , OX513A Larvae Pupae Sorting ‐ For Release.
1.2 UTO Prepare the wire sorter; check the wires are in good condition and that the box seal is intact. Fill the Sorting Box with water to a depth of approximately cm. This photograph shows the wire sorter.
1.3 UTO Ensure that the Wire Sorter is set so that males pass through and females are retained. - Using a Dissection Microscope manually sex sor female pupae - Add them to the Wire Sorter box - Lock the Wire Sorter box - Reduce the gap of the Wire Sorter so that no pupae may pass through - Submerge the Wire Sorter in the water‐filled Sorting Box - Gradually increase the gap until the first female pupa passes through - Decrease the gap by turning the adjusting knob five turns - Wash the Wire Sorter removing the female pupae for disposal as waste
Oxitec Ltd CONFIDENTIAL Page 1 of 7 Printed: 15 Dec 2016
OX513A Sex Sorting of Pupae for Release Effective Date 28JA...
SECTION 1. INSTRUCTIONS AND RECORD UNIT: BATCH:
5.4 UTO Assess whether the sorting criteria with respect to “Males in the Female Pupae” stated in OX513A Sex Sorting Criteria ‐ For Release is met. - If Fail, consult UTL. - If Pass, continue to next step. - Record Pass or Fail in the table.
6 PROCESS COMPLETION
6.1 UTO Confirm that the process has been performed in compliance with this procedure and that the record is complete and accurate.
6.2 UTO Proceed to , OX513A Male Eclosion in Release Device.
Comments (if required):
6.3 UTL Verify that the process has been performed in compliance with this procedure and that the record is complete and accurate.
Oxitec Ltd CONFIDENTIAL Page 5 of 7 Printed: 15 Dec 2016 OX513A Sex Sorting of Pupae for Release Effective Date 28JA...
SECTION 2. PUPA SEXING GUIDE
Oxitec Ltd CONFIDENTIAL Page 6 of 7 Printed: 15 Dec 2016
OX513A Emergency Response Procedures. Effective Date 18JUL1...
SECTION 1. INSTRUCTIONS
1 EVENTS DURING BUSINESS HOURS
1.1 The Principal Investigator (PI) should be immediately alerted (investigator name and contact information are posted in the lab)
1.2 Follow any directions from the emergency services.
2 EVENTS OUTSIDE OF BUSINESS HOURS
2.1 If an event is discovered outside of business hours, contact the key holder cascade as described in the biosafety manual, and then proceed as for events within business hours.
2.2 Follow any directions from the emergency services.
3 IN THE EVENT OF FIRE
3.1 In the event of fire follow the FKMCD fire procedure for individual safety.
4 IN THE EVENT OF SEVERE WEATHER/NATURAL DISASTERS (HURRICANE, FLOOD, ETC.)
4.1 The insectaries area housed in a building that is designated as a Category 4 hurricane protected zone.
4.2 Depending on the severity of the event the PI will take a decision regarding the safe devitalisation of adult and other life stages of the regulated arthropod. The decision will be documented and communicated to the Sponsor. This is done in accordance with procedure Hurricane Preparedness Policy.
4.3 Eggs of the regulated arthropod will be stored in the rearing facility which is within a category 4 hurricane rated building. This facility also has generator backup for several days.
4.4 Follow the FKMCD procedures for individual safety.
4.5 Follow Hurricane Preparedness Policy for arthropod containment.
5 IN THE EVENT OF POWER OUTAGES LONGER THAN A DAY (BACKUP GENERATORS AVAILABLE FOR SEVERAL DAYS OF OPERATIONAL USE OF THE FACILITY)
5.1 Do not open fridges or freezers.
5.2 After a power outage, keep any waste in the freezer(s) for a minimum of 24 hours after power is restored
5.3 Keep use of electrical items to a minimum.
6 IN THE EVENT THAT THE BUILDING CONTAINMENT IS BREACHED BY MALICIOUS MEANS
6.1 Local security and the local police (via 911) should be immediately alerted by the PI, FKMCD director/senior manager or appointed deputy.
6.2 Keep remaining doors and windows closed/ locked, where possible.
6.3 Staff should stay inside the building. Staff will be alerted as to when they may leave by a member of
Oxitec Ltd CONFIDENTIAL Page 1 of 4
Printed: 15 Dec 2016
OX513A Emergency Response Procedures. Effective Date 18JUL1...
SECTION 1. INSTRUCTIONS
basic first aid materials. A functional phone service and a directory with phone numbers of nearby emergency medical services will be posted in both work places.
9.3 Reporting of accidents: A concise accident report should be filled out for all incidents that occur on site. Accident reporting sheets are available.
Oxitec Ltd CONFIDENTIAL Page 3 of 4
Printed: 15 Dec 2016