Online Issues

ILAR Journal Vol 46(1)

Containment of Arthropod Disease Vectors

Thomas W. Scott

Thomas W. Scott, Ph.D., is Professor and Director of the Mosquito Research Laboratory in the Department of Entomology, University of California-Davis, Davis, California.

Abstract

Effective containment of arthropod vectors of infectious diseases is necessary to prevent transmission of pathogens by released, infected vectors and to prevent vectors that escape from establishing populations that subsequently contribute to increased disease. Although rare, past releases illustrate what can go wrong and justify the need for guidelines that minimize risks. An overview of recommendations for insectary facilities, practices, and equipment is provided, and features of four recently published and increasingly rigorous arthropod containment levels (ACLs 1-4) are summarized. ACL-1 is appropriate for research that constitutes the lowest risk level, including uninfected arthropods or vectors that are infected with micro-organisms that do not cause disease in humans, domestic animals, or wildlife. ACL-2 is appropriate for indigenous and exotic arthropods that represent a moderate risk, including vectors infected or suspected of being infected with biosafety level (BSL)-2 infectious agents and arthropods that have been genetically modified in ways that do not significantly affect their fecundity, survival, host preference, or vector competence. ACL-3 is recommended for arthropods that are or may be infected with BSL-3 infectious agents. ACL-3 places greater emphasis on pathogen containment and more restricted access to the insectary than ACL-2. ACL-4 is intended for arthropods that are infected with the most dangerous BSL-4 infectious agents, which can cause life-threatening illness by aerosol or arthropod bite. Adherence to these guidelines will result in laboratory-based arthropod vector research that minimizes risks and results in important new contributions to applied and basic science.

Key Words: arthropod containment levels; arthropod vector; biosafety; containment; insect vector; vector-borne disease

Introduction

For more than a century, since the pioneering work on malaria by Ronald Ross in India and on yellow fever by Walter Reed and the Yellow Fever Commission in Cuba (Harwood and James 1979), there has been a need to maintain in captivity arthropods that are known or believed to be involved in the transmission of pathogens to humans, domestic animals, and wildlife. Containment is necessary to gather critical information in a biologically safe and environmentally controlled setting on the behavior of arthropods and their life cycle, capacity to become infected with and transmit pathogens, and susceptibility to various approaches for interfering with pathogen transmission. Results from that kind of research are used to develop and refine strategies for disease prevention.

Although there have been few instances when caged arthropods have escaped and contributed to a public or veterinary health problem, preventing escape remains a high priority that requires appropriate procedures and facilities. Experimentally infected vectors that break out of containment represent an immediate risk because they could cause infections, augment transmission of a pathogen already present, or establish transmission of an agent novel to the release environment. Uninfected vectors represent a long-term risk because they could augment endemic populations or establish conditions receptive to the introduction of new pathogens. To minimize risks, a group of medical entomologists developed recommendations for proper vector containment. This review pertains to the evolution of those guidelines, general information on facilities and procedures for maintaining arthropod vectors in captivity, and recently published recommendations for four increasingly rigorous levels of containment. Interested readers are referred to Benedict et al. (2003) for the most recent and thorough presentation on this topic.

The discussion that follows is limited to arthropods that transmit pathogens to vertebrates. The discussion does not include species that are medically important due to their direct effects (e.g., scorpions, spiders, myiasis, bees, and wasps), species that transmit pathogens to nonvertebrate hosts (e.g., plants), or species that as part of their life cycle do not acquire vertebrate blood (e.g., Drosophila). Rather, the emphasis is on blood-sucking flies (e.g., mosquitoes, tsetse, black flies, sand flies, biting midges, horse and deer flies), bugs (e.g., kissing bugs and bed bugs), lice, fleas, ticks, and mites. To be consistent with literature on this topic, the text includes references to the facility where arthropod vectors are reared and maintained as an insectary, even though noninsect vectors (e.g., ticks and mites) can be held in such a structure. Also included is a review of recommendations for containment of arthropods that are infected with pathogens, that are uninfected, and that are genetically modified. An important motivation for the most recent containment guidelines was the explosion of research involving genetically modified vectors (Higgs 2003). Genetic modification represents a different set of risks than containment of infected and/or unaltered vectors. Consequently, it required its own set of appropriately designed responses. The ideal situation would be one in which containment is the same for arthropod vectors that are and are not genetically modified.

History

Although rare, introductions of exotic vectors have occurred; and four notable examples merit review. Three introductions resulted in substantial increases in disease. The third and fourth are still under way, with an evolving role in pathogen transmission. All four examples illustrate the risks associated with an exotic vector becoming established in a novel, receptive habitat (Benedict et al. 2003).

The first example began in 1915, when the Chagas disease vector, Rhodnius prolixus, is believed to have accidentally escaped from a research laboratory in El Salvador (Schofield 2000). It subsequently established itself throughout Central America where, because of its close association with humans, it remains the most important vector of Chagas disease throughout Central America and northern South America.

The second example similarly took place in Latin America. A campaign to reduce malaria in South America during the early part of the 20th century was successful, in part, because of severe reductions in populations of anopheline mosquito vectors. In 1930, however, the highly efficient African malaria vector, Anopheles gambiae, was detected in Natal, Brazil (Soper and Wilson 1943). Presumably it was accidentally introduced by ships from Africa. When malaria subsequently increased, A. gambiae was considered the culprit. The forceful campaign that eliminated it from the continent remains one of the great successes in the history of public health.

The third example concerns the classic range expansion of Aedes aegypti--the yellow fever mosquito--into the New World, presumably via ships from Africa that were part of European colonization of the western hemisphere and the slave trade (Tabachnick 1991). Introduction of this highly efficient arbovirus vector resulted in the establishment of New World peridomestic mosquito populations, urban yellow fever became a regular and devastating component of the public health landscape, and sylvan yellow fever cycles were established in jungle settings. An attempt to eradicate A. aegypti from the New World during the 1960s was successful in most regions, but incomplete. Development of a yellow fever vaccine and redirected health priorities led to a relaxation of the program. A. aegypti reinvaded areas from which it had been removed, and in many areas population densities are currently greater than they have ever been. The reinvasion of A. aegypti facilitated a Latin America dengue pandemic that began in the early 1980s and continues today as one of the most important public health burdens for that region of the world.

The fourth example began in North America during the later half of the 20th century and is still unfolding. In 1985, the Asian tiger mosquito, Aedes albopictus, was detected in Houston, Texas (Hawley 1988; Sprenger and Wuithiranyagool 1986). Its introduction was traced to the importation of tires from Asia. A. albopictus became the subject of considerable research and media attention because it had a documented capacity for efficient transmission of a variety of arthropod-borne viruses (arboviruses). Although several different viruses were recovered from wild A. albopictus, the significance of its role in the transmission of those agents in North America remained questionable. From the point of introduction, it expanded its range to the north and east (Lounibos 2002) and became an important pest mosquito in much of its new habitat. The introduction of West Nile virus (WNV¹) into North America during 1999 may have elevated A. albopictus from a pest to a public health threat. Laboratory studies on vector competence indicate that it is an efficient vector of WNV (Komar 2003). Research by a number of groups is currently under way to define its role in the epidemic transmission of WNV.

For most of the 20th century until the 1980s, containment of arthropod vectors was left to the good intentions of individual investigators or his/her institution. In 1980, a set of guidelines was published by the Subcommittee on Arboviral Laboratory Safety (SALS¹) (SALS 1980). This proactive group of arbovirologists wanted to establish safe and reasonable recommendations for laboratory arbovirus research. Based on surveys from hundreds of laboratories, they assigned each known arbovirus to one of four levels of containment and laboratory practice. Their goal was to protect people working in the laboratory and to prevent escape of viruses from laboratories. At each level, recommendations for containment of infected arthropod vectors were provided. The report did not address vectors that are not infected with an arbovirus, nonviral agents, or genetically modified vectors. In 1984, the US Department of Health and Human Services expanded the format developed in the SALS report to include all microbial infectious agents (DHHS 1999). Biosafety in Microbiological and Biomedical Laboratories (BMBL¹) emphasizes containment in the context of "laboratory practice and technique, safety equipment, and facility design." The 4th edition is available online (http://www.cdc.gov/od/ohs/biosfty/bmbl4/bmbl4toc.htm).

To this point, guidelines for vector containment were a component of recommendations that were aimed at preventing the escape of microbial agents. In 1996, vector containment became an area of primary emphasis unto itself with the publication of two articles. Hunt and Tabachnick (1996) have recommended equipment, procedures, and facilities to provide safe containment of small, zoophilic arthropods that are infected with biosafety level (BSL¹)-3 pathogens. Minute, flying, infected arthropod vectors (in this case, biting midges that transmit exotic bluetongue virus) are more difficult to contain and monitor for than larger vectors such as mosquitoes. The second article is a book chapter by Higgs and Beaty (1996). They used, as an example for other institutions, facilities and practices successfully applied for several decades of research on mosquitoes infected with BSL-3 arboviruses at Colorado State University. Their chapter includes valuable diagrams of insectary design and photographs illustrating equipment and procedures.

In 1999, two additional important events took place. First, the remarkable progress made in genetic manipulation of organisms stimulated the National Institutes of Health to revise their existing safety guidelines for research with animals containing recombinant DNA (NIH 1999). That document recommended maintaining arthropods containing recombinant DNA in facilities consistent with BSL-2. Second, the American Committee of Medical Entomology (ACME¹), a group affiliated with the American Society of Tropical Medicine and Hygiene (ASTMH¹), adopted a resolution to develop arthropod containment guidelines consistent with the format of and augmenting the BMBL. A subcommittee was formed, guidelines were developed, and draft documents were shared among the subcommittee and members of the scientific community beginning in 2000 (Higgs 2003). Subsequent draft documents were posted on a website, circulated via email to subcommittee members and researchers expected to be interested in vector containment guidelines, and discussed at ASTMH open meetings during 2000 and 2001. After a series of revisions, a final draft of Arthropod Containment Guidelines (ACG¹) was approved by ACME at the 2002 annual meeting of the ASTMH and published in a special issue of Vector-Borne and Zoonotic Diseases (Benedict et al. 2003). The guidelines are available at no cost via the internet (see References).

Insectary Design and Procedures

Information in this section is a general overview of insectary design and procedures based on previous reviews (Duthu et al. 2001; Higgs 2004a,b; Higgs and Beaty 1996; Higgs et al. 2003; Hunt and Tabachnick 1996; Olson et al. 2003; Richmond 2003). Specific containment levels are presented below, in Arthropod Containment Levels.

Field Laboratories

An important distinction for insectary guidelines is whether they are associated with a field site or with more permanent structures that are subject to laboratory guidelines recommended in ACG. Researchers, institutions, biosafety committees, other regulatory bodies, and funding agencies must determine the extent to which ACG applies to field research. The ACG describes field laboratories as temporary facilities in which research is carried out with locally collected arthropod vectors (Benedict et al. 2003). If arthropods escape from a field laboratory, they are not expected to augment the existing population density or genetic structure of local populations or to increase the risk of transmission of pathogens to humans or other animals. Escape of local arthropods maintained in field laboratories is minimized by appropriate handling practices and primary containment (i.e., caging). Field laboratories do not require significant modification of existing facilities. Because they are often located in places where the arthropods being studied participate in pathogen transmission, it is possible that specimens brought into a field laboratory may be infected. Appropriate precautions should be taken to protect researchers and people residing or working near the field laboratory from infection with microbial agents.

A dimension of field research that is currently receiving considerable attention is that of genetically modified vectors. The goal is to use genetic approaches to prevent vectors from transmitting pathogens or to reduce their population densities (Scott et al. 2002). One advocated scenario is transitional assessments of candidate genetically modified vectors from laboratory containment to large outdoor cages that simulate the natural environment before release of free-ranging, modified arthropods. Knols and colleagues (2003) present a helpful list of facilities and procedural recommendations for stringent containment of genetically modified mosquitoes in large outdoor cages, or what they call "semi-field structures." The remainder of this article concerns insectary space in permanent laboratories that are not considered field laboratories or semi-field facilities.

Insectary Design Features

All insectaries should be designed to prevent arthropod escape. The facility should be inspected and evaluated for proper containment during design, construction, and operational phases. Access should be restricted to people directly involved in research or support staff. Entry and exit by passing through two doors (i.e., double door access), with the first closing before the second is opened, creates a double barrier to escape. Cloth or air curtains (i.e., stiff blast of air) can be added to doors to dislodge insects from clothing of exiting personnel and to prevent flying insects, such as mosquitoes, from escaping. Doors should close automatically and fit tightly on all four sides. Doors that open inward are preferable because they tend to push flying insects back into the insectary. Entryways should be designated with proper signage that identifies the vectors held inside, whether infectious agents are being studied, and the principal investigator and other appropriate contacts. Screening should be the proper mesh to prevent escape by the species being held inside.

It is advisable not to have windows in an insectary, but if windows are present, they should be permanently sealed. Interior walls and surfaces should permit ready detection of escaped arthropods (e.g., all-white surfaces on walls, floors, and work surfaces) and allow decontamination (i.e., washing or spraying with disinfectants). Construction should minimize or eliminate opportunities for escaped arthropods to avoid detection. For example, there should be no gaps in the wall around entering water pipes or electrical conduits. Similarly, wastewater traps can be designed so that accumulated water is either disinfected or heated before discharge to avoid accidental production of vectors with aquatic life stages.

Within the insectary, screened rooms or vestibules fitted with automatically closing doors are desirable for maintaining different vector species or strains. Higgs and Beaty (1996) provide a valuable example for insectary design in a diagram of existing insectary space at Colorado State University (see Figure 35.4).

Only equipment items that are required for ongoing research should be kept in the insectary. Equipment should be located so that it can be accessed conveniently and without increasing risk of arthropod escape or contamination of insect colonies. Lighting should be installed so that it can be accessed easily for replacement of fixtures without compromising containment. It may be advantageous to have an autoclave in the insectary for decontamination capabilities depending on the level of containment or the particular research activities performed with arthropods.

Construction materials should be capable of withstanding environmental conditions appropriate for rearing and maintaining arthropods, because insectaries are often maintained at warm temperatures and high humidity to support arthropod survival. In some cases, environmental chambers are placed in insectaries to maintain appropriate environmental conditions for vector survival.

The physical association of the insectary to other laboratories or research activities is an important consideration. The risk of arthropod escape should not increase during transfer from an insectary to a laboratory where analyses are carried out. If it is not possible to locate the insectary next to the research laboratory, procedures should be developed that minimize the risk of escape. For example, it is possible to make nonbreakable transport containers, or to limit research to times of the year when outdoor temperatures are intolerant to escaped species (e.g., freezing temperatures outside). Guidelines for transportation of arthropods among laboratories or from field sites to a laboratory should adhere to state and federal regulations.

Insectary Procedures and Equipment

Procedures and equipment constitute the primary barrier for effective arthropod vector containment. Facilities are able to contain vectors effectively only when procedures are carried out properly and equipment is used correctly. Staff training in the biology and containment of the arthropod is a fundamental component of proper insectary procedures. Training is an ongoing process that requires periodic updates as new species are brought into the insectary, new equipment is used, and new guidelines for the insectary are developed.

General procedures should be developed for maintaining an effective insectary. The facility should be kept clean at all times and inspected regularly to ensure that containment has not been compromised (e.g., there are no holes in the screening) and that equipment is functioning correctly. Procedures for effective containment vary with the risk associated with infected versus uninfected versus genetically modified arthropods. Some general procedures are reviewed below; however, laboratory-specific procedures should be developed for particular species and activities.

Standard practices for rearing and safely containing colonies of arthropod vectors are available for mosquitoes (Gerberg 1970; Gerberg et al. 1994; Singh and Moore 1985), biting midges (Hunt 1994; Hunt and Tabachnick 1996; Jones 1960), ticks (Jellison and Philip 1933; Jones et al. 1988; Waladde et al. 1991), and fleas (Hudson and Prince 1958; Wade and Georgi 1988). Of critical importance is the establishment of a monitoring system for escaped vectors. In an experimental setting, it is important to record and account for the number of specimens under study. Depending on research and colony maintenance activities, different kinds of traps can be used as surveillance tools to capture escaped arthropods. Traps can include physical barrier substances such as "Tangle Trap" (Professional Pest Control Products of Pensacola, Inc., Pensacola, FL), which can be applied around the perimeter of doors to trap and/or monitor for both escaped crawling arthropods (e.g., tick, kissing bugs) and/or potential pests like ants and cock roaches. Caution should be used, however, so that traps effectively capture the species held in the insectary. Some trap designs do not attract and capture all arthropod species, and traps that work well under field conditions do not necessarily work well in a confined insectary space. A monitoring trap system can be based on direct (i.e., collection of the escaped arthropod itself) or indirect (i.e., collection of eggs in ovitraps) detection of arthropods that have escaped primary containment.

Cages for arthropod containment should be designed to prevent vector escape and include ways to access arthropods for maintenance and/or experimentation without increasing the risk that they will escape. A glove box is a useful device for manipulating vectors, removing them from cages, and so forth. Arthropods that escape during the procedure remain in the glove box from which they can be captured and destroyed. Depending on the species under consideration, it is possible to deploy secondary barriers to escape from a cage (e.g., a moat of oil under and around cages containing ticks or sand flies, which will trap and kill arthropods that escape). Development of immature arthropods should be monitored so that they do not complete their metamorphosis and emerge as free-flying adults. For example, mosquito pupae should be checked and collected daily before adults emerge and fly freely in the insectary.

Most arthropods involved in pathogen transmission require a meal of vertebrate blood to support their survival and reproduction. When intact vertebrates are used as a source of blood, they should not be housed in the same room as the arthropods to avoid accidental infection with a vector-borne pathogen. Transport of vertebrates or vectors after blood feeding should be done in unbreakable containers. Details of the procedures for feeding arthropods on vertebrates must be approved before their use by the appropriate institutional animal care and use committee (IACUC¹) or other institutional or governmental oversight organization.

To avoid using intact vertebrates for maintaining arthropod colonies or during experiments, procedures have been developed to feed vectors on artificial blood meals. A variety of devices and protocols exist for different arthropod species (Cosgrove et al. 1994; Fahrner and Barthelness 1988; Hastriter and Cavanaugh 1981; Hastriter et al. 1980; Hunt and McKinnon 1990; Kogan 1990; Richman et al. 1999; Rutledge et al. 1964; Waladde et al. 1991). In general, artificial blood meals consist of blood that has been treated to prevent clotting, warmed to stimulate arthropod feeding, and presented through a membrane that the mouth parts of the arthropods can penetrate. Artificial blood meals are advantageous because they reduce the need for vertebrate exposure to biting and/or infected vectors. They are disadvantageous because they typically result in fewer eggs produced and lower levels of arthropod infection than when an intact vertebrate is the source of blood or infectious agent. Some arthropod species or strains will not imbibe blood from an artificial meal.

Arthropod Containment Levels

Risk Assessment

The assignment of particular arthropods and vector research activities to a specific biosafety level requires qualitative judgment because no set criteria exist. In the process of assessing risk, Benedict and colleagues (2003) point out that "several factors must be considered in combination: the agents transmitted, whether the arthropod is or may be infected, the mobility and longevity of the arthropod, its reproductive potential, biological containment, and epidemiological factors influencing transmission in the proposed location or region at risk" (p. 69). The principal investigator or laboratory director, who is responsible for the initial steps in risk assessment, should work in concert with the institutional biosafety committee (IBC¹) to guarantee compliance with biosafety guidelines and regulations.

The ACGs list four discrete risk categories; there can be variation within each grouping. After each category, the ACG provides lists of question that are helpful in assigning a level of containment. The questions and categories are not intended to be all inclusive. Instead, they illustrate how investigators and their IBCs can assess and minimize risk associated with arthropod vector research. The first category concerns arthropods that are known to be free of specific pathogens. This category typically involves a low level of risk category, except when arthropod escape could lead to an immediate or future increase in pathogen transmission. The second category involves arthropods that are known to contain a specific pathogen. Here, risk is associated with patterns and efficiencies of transmission as well as severity of disease. Higher levels of containment are recommended for vectors that contain more virulent agents. When infectious agents are involved, a medical surveillance program should be established to evaluate containment practices and to provide the basis for rapid medical response if laboratory personnel become infected. The third category relates to arthropods whose infection status is unknown or who may contain unknown infectious agents. Because the risk cannot be clearly defined, a conservative approach is recommended for assigning a containment level. The fourth category relates to vectors that have been genetically modified so that they express recombinant DNA molecules. Genetic modification can be of arthropods themselves or micro-organisms in the vector. Risk involves (1) phenotypic change associated with genetic modification and its potential impact on wild-type populations if the arthropod escapes, and (2) the presence of an autonomous element that could potentially move within the genome, thus leading to unpredicted genetic changes. Guidelines for Research Involving Recombinant DNA Molecules (NIH 1999) is a basic reference for assessing risk and assigning containment for genetically modified arthropod vectors and micro-organisms in vectors.

Containment Levels

In 2003, a subcommittee of ACME (Benedict et al. 2003) published the following recommendations for practices, equipment, and facilities that constitute arthropod containment levels (ACLs¹) 1 through 4. Note that each successive ACL builds on and, unless otherwise stated, includes components of the preceding level of containment. Personnel who work with pathogen-infected vectors and vertebrates should simultaneously consider vertebrate animal biosafety levels, which are discussed in the BMBL.

ACL-1

ACL-1 is appropriate for research that constitutes the lowest level of risk; for example, uninfected arthropods or vectors that are infected with micro-organisms that do not cause disease in humans, domestic animals, or wildlife. This group includes arthropods already found in the area where the research is being performed and exotic vectors that would not establish a local population or would not contribute to pathogen transmission if they escaped.

Procedures in an ACL-1 facility are consistent with standard insectary practices reviewed above. Arthropods are held in ways that minimize the possibility of escape or of contact with humans or other vertebrates. Cages (i.e., primary containment) are constructed of materials that prevent escape and can be thoroughly cleaned and disinfected after each use. All cages or containers that hold arthropods are secure and properly labeled for identification (e.g., species, strain, data collected). Arthropods of all life stages are killed before disposal. Care is taken to prevent arthropod transport from the insectary on personnel. A system is in place to exclude arthropod and vertebrate pests (i.e., cockroaches, houseflies, mice, and rats). A monitoring system is in place to detect escaped arthropods. Potential hiding and development sites are eliminated. Standard microbiology laboratory procedures are established to minimize risks associated with using syringes, needles, and other sharp implements. Entry into the insectary should include the proper signage.

Special practices, primary barriers (safety equipment), and secondary barriers (facilities) are similarly consistent with standard insectaries. Any accidental release is promptly reported to the insectary director, and escaped arthropods are captured and/or killed using a mechanical aspirator or by crushing. They should not be crushed with a bare hand or collected with a mouth aspirator. Use of vertebrate animals as a blood source requires approval from an IACUC. Vertebrates not being used to feed arthropods either are housed in a room separate from the vectors or, if in the same room, are protected from vectors that might escape and try to bite them. Procedures are instituted to prevent arthropod escape during blood feeding and when vertebrates are placed into or removed from the cage holding the arthropods. Precautions are taken to prevent the unintentional infection of arthropods that imbibe blood meals.

Unless it is exempt, research with recombinant DNA requires IBC approval. Gloves and white laboratory coats should be worn when handling vertebrates or their blood. When appropriate (e.g., for allergies or head covers), personal protective equipment should be worn. Insectaries should be located so that they are not accessible to general traffic in the building. Entryways should minimize arthropod escape from within and entry from outside. If windows are present, they are sealed to prevent exit or entry of arthropods.

ACL-2

ACL-2 is appropriate for arthropods that represent a moderate potential risk. This level is recommended for vectors infected or suspected of being infected with BSL-2 infectious agents and arthropods that have been genetically modified in ways that do not significantly affect their fecundity, survival, host preference, or vector competence. ACL-2 is applicable to indigenous and exotic arthropods.

ACL-2 includes and extends recommendations for ACL-1 so that the requirements for procedures, equipment, and facilities are more rigorous than for ACL-1. Facilities and procedures are in place for improved detection of escaped arthropods. Nonbreakable cages and other arthropod rearing or holding equipment are designed to prevent escape and are sterilized or incinerated after use, whether or not the arthropods in them were infected. All arthropods are autoclaved or incinerated before disposal. Uninfected vectors are isolated from those that are known to be infected. Enhanced procedures are in place to prevent, detect (e.g., monitoring system), and record arthropod escape from the insectary. Escape of infected arthropods within the insectary is prevented by handling them in a glove box, biosafety cabinet, or special devices that prevent escape. Places where arthropods can hide or breed are eliminated. All surfaces are constructed so that they can be decontaminated; work surfaces are routinely decontaminated, and equipment is decontaminated before transfer to different rooms in the insectary or transport out of the insectary. Signage on the exterior of the insectary lists the arthropods, BSL-2 infectious agents known or suspected to be inside, and the name and contact information for the insectary or research director.

A safety manual that is developed and approved by the local IBC is in the insectary. Personnel working in the insectary receive training consistent with this level of containment. Access is limited to trained personnel and service staff who are aware of the hazards in the insectary. A medical surveillance system is advisable. IBC and IACUC approvals are obtained as necessary. Vertebrate housing is consistent with ACL-1, but arthropods are more rigorously contained during blood feeding, whether or not vertebrates are known to be infected with a vector-borne agent. Inadvertent releases of arthropods are reported, and arthropods that escape but remain in the insectary are captured and killed in ways that do not expose personnel to pathogens. When personnel are in the insectary, they should wear their laboratory coats, and gloves when handling potentially infectious material; and they should not wear open-toed shoes.

ACL-2 insectaries are more physically removed and have more barriers to arthropod escape than ACL-1. The insectary is separate from the main flow of activity in the building and has a double-door access with doors that self-close inward, are not both opened at the same time, and fit tightly on all sides. It is advisable to have no windows, but if they are present, they are sealed. Interior surfaces are a light color so that escaped arthropods can be seen easily, and they are constructed of materials that can be decontaminated. If present, vacuum systems and floor drains must have traps or other barriers that prevent arthropod escape. Ventilation systems should be designed so that they prevent arthropod escape (e.g., screens or filters at locations where air enters or exits the insectary). It is advisable to have negative air pressure in the insectary; air must flow into the insectary. Lighting does not create opportunities for arthropods to hide or escape. The insectary director inspects the facility regularly and at least annually, or perhaps makes alterations as necessary to maintain containment.

ACL-3

ACL-3 is recommended for research with arthropods that are or may be infected with BSL-3 infectious agents. The assignment of this level extends facilities and practices beyond ACL-2 by placing greater emphasis on pathogen containment and more restricted access to the insectary. Guidelines for work with BSL-3 agents in ACL-3 require use of Class II biosafety cabinets fitted with a high-efficiency particulate air-filtered exhaust system to protect laboratory personnel and to prevent pathogen release. Cases exist, however, in which the use of biosafety cabinets in insectaries may increase rather than decrease safety risks. For example, during the manipulation of small arthropods, a biosafety cabinet may create air currents that blow them away. After recovery from sedation, arthropods may become fully mobile and constitute a risk to insectary personnel or escape from the insectary. In those cases, it may be best to perform arthropod manipulations in a protected, isolated area free from harmful air currents. When there are no risks of arthropod escape (e.g., assaying for a pathogen in cell culture), the use of biosafety cabinets is recommended. Similarly, personnel in ACL-3 should wear proper personal protective clothing and equipment. Because protective clothing and equipment can impede dexterity, it may be necessary to modify its use so that risks of personal exposure and arthropod or pathogen escape are minimized. Protocols that address insect manipulation in biosafety cabinets and while wearing protective gear should be developed and approved with the local IBC.

An isolated area within the insectary should be identified and used for work with arthropods known or suspected to be infected with a BSL-3 agent. A small, secure room that constitutes several layers of containment and where escaped arthropods can be readily detected is often best for this purpose. Glove boxes are useful for manipulating infected arthropods in ACL-3.

Standard ACL-3 practices emphasize pathogen containment and decontamination, and take into consideration any risks associated with the possible generation of infectious aerosols. According to the recommendations, only arthropods requiring ACL-3 containment should be held in an ACL-3 insectary, and the efficiency of containment barriers should be evaluated. Materials used to wipe down surfaces, arthropods that will be eliminated, arthropod waste, and primary containers are autoclaved or incinerated before they are discarded. In addition to duplicating the information posted outside an ACL-2 insectary, there should be proper signage at the entrance of the ACL-3 facility to inform people of its designation.

Training of personnel who work in the ACL-3 insectary is more thorough than for ACL-2. A medical surveillance system is in place, the facilities director limits access to only essential personnel, and untrained personnel (maintenance staff) who enter the facility are accompanied by trained staff and apprised of the potential hazards. Routine facility maintenance is carried out by trained laboratory personnel. Known or potentially infected arthropods are handled or manipulated only in the proper containment devices. The design of those devices depends on the life history and behavior of the arthropod being studied. Guidelines for containment of blood-feeding arthropods are followed meticulously, and escaped arthropods are captured and/or killed properly so that laboratory personnel are not exposed to the infectious agent. Use of personal protection (e.g., gloves, garments, and foot wear) is strictly enforced. Pesticides are available in case they are needed to kill escaped arthropods.

As is the case for a BSL-3 laboratory, an ACL-3 insectary is isolated from areas of main traffic in the building. The insectary has a double-door entrance with a secure outer door (e.g., key lock or card key). Windows should be avoided, but if they are present, they should be sealed and resistant to breakage. An autoclave is available in the insectary area. ACL-3 insectaries can include a shower. It is advisable to avoid floor drains, but if they are present, they should be fitted with proper traps to prevent arthropod escape. Ventilation should always be flowing into the insectary and designed so that it does not allow arthropod escape. Design and operations in the ACL-3 insectary are approved before use and annually by the IBC.

ACL-4

ACL-4 is intended for research with arthropods that are infected with the most dangerous BSL-4 infectious agents, which can cause life-threatening illness by aerosol or arthropod bite. Containment guidelines include those required for ACL-3 plus the following: personnel working in the facility must shower before entering and when leaving, and work with infectious agents is performed either in a Class III biosafety cabinet or while wearing a positive pressure laboratory safety suit. No compromise of the containment guidelines is allowed. The design of BSL-4 facilities and procedures must be followed strictly. Arthropods in an ACL-4 must be contained properly at all times, whether or not they are being manipulated. Facilities and procedures should be appropriate for the species being manipulated or studied. Although insects can be used experimentally in BSL-4 agent research, most vectors of agents in this category are ticks. The highly infectious and virulent nature of BSL-4 agents mandates rigorous training of staff who work in the facility, specialized equipment specifically designed for ACL-4 research, and facilities and protocols that have been approved by the IBC.

Conclusions

Laboratory research with arthropod vectors of vertebrate disease is essential for an improved understanding of the role of the vector in pathogen transmission and for development of enhanced disease prevention strategies. Facilities and practices should minimize or eliminate the escape of arthropods and pathogens, protect personnel who work in the insectary, and protect the community in which the insectary resides. Pathogen-infected vectors represent an immediate threat, but even uninfected arthropods that escape captivity can establish populations that subsequently transmit pathogens. Recently published guidelines (Benedict et al. 2003) provide thorough recommendations for appropriate insectary construction, procedures, and equipment for various kinds of arthropod vectors that represent different risks. Containment and a safe working environment require a sustained effort to correct or upgrade, as needed, components of laboratory work with arthropod vectors. Therefore, while engaging in the formulation of answers to compelling research questions, successful vector research programs must focus on the critical component of laboratory safety.

Acknowledgments

I thank M.Q. Benedict and S. Higgs for their leadership in the development of arthropod vector containment guidelines. J.M. Grumstrup-Scott edited an earlier version of the manuscript. Three anonymous reviewers provided helpful suggestions for modification of this review.

¹Abbreviations used in this article: ACG, Arthropod Containment Guidelines; ACL, arthropod containment level; ACME, American Committee of Medical Entomology; ASTMH, American Society of Tropical Medicine and Hygiene; BMBL, Biosafety in Microbiological and Biomedical Laboratories; BSL, biosafety level; IACUC, institutional animal care and use committee; IBC, institutional biosafety committee; SALS, Subcommittee on Arboviral Laboratory Safety; WNV, West Nile virus.

References

Benedict, MQ, Tabachnick WJ, Higgs S, Azad AF, Beard CB, Beier JC, Handler AM, James AA, Lord CC, Nasci RS, Olson KE, Richmond JY, Scott TW, Severson DW, Walker ED, Wesson DM. 2003. Arthropod Containment Guidelines. Vector-Borne Zoonotic Dis 3:57-98. Available online (http://iris.ingentaselect.com/vl=11142254/cl=47/nw=1/rpsv/cw/mal/15303667/v3n2/contp1-1.htm).

Cosgrove JB, Wood RJ, Petric D, Evans DT, Abbott HR. 1994. A convenient mosquito membrane feeding system. J Am Mosq Control Assoc 10:434-436.

DHHS [US Department of Health and Human Services]. 1999. Biosafety in Microbiological and Biomedical Laboratories. Washington DC: DHHS. Available online (http://bmbl.od.nih.gov).

Duthu, DB, Higgs, S, Beets, RL, McGlade, TJ. 2001. Design issues for insectaries. In: Richmond JY, ed. Anthology of Biosafety. IV. Issues in Public Health. Mundelein IL: American Biological Safety Association. p 227-244.

Fahrner J, Barthelmess C. 1988. Rearing of Culicoides nuberculosus (Diptera: Ceratopogonidae) by natural and artificial feeding in the laboratory. Vet Parasitol 28:307-313.

Gerberg EJ. 1970. Manual for mosquito rearing and experimental techniques. American Mosquito Control Association Bulletin 5. Selma CA: American Mosquito Control Association.

Gerberg EJ, Barnard DR, Ward, RA. 1994. Manual for mosquito rearing and experimental techniques. American Mosquito Control Association Bulletin 5, Revised. Lake Charles LA: American Mosquito Control Association.

Harwood RF, James MT. 1979. Entomology in human and animal health. New York: Macmillan Publishing Co., Inc.

Hastriter MW, Cavanaugh DC. 1981. An apparatus for colonizing fleas (Siphonaptera) and collecting pupal cocoons. J Med Entomol 18:251-252.

Hastriter MW, Robinson DM, Cavanaugh DC. 1980. An improved apparatus for safely feeding fleas (Siphonaptera) in plague studies. J Med Entomol 17:387-388.

Hawley, WA. 1988. The biology of Aedes albopictus. J Am Mosq Control Assoc Suppl 4:1-39.

Higgs S. 2003. Editorial. Vector-Borne Zoonotic Dis 3:57-58.

Higgs S. 2004a. Care, maintenance, and experimental infection of mosquitoes. In: Marquardt, WC, Kondratieff, B, Freier, J, Hagedorn, HH, eds. The Biology of Disease Vectors. New York: Academic Press (In Press).

Higgs S. 2004b. The containment of arthropod vectors. In: Marquardt, WC, Kondratieff, B, Freier, J, Hagedorn, HH, eds. The Biology of Disease Vectors. New York: Academic Press (In Press).

Higgs S, Beaty BJ. 1996. Rearing and containment of mosquito vectors. In: Beaty BJ, Marquardt WC, eds. The Biology of Disease Vectors. Niwot: University Press of Colorado. p 595-605.

Higgs S, Benedict MQ, Tabachnick WJ. 2003. Arthropod containment guidelines. In: Richmond JY, ed. Anthology of Biosafety. VI. Arthropod Borne Diseases. Mundelein IL: American Biological Safety Association. p 73-84.

Hudson BW, Prince FM. 1958. A method for large-scale rearing of the cat flea Ctenocephalides felis felis (Bouché). Bull WHO 19:1126-1129.

Hunt GJ. 1994. A procedural manual for the large-scale rearing of the biting midge Culicoides variipennis (Diptera: Ceratopogonidae). Springfield VA: National Technical Information Service. USDA Research Service, ARS-121.

Hunt GJ, McKinnon CN. 1990. Evaluation of membranes for feeding Culicoides variipennis (Diptera: Ceratopogonidae) with an improved artificial blood-feeding apparatus. J Med Entomol 27:934-937.

Hunt GJ, Tabachnick WJ. 1996. Handling small arbovirus vectors safely during biosafety level 3 containment: Culicoides variipennis sonorensis (Diptera: Ceratopogonidae) and exotic bluetongue viruses. J Med Entomol 33:271-277.

Jellison WL, Philip CB. 1933. Technique for routine and experimental feeding of certain ixodid ticks on guinea pigs and rabbits. Public Health Rep 48:1081-1082.

Jones RH. 1960. Mass-production methods in rearing Culicoides variipennis sonorensis. J Econ Entomol 53:731-735.

Jones, LD, Davies CR, Steele GM, Nutall PA. 1988. The rearing and maintenance of ixodid and argasid ticks in the laboratory. Anim Technol 39:99-106.

Knols, BGJ, Njiru BN, Mukabana RW, Mathenge EM, Killeen GF. 2003. Contained semi-field environments for ecological studies on transgenic African malaria vectors: Benefits and constraints. In: Takken W, Scott TW, eds. Ecological Aspects for Application of Genetically Modified Mosquitoes. Wageningen: Fontis Press. p 91-106.

Kogan PH. 1990. Substitute blood meal for investigating and maintaining Aedes aegypti (Diptera: Culicidae). J Med Entomol 27:709-712.

Komar, N. 2003. West Nile virus: Epidemiology and ecology in North America. Adv Virus Res 61:185-234.

Lounibos, LP. 2002. Invasions by insect vectors of human disease. Ann Rev Entomol 47:233-266.

NIH [National Institutes of Health]. 1999. Guidelines for Research Involving Recombinant DNA Molecules. 63 FR 25361. Washington DC: GPO. Available online (http://www4.od.nih.gov/oba/rac/guidelines/guidelines.html).

Olson, KE, Larsen, RE, Ellis RP. 2003. Biosafety issues and solutions for working with infected arthropods. In: Richmond JY, ed. Anthology of Biosafety. VI. Arthropod Borne Diseases. Mundelein IL: American Biological Safety Association. p 25-38.

Richman DL, Koehler PG, Brenner RJ. 1999. Spray-dried bovine blood: An effective laboratory diet for Ctenocephalides felis felis (Siphonaptera: Pulicidae)isiknowledge.com/CIW.cgi?SID=QLN3jArg-GgAAFntSE8&Func=Abstract&doc=4/1. J Med Entomol 36:219-221.

Richmond JY, ed. 2003. Anthology of Biosafety. VI. Arthropod-Borne Diseases. Mundelein IL: American Biological Safety Association.

Rutledge LC, Ward RA, Gould DJ. 1964. Studies on the feeding response of mosquitoes to nutritive solutions in a new membrane feeder. Mosq News 24:407-419.

SALS [Subcommittee on Arbovirus Laboratory Safety of the American Committee on Arthropod-Borne Viruses]. 1980. Laboratory safety for arboviruses and certain other viruses of vertebrates. Am J Trop Med Hyg 29:1359-1381.

Schofield CG. 2000. Challenges of Chagas disease vector control in Central America. WHO, Communicable Disease Control, Prevention and Eradication, WHO Pesticide Evaluation Scheme. Geneva: WHO.

Scott TW, Takken W, Knols BGJ, Boëte C. 2002. The ecology of genetically modified mosquitoes. Science 298:117-119.

Singh P, Moore RF, eds. 1985. Handbook of Insect Rearing. Vol 2. New York: Elsevier.

Soper FL, Wilson DB. 1943. Anopheles gambiae in Brazil: 1930 to 1940. New York: The Rockefeller Foundation.

Sprenger D, Wuithiranyagool T. 1986. The discovery and distribution of Aedes albopictus in Harris County, Texas. J Am Mosq Control Assoc 2:217-219.

Tabachnick WJ. 1991. Evolutionary genetics and arthropod-borne disease: The yellow fever mosquito. Am Entomol 37:14-24.

Wade SE, Georgi JR. 1988. Survival and reproduction of artificially fed cat fleas, Ctenocephalides felis Bouché (Siphonaptera: Pulicidae). J Med Entomol 25:186-190.

Waladde SM, Ochieng SA, Gichuhi PM. 1991. Artificial-membrane feeding of the ixodid tick Rhipicephalus appendiculatus to repletion. Exp Appl Acrol 11:297-306.