The Use of Granular Cyclopentanone as Alternative to Artificial Source of Carbon Dioxide in Improved Passive Outdoor Host Seeking Device (POHD)

Reliable sources of CO2 that are relatively cheap, obtainable, and easy to sustain are immediately required for scaling up of odor-baited mosquito surveillance and control devices. Several odor-baited devices are in the pipeline; however, their scale-up against residual malaria transmission, particularly in resource poor areas, is limited by the unavailability of reliable sources of CO2 and reliance on electrical power sources among other factors. We evaluated the use of granular cyclopentanone as an alternative to artificial or yeast fermentation-produced CO2 in passive outdoor host seeking device (POHD). Experiments were conducted against semifield reared An. arabiensis within the semifield system (SFS) at Ifakara Health Institute. Mosquitoes were tested against odor-baited POHDs augmented with yeast fermentation-produced CO2, granular cyclopentanone, attractive blends (Mbita or Ifakara), or their combinations. An insecticide, bendiocarb, was a killing agent used as a proxy for marking the mosquitoes visit the POHDs. Relative attractiveness of different treatment combinations was compared based on the proportion of dead mosquitoes that visited the POHD. The POHD augmented with granules of cyclopentanone alone was attractive to An. arabiensis as much as, or more than, POHDs augmented with yeast fermentation-produced CO2. The POHD baited with CO2 attracted more mosquitoes than those POHDs baited with synthetic blends alone; when these blends are combined with CO2, they attracted more mosquitoes than individual blends. More importantly, such POHDs baited with cyclopentanone attracted far greater proportion of mosquitoes than the POHD baited with either Mbita or Ifakara blend alone. The granular cyclopentanone strongly enhanced/potentiated the attractiveness of POHD baited with Mbita blends against mosquitoes compared to that of POHD baited with Ifakara blend. Moreover, the granular cyclopentanone retained its residual activity against An. arabiensis for up to 2 months after application particularly when used in combination with Mbita blend. In conclusion, this study demonstrates that cyclopentanone granules have the potential to substitute sources of CO2 in outdoor-based surveillance and control devices, thus warranting evaluation of such alternative under realistic field conditions.

ese conventional sources of CO2 include pressurized gas cylinders of industrial CO 2 [30,40], dry ice [33,39,44], and burning of propane [45]. Examples of traps baited with CO 2 include BG-sentinel traps [40,46], CDC light traps [39,40], counterflow geometry traps [47], Mosquito Magnet-X traps (MMX-trap) [38], mosquito landing boxes [35], and mosquito trapping boxes [33]. However, these conventional sources of CO 2 are expensive, unobtainable, and labor intensive to transport and sustain for large scale surveillance and control programs in resource poor areas [48]. erefore, reliable sources of CO 2 that are relatively less expensive, easy to use, and easy to sustain in remote poor resource settings are urgently required for use in odor-baited devices. Otherwise, the control of residual malaria transmission, which occurs almost exclusively outdoors, will remain a difficult endeavor.
Recently, several organic sources of CO 2 have been proposed for use in mosquito traps for surveillance [32,49,[50][51][52]. ese organic sources of CO 2 include yeast fermentation of sugar/sucrose and molasses [32,[52][53][54], electrolyzed oxalic acids (Harwood et al., 2014), granular CO 2 sachets [55], and food-grade sources such as mixture of citric acid, water, and powdered sodium bicarbonate (NaHCO3), or vinegar (5%) mixed with sodium bicarbonate (NaHCO3) [56]. However, traps baited with CO 2 from organic sources attract relatively fewer mosquitoes than those baited with industrial CO 2 in pressurized gas cylinders or dry ice [55,56]. ese organic sources of CO 2 are still expensive and logistically difficult to use for large scale surveillance and control programs because they require frequent replenishment of materials to extent residual activity. erefore, the portable, easy to use, long-lasting, and cheaper novel sources of CO 2 that may attract high proportions of mosquitoes, comparable to or higher than those attracted with traditional sources of CO 2 , are urgently required for large scale application of odor-baited devices in poor resource countries.
One of the desirable options could be the use of novel compounds that mimic CO 2 such as cyclopentanone (C 5 H 8 O) [57,58], acetone [59], and 2-butanone [60,61]. ese compounds activate CO 2 receptor (cpA) neuron on the maxillary palps of mosquitoes, and some of them have been tested against Anopheles, Culex, and Aedes mosquitoes as substitutes for CO 2 in odor-baited traps without success [17,58,60]. Traps baited with liquid formulation of cyclopentanone had similar catches of Culex quinquefasciatus to those traps baited with industrial CO 2 under semifield conditions [17], but not in the field [58]. Similarly, the liquid formulation of 2-butanone increased attractiveness of synthetic blend (e.g., Mbita blend) to mosquitoes including Anopheles gambiae and Anopheles funestus the same as CO 2 in the odor-baited traps under the field but not semifield environments [60]. However, it was hypothesized that liquid formulation of cyclopentanone or 2-butanone in a cotton wick may depreciate relatively quickly due to variations of wind speed, moisture, temperature, and competing odors from the surrounding natural vegetation [58,60]. erefore, there is a need for slow release formulations of cyclopentanone or other CO 2 mimics such as 2butanone to enhance further development and large scale use of traps and odor-baited outdoor control devices [17,58,60]. is study aimed to (i) evaluate the potential of cyclopentanone granules to enhance/potentiate the attractiveness of synthetic blends in POHD against An. arabiensis, (ii) compare the efficacy of POHD baited with granular and strip formulation of cyclopentanone against An. arabiensis, and (iii) assess the residual activity/persistence of such formulations against An. arabiensis.

Study Site.
Experiments were conducted within the semifield system (SFS) located at Kining'ina village in Kilombero Valley (Figure 1), about 6 km from Ifakara town. e SFS has several chambers (2.97 × 6.70 × 2.80 m) with temperature range from 26 to 32°C.

Mosquitoes.
e colony of An. arabiensis was established in 2008 by collection of gravid females from the wild population at Sagamaganga village, Kilombero Valley (8.0667 S, 36.8000 E) [62].
is colony of mosquitoes is reared under ambient conditions in one chamber (3.30 × 2.70 × 2.50 m) of the semifield system (60 × 20 m) [62]. e temperature and relative humidity in this system range from 25 to 32°C and from 70 to 90%, respectively. e larvae are kept in rearing basins and are fed on TetraMin ® (Tetra GmbH, Germany), as finely ground baby fish food flakes twice to three times a day. e adult mosquitoes are maintained inside the cages (0.45 × 0.45 × 0.45 m) with 10% glucose solution. Female An. arabiensis mosquitoes used throughout this study were 3-7 days old.

Improved POHD and Different Treatments.
e improved POHD with bottom mosquito entrance was used in these experiments (Figure 2, Kessy et al., unpublished). Cyclopentanone, CO 2, and synthetic blend (Mbita or Ifakara) [36,63] were deployed as attractants in the improved POHD, singly or in combination to compare their attractiveness against An. arabiensis female mosquitoes. e Mbita and Ifakara blends were designed and developed to either sachet containing granules or soaked nylon strips by 2 e Scientific World Journal Biogents AG, Germany. Two forms of CO 2 were used: (i) the yeast fermentation-produced CO 2 from modified mixture of yeast (8.75 g) and molasses (250 g) [50], and a total volume of 1 L of warm water based on previous studies that produced CO 2 using at least 1 g of yeast in a total volume of ≤1 or >1 L of water [32,52,64,65], (ii) 20% cyclopentanone (C 4 H 8 O) used in previous studies [17], which was modified and formulated to granules and impregnated nylon strips by Biogents AG, Germany. e synthetic attractive blends and cyclopentanone were stored in the refrigerator at 4°C between experiments. A powder formulation of bendiocarb (Ficam D) was applied on electrostatic charged netting [66] to kill mosquitoes visiting the improved POHD. Bendiocarb is a nonrepellent insecticide that acts by contact against mosquitoes including the population of An. arabiensis in Kilombero Valley [72]. us, bendiocarb was a suitable carbamate for use in POHD baited with lures to attract and kill any visiting mosquitoes A large plastic jug filled with a mixture of warm water, molasses, and yeast to produce CO 2 . A plastic tube directs the flow of CO 2 downwards.
The CO 2 -producing container is inserted and suspended inside a 6 inch diameter PVC pipe by cross brace of wood or metal. The open end of plastic pipe points downwards.
Treated or untreated electrostatic net is sewn to form a cone with a small top opening allowing a sachet of cyclopentanone, attractive blend, or their combination to protrude. This is then inserted into the lower end of the pipe.
The completed POHD is suspended by cords to allow downward flow of CO 2 or cyclopentanone and synthetic odour plumes which attract mosquitoes to come in contact with the treated or untreated surface of the net inside the tube Figure 2: Schematic drawing and pictures showing improved passive host seeking device (POHD) with bottom placement of synthetic attractive blends and mosquito killing agent (bendiocarb). Components of the POHD are (a) the inner red jug containing mixture of warm water, molasses, and yeast for production of carbon dioxide (CO 2 ) and a rubber tube for channelling CO 2 downwind; (b) the CO2-producing plastic jug is inserted and suspended inside 6-inch PVC pipe using fixed wood pieces; (c) a holed conical electrostatic netting untreated or treated with powdered bendiocarb and plugged with a bag/sachet containing granules of synthetic attractive blend, the granules of novel CO 2 mimic compound, and 20% cyclopentanone as alternative to yeast fermentation-produced CO 2 ; (d) a complete POHD with polyvinyl chloride (PVC) outer cover to allow downwind flow of odor and CO 2 plumes to attract mosquitoes towards treated or untreated netting inside the device. and to quantify the number of dead mosquitoes as a proxy for attractiveness of the POHD to mosquitoes.

Potentiating/Enhancing Effects of Cyclopentanone.
e sachets of granular formulation or impregnated nylon strips of either Mbita (Mb) or Ifakara blend (Ib) were hung inside the improved POHD. Eight (8) different treatment combinations were evaluated: (1) CO 2 + cyclopentanone + odor blend + untreated netting (untreated), (2) bendiocarbtreated netting alone (Be), (3) odor blends + Be (Mb or Ib), (4) CO 2 + Be (CO), (5) cyclopentanone + Be (Cy), (6) CO 2 + odor blends + Be (MbCO or IbCO), (7) cyclopentanone (Cy) + odor blends + Be (CyMb, or CyIb), and (8) odor blend + CO 2 + Cy + Be (CyMbCO or CyIbCO). e improved POHD components were assembled and hung at the middle of the chamber within the SFS, 0.25 m off the ground (Figures 2 and 3(a)). In each experiment, a total of 100 starved female mosquitoes (25 individuals per cup) were released at four different corners of the SFS against the aforesaid treatment combinations. Mosquitoes were left to forage for overnight. e next morning, all live or dead mosquitoes from inside the POHD and other parts of the SFS chamber were independently recovered, counted, and recorded. All live mosquitoes were kept in the semifield insectary, provided with 10% glucose solution, and monitored for mortality after 24 hrs. Treatment combinations were randomly alternated between the days of the experiments. Between experiments, any mosquitoes remaining inside the experimental SFS chamber were removed using CDC backpack aspirator to avoid spillover effect. ese experiments were replicated three times for each treatment combination.

Persistence of Granular Cyclopentanone.
Persistence of cyclopentanone was evaluated using a rectangular bioassay box (1.87 × 2.12 × 1.15 m), inside three different SFS chambers (2.50 × 9.50 × 9.00 m), one box per chamber ( Figure 3(b)). e synthetic blend that was strongly enhanced by granular cyclopentanone from the efficacy experiment above, Mbita blend in this case, was selected to be combined with cyclopentanone in these experiments. Persistence was counted from the day the cyclopentanone started being used after preparation, and month is abbreviated as "mo" in the text and the figures. e treatment combinations were as follows: (1) fresh granules of cyclopentanone (Cy) + Mbita blend (Mb) + untreated netting (untreated), (2) fresh granules of Mbita blend alone (Mb), (3) 2 mo old granules of cyclopentanone + bendiocarbtreated netting-Be (2moCy), (4) 2 mo old granules of cyclopentanone + Mb + Be (2moCyMb), (4) fresh granules of cyclopentanone + Be (FreshCy), and (5) fresh granules of cyclopentanone + Mb + Be (FreshCyMb). en, improved POHD was assembled and hung at the middle of the rectangular box at 0.25 m off the ground (Figure 3(b)). e two sides of the box were closed with cardboard, but two holes (0.11 cm in diameter) were made for the release of mosquitoes inside during experiments. Each time of the experiment, a total of 100 female mosquitoes starved for 6 hours without glucose were released inside the box through 4 holes (25 individual mosquitoes/holes × 4 holes � 100 mosquitoes) in the evening at 7:00 pm. Recapture and monitoring of mosquitoes were done following the same procedures as in the efficacy experiments above. ese experiments were replicated three times for each treatment combination.

Comparing Efficacy of Different Formulations of Cyclopentanone.
ese experiments were conducted using rectangular bioassay boxes (in experiment 2 above), with the aim of comparing the efficacy of cyclopentanone applied on granules and nylon strips delivery formats against mosquitoes. e formulations of cyclopentanone were tested in combination with long-lasting granular formulation of Mbita blend (depicted from persistence experiments). Different treatment combinations evaluated in the improved POHD were as follows: (1) fresh cyclopentanone + Mbita blend + untreated netting, (2) fresh cyclopentanone strip + Mb + Be (strips), and (3) fresh cyclopentanone granules + Mb + Be (granules). e improved POHDs incorporated with different treatment combinations were assembled (independently) and hung at the center of the box (Figure 3(b)). Mosquitoes were exposed to improved POHDs, recaptured, and monitored for 24-hour mortality, in the same way described in experiments 1 and 2 above. ese experiments were replicated three times for each treatment combination.

Ethical Considerations.
e research team and technical assistants responsible for rearing and handling mosquitoes were routinely (i.e., after every 7 days) screened for malaria to ensure that experimental materials are free of malaria parasites. e semifield had double doors, and it was routinely checked to ensure that the screens were always intact to prevent escape of mosquitoes to the environment. e remains of bendiocarb and/or treated materials were disposed accordingly and thereafter incinerated. is work was also granted permission for publication by National Institute of Medical Research in Tanzania (NIMR/HQ/P.12/Vol. XXVIII/93).

Statistical Analysis.
Statistical analyses were conducted to test the efficacy of cyclopentanone as alternative to CO 2 for enhancing/potentiating the attractiveness of synthetic blends, to determine the duration of their effects (persistence), and to compare between nylon strips and granular formulations against Anopheles arabiensis. Response variable measured in these experiments was proportion of mosquitoes killed by different treatments of 4 e Scientific World Journal POHD.
e response variable of proportion of dead mosquitoes was analyzed using generalized linear mixed effect models with binomial errors (glmer) in the R statistical software package [71]. e explanatory variables, "treatments," "blend type," and "formulations," were considered as the main effects, whereas "days" of the experiments were considered as random effects. A base model including only random effect of "day" was constructed. A sequential addition of the "main effects" and their interaction (treatment × blends or treatment formulation) to the base model was conducted to construct a maximal model (forward stepwise approach). A statistical significance of fixed effects and interaction term was generated and evaluated using likelihood ratio tests (LRTs). When the interaction terms were statistically significant, the main effect of either "blend type" or "formulation type" for each synthetic attractant was analyzed separately to generate estimates for the main effects. en, the full model was used to perform a two-ways multiple comparisons using Tukey post hoc tests (adjusting for multiple comparison) to establish statistical significant differences between treatments.

Discussion
is study revealed the potential of granular formulation of cyclopentanone as an alternative to artificial CO 2 applied in odor-baited traps against An. arabiensis that maintains residual transmission of malaria in most of African countries. e improved POHD baited with granular cyclopentanone was attractive as much as, or more than, the POHD baited with fermented CO 2 to An. arabiensis. Plumes of both the granular cyclopentanone and CO 2 act in the same way on mosquitoes CO 2 receptors (cpA). is finding compares with the results of recent semifield and field experiments that tested liquid formulations of 2-butanone combined with nylon strips of Mbita blend [60] and previous semifield experiment that used liquid formulation of 20% cyclopentanone [17]. However, these results were both different from that of the field trial conducted by Philippe-Janon et al. [58]. Such contrast was attributed to the quick loss of attractiveness by liquid formulation cyclopentanone under field settings consequent to strong wind as well as fluctuations on temperature and relative humidity. erefore, the slow release formulation (granules) of cyclopentanone offers an exciting option for use in odor-baited mosquito surveillance and/or control devices. Sachet of granular cyclopentanone is less bulky and easy to apply and distribute in large scale surveillance compared to other sources of CO 2 such as pressurized cylinder of CO 2 and gallons for yeast fermentation-produced CO 2 , and it lasts relatively longer than liquid formulations [17,58,60].
us, it can be deployed in resource poor areas like rural Tanzania. Although production of CO 2 from a mixture of yeast and molasses was in certain circumstances as effective as, or less effective than, cyclopentanone, it has a drawback in that the mixture of materials in portable containers can release CO 2 for approximately 8-12 hours and resources are highly needed for domestic consumption and logistically difficult to change and distribute on daily basis for large scale surveillance and control of mosquitoes.
Also, this study indicated that granular cyclopentanone had stronger potentiating/enhancing effect on synthetic blends than that of the yeast fermentation-produced CO 2 . Such effect of cyclopentanone on synthetic blends varied between blend types applied on the POHD. Nevertheless, the granular formulation of cyclopentanone enhanced/potentiated attractiveness of POHD baited with Mbita blend to An. arabiensis by 70%, but it enhanced the attractiveness of POHD baited with Ifakara blend by 56%. Previous semifield and field experiments indicated that Mbita blend is relatively more attractive to mosquitoes than Ifakara blend when augmented with CO 2 [63]. e current study also found that the combinations of either blend with cyclopentanone or CO 2 in POHD attracted far greater proportion of mosquitoes relative to the individual blends, Cy alone, or CO 2 alone. is observation that CO 2 enhances attractiveness of odor blends is also emphasized in several studies conducted elsewhere [11,16,30,37,38,43]. e results of this study also confirms the recent findings that traps baited with Mbita blend and augmented with CO 2 mimic compounds such as 2-butanone attracted similar number of mosquitoes as those traps baited with odor blends combined with CO 2 gas [60]. Since plumes of cyclopentanone and CO 2 act in the same way on mosquito's CO 2 receptors [17], they may be used to augment synthetic blends in POHD and other devices of the sort.
Moreover, the residual activity of 2-month-old granular cyclopentanone was reduced by 47% relative to the fresh formulation when applied in POHD. Nevertheless, when such formulation was complemented with Mbita blend, the efficacy was elevated to 76%. Such stable residual activity of cyclopentanone could have been contributed by their storage under refrigeration temperature between experiments and slow release granular formulation.
is slow release formulation may have enabled the older cyclopentanone to retain its enhancement/potentiation effects on synthetic blends in the POHD that attracted and killed >80% of the exposed mosquitoes. Similarly, previous studies showed that the application of BG lures as granules increased their residual activity for up to 5 months [34,72]. ese findings therefore indicate that attractiveness of cyclopentanone granules lasts for relatively longer time after application than other CO 2 sources and thus will need to be replaced less frequently when applied in odor-baited traps for large scale vector surveillance and control.
On the other hand, fresh granules and nylon strips of cyclopentanone were equally effective in enhancing attractiveness of POHD baited with Mbita blend against An. arabiensis.
e possible explanation could be that these experiments were conducted within a small sized chamber (1.87 × 2.12 × 1.15 m) where cyclopentanone with high volatility, when used alone or combined with Mbita blend, yielded equal concentration of odor plumes in a short range/ distance to mosquitoes regardless of the formulations. erefore, mosquitoes equally detected the plumes from different treatments. Variation between the two formulations could have been observed if they were compared under natural field conditions. Previous field work demonstrated that traps baited with nylon strip impregnated with Mbita blend combined with CO 2 attracted as much Aedes and  Figure 6: Estimated proportion (±1 s.e.) of An. arabiensis mosquitoes that were attracted and killed after exposure to an improved passive host seeking device that was untreated or treated with bendiocarb (BE) and baited with impregnated nylon strips or granular formulations of cyclopentanone (Cy) in combination with Mbita blend (Mb) in a rectangular bioassay box within the semifield system. e treatments were abbreviated as follows: control: attractants without benthiocarb-treated nettings, strips: cyclopentanone impregnated nylon strips, granules: granular cyclopentanone.
e statistical significant differences between treatments are indicated as follows: the asterisks indicate statistical significant differences from all other treatments with open boxes without asterisks. Error bars represent ±1 standard error. e Scientific World Journal Culex mosquitoes as the traps baited with combination of long-lasting granular formulation of BG lures and CO 2 [37]. Contrastingly, the recent study demonstrated that fresh liquid formulation of 2-butanone combined with synthetic odor blend attracted more Anopheles mosquitoes to the traps than traps baited with odor blend alone in the field, but not under the semifield conditions [60].
Although impregnated nylon strips of synthetic blends have been shown to remain attractive to mosquitoes for several weeks to months [34,35,43,63,73], they are not suitable for mass production and use in large scale mosquito surveillance and control programs [54]. Overall, the granular formulation of cyclopentanone may be portable, durable, lightweight, and easy to produce and use in odor-baited devices for large scale surveillance and control of mosquitoes. However, further investigations are required to assess long-lasting enhancing effects of granular formulations of cyclopentanone on synthetic blends in natural environments.
In conclusion, the present study demonstrates that cyclopentanone has the potential to substitute artificial source of CO 2 in odor-baited devices for surveillance and control of An. arabiensis, thus warranting evaluation of such alternative under realistic field conditions.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.