Effects of copper and zinc oxide nanoparticles on German cockroach development, indoxacarb resistance, and bacterial load

Abstract

BACKGROUND:

The German cockroach, Blattella germanica, is a ubiquitous and medically significant urban pest. The ongoing development of insecticide resistance in global populations of B. germanica has complicated control efforts and created a need for improved tools. We previously reported that disruption of the gut microbiota by oral administration of the antimicrobial doxycycline reduced resistance in an indoxacarb resistant field strain and also delayed nymphal development and reduced adult fecundity. However, the application of doxycycline for cockroach control in the field is impractical. Here, we sought to determine whether two metal nanoparticles with known antimicrobial properties, copper (Cu) and zinc oxide (ZnO), have similar effects to doxycycline on the physiology of B. germanica and could provide more practical alternatives for control.

RESULTS:

We found that dietary exposure to 0.1% Cu nanoparticles, but not ZnO, significantly delays the development of nymphs into adults. However, neither of the nanoparticles altered the fecundity of females, and ZnO surprisingly increased resistance to indoxacarb in a resistant field strain, in contrast to doxycycline. Semi-quantitative polymerase chain reaction (qPCR) further revealed that prolonged dietary exposure (14 days) to Cu or ZnO nanoparticles at the low concentration readily consumed by cockroaches (0.1%) does not reduce the load of the bacterial microbiota, suggesting alternative mechanisms behind their observed effects.

CONCLUSIONS:

Together, our results indicate that ingestion of Cu nanoparticles can impact German cockroach development through an undetermined mechanism that does not involve reducing the overall load of the bacterial microbiota. Therefore, Cu may have some applications in cockroach control as a result of this activity but antagonistic effects on insecticide resistance should be considered when evaluating the potential of nanoparticles for cockroach control.

1. INTRODUCTION

The German cockroach, Blattella germanica, is a widespread urban pest of public health importance.¹ In addition to contributing to the transmission of pathogenic microbes,^1–3 the German cockroach is also a significant source of indoor allergens linked to asthma. Interventions to control cockroach infestation can reduce allergen levels and asthma morbidity^4,5 as well as mitigate pathogen transmission.⁶ However, the spread of insecticide resistance across populations of B. germanica has reduced the effectiveness of some chemical control strategies.^7–10 In some instances, resistance has resulted in the failure of commonly used bait insecticides such as indoxacarb to control infestations in the field.^11,12

Targeting the microbiota of B. germanica for control (i.e. dysbiosis-based control) is an alternative strategy to supplement conventional insecticides that is garnering growing interest.¹³ German cockroaches harbor a highly diverse community of commensal microorganisms in their gut.¹⁴ The optimization of dysbiosis-based control requires fundamental knowledge of the contributions of these microbes to cockroach physiology as well as the identification of feasible and effective ways to manipulate the microbiota. Our group previously determined that bacteria in the gut contribute to resistance against the commonly used insecticide indoxacarb in B. germanica.¹⁵ This role was demonstrated using the common antimicrobial doxycycline, which enhanced the toxicity of indoxacarb when ingested. We also found that prolonged exposure to doxycycline delayed nymphal development and reduced the fecundity of adult females. Since this seminal work, others have identified further links between the cockroach microbiota and insecticide resistance and have demonstrated the utility of other antimicrobial compounds for disrupting the microbiota.^16–19 However, the administration of small molecule antimicrobials for control of cockroaches in the field is impractical and poses environmental health risks. For example, antimicrobials such as doxycycline, kanamycin, gentamicin, and rifampin have important clinical applications and their dissemination into the environment for cockroach control could exacerbate the emergence of antimicrobial resistant pathogens, a process in which cockroaches are already suspected to be involved.²⁰

Metal nanoparticles with antimicrobial activity are potential alternatives to small molecule antimicrobials for disrupting the gut microbiota of cockroaches to perturb their physiology. A number of metal nanoparticles are considered generally safe for humans, though research on the safety of many nanoparticles is lagging.²¹ Moreover, due to their non-specific mechanisms of action, metal nanoparticles are difficult for bacteria to develop resistance against and therefore their use does not pose one of the major risks that accompanies the use of small molecule antimicrobials.²² Some metal nanoparticles also have direct impacts on insect physiology,^23,24 which may be additionally helpful for control.

Both copper (Cu) and zinc oxide (ZnO) nanoparticles are known to possess potent antimicrobial properties and are relatively inexpensive.^25–29 The goal of the present study was therefore to determine whether Cu and ZnO nanoparticles have similar effects to doxycycline on the physiology of B. germanica and could provide more practical alternatives for control. To answer this question, we examined nymphal development, adult fecundity, and resistance to indoxacarb after dietary exposure to Cu and ZnO nanoparticles. To determine whether any observed effects on physiology could be explicitly linked to disruption of the gut microbiota or may involve alternate mechanisms, we also examined the effects of dietary exposure to nanoparticles on the load of the bacterial microbiota.

2. MATERIALS AND METHODS

2.1. Metal nanoparticle formulations

The Cu nanoparticles (99.8%, 500 nm) and ZnO nanoparticles (99.8%, 10–30 nm) were purchased from SkySpring Nanomaterials, Inc. (Houston, TX, USA). These nanoparticles were incorporated separately into either a highly palatable, proprietary gel bait matrix (Apex Bait Technologies, Inc., Santa Clara, CA, USA) that is consistent with several commercially available cockroach gel baits or an experimental dog chow gel formulation consisting of 50% water, 0.5% agar, and a remaining balance of finely ground dog chow (Purina, St Louis, MO, USA). Nanoparticles were incorporated at two different concentrations: 1% or 0.1% (w/w).

2.2. Feeding assay

To assess if Cu or ZnO nanoparticles have inhibitory effects on cockroach feeding, consumption of proprietary gel baits containing the nanoparticles was quantified in an overnight choice assay. This is a highly sensitive assay designed to detect small variations in bait palatability. In brief, an enclosure with approximately 200–300 cockroaches of mixed stages and sexes was prepared. The cockroaches were provided shelter, dog chow as an alternative food source, and water ad libitum. Gel baits containing the nanoparticles at each concentration, as well as a control gel with no nanoparticles, were weighed into roughly equal portions (approximately 4 g) and placed on individual Petri dishes side-by-side inside the cockroach enclosure. The next day, each of the gel baits was weighed again and the difference in weight was used to calculate overnight consumption in grams after adjusting for moisture loss. Moisture loss was determined as a percentage of weight lost by a control bait upon environmental exposure in the absence of cockroaches. Data were collected from three independent replicates and were analyzed by matched analysis of variance (ANOVA) with Tukey’s post hoc test and adjustment for multiple comparisons to compare consumption of the nanoparticle-containing baits and the control bait. The feeding assays were performed at the Apex Bait Technologies, Inc., facility using the laboratory adapted American Cyanamid Orlando Normal strain of B. germanica from a colony maintained as previously described.¹⁰

2.3. Development assay

To determine if consumption of Cu or ZnO nanoparticles affects nymphal development, groups of 50 first instar nymphs were placed in a 5 in × 8 in enclosure with shelter and water ad libitum and were provisioned 5 g of an experimental dog chow gel formulation containing 0.1% Cu or ZnO as a sole food source. Cockroaches provided the same dog chow gel without nanoparticles served as the control group. The cockroaches were monitored throughout development and the number of individuals molting to adulthood was recorded at regular intervals. Insects that did not develop to adulthood after a prolonged period were censored at the last timepoint of the experiment. Two independent replicates were conducted for each nanoparticle, and in each replicate development curves were compared to the control using a Mantel–Cox log-rank test. The development assays were performed at the University of South Dakota insectary facility using the laboratory adapted American Cyanamid Orlando Normal strain of B. germanica from a colony maintained as previously described.³⁰

2.4. Fecundity assay

To examine if consumption of Cu or ZnO nanoparticles affects fecundity, groups of five adult males and ten non-gravid adult females (with no visible oothecae) were placed in a 5 in × 8 in enclosure with shelter and water ad libitum and were provisioned 5 g of experimental dog chow gel formulation containing 0.1% Cu or ZnO as a sole food source. Cockroaches provided the same dog chow gel without nanoparticles served as the control group. The insects were allowed to mate and exposure to nanoparticles was continued as females developed oothecae. The number of viable offspring produced from each ootheca was recorded upon hatching. Two independent replicates were conducted and counts from individual ootheca were pooled (N = 7–11 total hatched oothecae per treatment group). The data were then analyzed by ANOVA with Dunnett’s post hoc test and adjustment for multiple comparisons to compare fecundity between nanoparticle treated and control females. The fecundity assays were performed at the University of South Dakota insectary facility using the laboratory adapted American Cyanamid Orlando Normal strain of B. germanica from a colony maintained as previously described.³⁰

2.5. Insecticide resistance assay

To determine if consumption of Cu or ZnO nanoparticles can reduce resistance to indoxacarb in a resistant field strain, as we previously demonstrated for doxycycline,¹⁵ we tested indoxacarb efficacy when administered in baits with or without the nanoparticles. In these experiments, enclosures containing approximately five adult males, five adult females, and 15 middle instar nymphs were provided shelter, dog chow, and water ad libitum and allowed to acclimate for 3 days. Following the acclimation period, proprietary gel baits containing 0.6% indoxacarb with or without 0.1% Cu or ZnO nanoparticles were added to separate enclosures. The concentration of indoxacarb was chosen to be in line with that found in commercially available cockroach baits. A control enclosure that received blank bait matrix without indoxacarb or nanoparticles was also included in each experiment. Over a period of 14 days, dead cockroaches (defined as exhibiting no movement when physically stimulated) were counted in each experimental group. Data were collected from three independent replicates and were analyzed by matched ANOVA with Tukey’s post hoc test and adjustment for multiple comparisons to compare the percent mortality at the termination of the 14 day experimental period. The insecticide resistance assays were performed at the Apex Bait Technologies, Inc., facility using the moderately indoxacarb resistant DCA field strain of B. germanica from a colony maintained as previously described.¹⁰

2.6. Quantitation of the microbiota by real-time polymerase chain reaction

To determine if consumption of Cu or ZnO nanoparticles can reduce the load of the bacterial microbiota or whether any observed physiological effects were due to alternate mechanisms, a semi-quantitative real-time polymerase chain reaction (qPCR) assay was carried out. Groups of first instar nymphs were provided shelter and water ad libitum and were separately provisioned an experimental dog chow gel formulation with 0.1% Cu or ZnO as a sole food source for 14 days. Cockroaches provided the same dog chow gel without nanoparticles served as the control group. Individual nymphs were then collected, and DNA was isolated from whole bodies using the DNeasy blood and tissue kit (Qiagen, Germantown, MD, USA) according to the manufacturer’s protocol. The qPCR was performed on a QuantStudio 3 instrument (Applied Biosystems, Waltham, MA, USA) using the PowerUp SYBR Green Master Mix (Applied Biosystems), 50 ng of cockroach DNA as template, and previously published primers targeting a conserved region of the bacterial 16S rRNA gene (331F/797R, F: 5’-TCCTACGGGAGGCAGCAGT-3’, R: 5’-GGACTACCAGGGTATC-TAATCCTGTT-3’).³¹ These primers cover 83.1% of bacterial taxa based on estimation using the SILVA TestPrime tool,³¹ allowing for simple simultaneous quantitation of diverse microbiota constituents, which would not be possible via culture. Amplification conditions were as follows: 95 °C for 5 min, followed by 40 cycles of denaturation at 95 °C for 15 s, annealing at 50 °C for 20 s, and extension at 72 °C for 30 s. Cycle threshold (CT) values were obtained for individual insects as a proxy for the bacterial load and analyzed by ANOVA with Dunnett’s post hoc test and adjustment for multiple comparisons. The qPCR assays were performed at the University of South Dakota using the laboratory adapted American Cyanamid Orlando Normal strain of B. germanica from a colony maintained as previously described.³⁰

3. RESULTS

3.1. Effects of nanoparticles on feeding

Both Cu and ZnO nanoparticles affected bait consumption by B. germanica to some degree in the overnight choice assay, although neither fully inhibited feeding even in the presence of alternative food choices (Fig. 1). We observed both nanoparticle-specific and dose-dependent effects. Incorporation of 0.1% ZnO into a gel bait did not affect cockroach consumption relative to the control, as mean overnight consumption was almost identical in the two groups (ANOVA, N = 3, P = 0.99). However, incorporation at 1% significantly inhibited consumption, as mean consumption of gel bait containing 1% ZnO was more than four-fold lower than the control (ANOVA, N = 3, P < 0.001). Incorporation of Cu into a gel bait even at the lower concentration of 0.1% resulted in a modest but statistically significant reduction in overnight consumption relative to the control bait (ANOVA, N = 3, P = 0.049). Moreover, 1% Cu also significantly inhibited feeding relative to the control (ANOVA, N = 3, P = 0.003). The inhibitory effect of 1% Cu was on average greater than 0.1% Cu, although this difference was not statistically significant (ANOVA, N = 3, P = 0.28). Based on the significant inhibition of feeding by 1% Cu or ZnO, this concentration was excluded from further investigation and studies of the effects of nanoparticle consumption on cockroach physiology were carried out using 0.1% Cu or ZnO.

Figure 1.

Effects of nanoparticles on feeding by Blattella germanica. Over-night consumption of a proprietary gel bait matrix containing copper (Cu) or zinc oxide (ZnO) nanoparticles by groups of cockroaches was quantified in a side-by-side choice assay. Gel bait matrix with no nanoparticles added served as the control. Three independent replicates were performed (N = 3), and data were analyzed by ANOVA with Tukey’s post hoc test. Shown are the mean, individual data points, and standard error. The P-values are adjusted for multiple comparisons.

3.2. Development of B. germanica nymphs exposed to nanoparticles in the diet

Dietary exposure to 0.1% Cu and ZnO nanoparticles had differing effects on the development of B. germanica nymphs (Fig. 2). In two independent replicates, 0.1% Cu caused statistically significant delays in time to adulthood (Mantel–Cox log-rank, P = 0.0025 and P < 0.001) (Fig. 2(A),(B)). In replicate 1, the median times to adulthood were 44 days for controls and 50 days for insects exposed to Cu. In replicate 2, the median times to adulthood were 50 days for controls and 71 days for insects exposed to Cu. However, 0.1% ZnO had no significant effect on time to adulthood in one replicate (Mantel–Cox log-rank, P = 0.23) (Fig. 2(C)) but resulted in a marginal though statistically significant hastening of development in a second replicate (Mantel–Cox log-rank, P = 0.01) (Fig. 2(D)). In replicate 1, the median time to adulthood was 58 days for both control insects and those exposed to ZnO. In replicate 2, the median times to adulthood were 61 days for controls and 54 days for insects exposed to ZnO. Importantly, while 0.1% Cu demonstrated some modest negative effects on bait palatability in the choice assay (Fig. 1), the effect of 0.1% Cu on food consumption was insignificant when considered in the absence of alternative food choices, as in the development assay (Supporting Information Fig. SS1).

Figure 2.

Development of Blattella germanica nymphs exposed to nanoparticles in the diet. Groups of first instar nymphs were provided an experimental dog chow gel formulation with 0.1% copper (Cu) (A, B) or zinc oxide (ZnO) (C, D) as a sole food source. Cockroaches provided the same dog chow gel without nanoparticles served as the control group. The cockroaches were monitored throughout development and the number of individuals molting to adulthood was recorded at regular intervals. Two independent replicates were conducted for each nanoparticle, and in each replicate development curves were compared to the control using a Mantel–Cox log-rank test.

3.3. Fecundity of adult female B. germanica exposed to nanoparticles in the diet

Dietary exposure to 0.1% Cu or ZnO did not meaningfully affect female fecundity (Fig. 3). The mean number of viable offspring produced from the oothecae of control females was 41.90 (N = 11), whereas the mean numbers of offspring produced from the oothecae of females exposed to 0.1% Cu or ZnO were 38.28 (N = 7) and 38.27 (N = 11), respectively. These differences were not statistically significant (ANOVA, P = 0.283 and P = 0.204) and are negligible in the context of pest management.

Figure 3.

Fecundity of adult female Blattella germanica exposed to nanoparticles in the diet. Groups of cockroaches were provided an experimental dog chow gel formulation containing 0.1% copper (Cu) or zinc oxide (ZnO) as a sole food source. Cockroaches provided the same dog chow gel without nanoparticles served as the control group. The number of viable offspring produced from the oothecae of individual females was recorded upon hatching. Data were combined from two independent experiments (N = 7–11 total hatched oothecae per treatment) and analyzed by ANOVA with Dunnett’s post hoc test to compare fecundity between nanoparticle treated and control females. Shown are the mean, individual data points, and standard error. The P-values are adjusted for multiple comparisons.

3.4. Indoxacarb resistance in B. germanica exposed to nanoparticles

The Cu or ZnO nanoparticles were incorporated into a proprietary gel bait containing indoxacarb to examine their effects on resistance to the insecticide in a resistant field strain of B. germanica (Fig. 4). In this assay, average mortality in cockroach groups exposed to blank bait matrix as a negative control was 3.4% after 14 days. Meanwhile, average mortality in cockroaches exposed to indoxacarb bait for 14 days was 62.3%, indicating that resistance was moderate but incomplete in the DCA field strain. Incorporation of 0.1% Cu into bait in combination with indoxacarb resulted in average mortality of 50.56% after 14 days. This reduction neared statistical significance relative to the indoxacarb only treatment (ANOVA, N = 3, P = 0.054). However, incorporation of 0.1% ZnO into bait with indoxacarb resulted in average mortality of only 40.76%, which was a significant reduction relative to indoxacarb alone (ANOVA, N = 3, P = 0.003). Together these results indicate that Cu and ZnO nanoparticles do not reduce and may possibly increase resistance to indoxacarb, in contrast to the effects of doxycycline, which can reduce resistance when administered in conjunction with the insecticide.

Figure 4.

Indoxacarb resistance in Blattella germanica exposed to nanoparticles. Enclosures containing mixed stages and sexes were administered proprietary gel baits containing 0.6% indoxacarb (Indx) with or without 0.1% copper (Cu) or zinc oxide (ZnO) nanoparticles. A control enclosure that received blank bait matrix without indoxacarb or nanoparticles was also included in each experiment. Over a period of 14 days, dead cockroaches (defined as exhibiting no movement when physically stimulated) were counted in each experimental group. Data were collected from three independent replicates and were analyzed by ANOVA with Tukey’s post hoc test to compare the percent mortality at the termination of the 14 days experimental period. Shown are the mean, individual data points, and standard error. The P-values are adjusted for multiple comparisons.

3.5. Effects of nanoparticle exposure on the microbiota

Our previous work attributed the physiological effects of doxycycline to its ability to perturb the gut microbiota of B. germanica.¹⁵ To determine if Cu and ZnO nanoparticles with known antimicrobial properties could similarly perturb the gut microbiota when consumed by cockroaches or whether the physiological effects we observed were due to alternate mechanisms, we examined the bacterial load in nymphs exposed to 0.1% Cu or ZnO in the diet for 14 days using a qPCR assay that targeted the bacterial 16S rRNA gene (Fig. 5). Surprisingly, there were no significant differences between the average CT values of control insects and those of insects exposed to either nanoparticle (ANOVA, N = 3, P ≥ 0.98), indicating that the load of the bacterial microbiota was not reduced by prolonged consumption of Cu or ZnO. The average CT values were 20.34, 21.39, and 19.52 for control insects, insects exposed to Cu, and insects exposed to ZnO, respectively.

Figure 5.

Effect of nanoparticle exposure on the load of the bacterial microbiota. Groups of first instar nymphs were provisioned an experimental dog chow gel formulation with 0.1% copper (Cu) or zinc oxide (ZnO) as a sole food source for 14 days. Cockroaches provided the same dog chow gel without nanoparticles served as the control group. DNA was then isolated from individual insect bodies and used in a semi-quantitative real-time polymerase chain reaction (qPCR) assay targeting a conserved region of the bacterial 16S rRNA gene. Cycle threshold (CT) values were obtained for individual insects as a proxy for the bacterial load and analyzed by ANOVA with Dunnett’s post hoc test. Shown are the mean, individual data points, and standard error. The P-values are adjusted for multiple comparisons.

4. DISCUSSION

Our results demonstrate that Cu and ZnO nanoparticles largely do not recapitulate the effects of doxycycline on cockroach physiology when given at a concentration that is readily consumed. For instance, we did not observe any effect of the nanoparticles on female fecundity (Fig. 3), nor of ZnO on development (Fig. 2). The lack of effect of both nanoparticles on the load of the bacterial microbiota (Fig. 5) may explain why their physiological effects were minimal and inconsistent with doxycycline. It is likely that the concentration of nanoparticles that is readily consumed by cockroaches is too low to significantly perturb the load of the bacterial community in the gut. Subtle effects of nanoparticle consumption on the composition of the bacterial community or on eukaryotic constituents of the microbiota cannot be ruled out as we did not conduct microbial community profiling. However, it is expected that if the nanoparticles were exerting significant antibacterial effects, like doxycycline, this would be apparent as a logarithmic reduction in bacterial load, as seen in other studies that targeted the microbiota of B. germanica with antimicrobials.¹⁷

Interestingly, we observed an unexpected, minor enhancement of insecticide resistance when the nanoparticles were administered in combination with indoxacarb (Fig. 4). This result remains unexplained but could be due to either effects of the nanoparticles on insecticide metabolism pathways or lower consumption of bait containing nanoparticles. The latter explanation is unlikely as incorporation of 0.1% ZnO into bait did not reduce consumption by cockroaches at all in a sensitive assay designed to specifically detect small effects of the nanoparticles on bait palatability (Fig. 1). However, prior work has reported microbe-mediated activation of indoxacarb in B. germanica as well as an association between selection for indoxacarb resistance and a reduction in commensal microbe transcripts.^12,32 Therefore, it is possible that Cu or ZnO nanoparticles may have targeted some specific eukaryotic or low abundance prokaryotic constituents of the microbiota that are inversely related to indoxacarb tolerance without reducing the overall bacterial load, contributing to the results in Fig. 4.

Although ZnO nanoparticles did not delay development significantly, in contrast to doxycycline, Cu nanoparticles produced notable delays in nymphal development in two independent replicates (Fig. 2). Considering that Cu nanoparticles did not decrease the bacterial load (Fig. 5) nor food consumption in the absence alternative food choices (Fig. SS1), the stunting of development appears to occur through a yet to be determined mechanism that is probably independent of any major disruption of the microbiota.

Prior work has documented detrimental effects of other metal nanoparticles on B. germanica, but their mechanisms also remain unknown. Feeding of silver nanoparticles (AgNO₃) at a concentration of 300 ppm (0.03%) resulted in > 90% mortality in both adults and nymphs within 3 days.²³ Moreover, consumption of gold nanoparticles (Au) at a concentration of 87.44 ug g⁻¹ did not cause mortality in adults but did reduce the number of nymphs hatching from oothecae, successful molting, and nymph lifespan.²⁴ The use of silver or gold nanoparticles for cockroach control on a large scale is cost prohibitive. However, our results suggest that Cu may have some limited applications in cockroach control due to its ability to affect growth. For instance, it could potentially be incorporated into some baits to enhance their effectiveness. Future studies should further investigate its mechanism(s) of action and confirm whether its effects are conserved under semi-field and field conditions. Other metal nanoparticles with known antibacterial properties could also be explored. Critically, our results suggest that antagonistic effects on insecticide resistance should be considered when evaluating the potential of nanoparticles for cockroach control.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.