The gut microbiota induces melanin deposits that act as substrates for fimA-mediated aggregation of Salmonella Typhimurium and enhance infection of the German cockroach vector

ABSTRACT When Salmonella Typhimurium is ingested by German cockroaches, the bacteria replicate in the gut and persist for at least 7 d, enabling transmission in the feces. However, the mechanisms that facilitate survival and persistence in the cockroach gut remain poorly detailed. We previously reported the formation of biofilm-like aggregate populations of S. Typhimurium in the gut of cockroaches upon ingestion. We also reported that deletion of the type-1 fimbrial subunit of S. Typhimurium, fimA, leads to a reduced bacterial load in the cockroach gut. Here, we link these observations and provide further insight into the mechanism and function of S. Typhimurium aggregation in the gut of the cockroach. We show that S. Typhimurium but not Escherichia coli forms aggregated populations in the cockroach gut, and that aggregate formation requires fimA but not the biofilm formation-related genes csgA and csgD. Furthermore, we show that S. Typhimurium aggregates are formed using small granular deposits present in the cockroach gut, which exhibit properties consistent with melanin, as substrates. These melanin deposits are prevalent in the guts of both immature and adult cockroaches from laboratory colonies and are correlated with increased gut bacterial density while being entirely absent in gnotobiotic cockroaches reared without exposure to environmental bacteria, indicating they are induced as a response to the gut microbiota. When cockroaches lacking melanin deposits in the gut are fed S. Typhimurium, they exhibit lower rates of infection than those harboring melanin deposits, demonstrating that microbiota-induced melanin deposits enhance infection of the gut of the vector. IMPORTANCE Cockroaches, including the German cockroach (Blattella germanica), can be both mechanical and biological vectors of pathogenic bacteria. Together, our data reveal a novel mechanism by which S. Typhimurium interacts with the cockroach gut and its microbiota that promotes infection of the vector. These findings exemplify the emerging but underappreciated complexity of the relationship between cockroaches and S. Typhimurium.

. Other experimental studies have demonstrated the ability of cockroaches to act as vectors of enteric pathogens via transfer on the cuticle and shedding in the feces (7)(8)(9)(10)(11).Although mechanical transmission of pathogens by cockroaches is somewhat appreciated, understanding of their potential for dynamic, biological transmission of pathogens is poorly developed.For example, when German cockroaches ingest Salmonella enterica serovar Typhimurium (S.Typhimurium), these bacteria replicate in the gut, reaching a steady state and persisting for at least 7 days without eliciting pathogenesis (11,12).Yet, the underlying mechanisms that facilitate S. Typhimurium survival, persistence, and shedding in the cockroach gut are almost entirely unknown.
In addition to documenting multiple phases of S. Typhimurium replication in the gut of B. germanica, we previously reported the presence of two distinct populations of the bacteria within the gut (11).As early as 3 h after S. Typhimurium is ingested, biofilm-like aggregates form around melanized granular deposits present in the foregut, which we hypothesize are generated as an immune response (melanization) to other microbes present in the gut or to preexisting tissue damage (13).The aggregated populations of S. Typhimurium exist alongside planktonic bacteria.However, the regulation of the formation of these aggregates, including whether they represent true biofilms, as well as their importance for infection of the cockroach vector, remains undetermined.
The formation of biofilms and aggregates allows bacteria to survive harsh environ mental conditions, including treatment with disinfectants and shifts in temperature, osmolarity, O 2 , CO 2 , pH, and nutrient availability (14).A variety of genes and regula tory networks are involved in biofilm and aggregate formation by S. Typhimurium (15).Furthermore, S. Typhimurium biofilms and aggregates have been shown to play important roles in colonization of both vertebrate and invertebrate hosts (16,17).
In Salmonella, csgD is the master transcriptional regulator responsible for the switch from a planktonic to a biofilm lifestyle (15).csgD regulates csgA and bcsA to produce curli fibers and cellulose, respectively, which are key components of biofilms (18)(19)(20).These components are also co-regulated with the O-antigen of the Salmonella capsule to remodel the extracellular matrix for biofilm formation (21).One study showed that biofilms enhanced virulence and facilitated vertical transmission in chickens, with the genes csgD and bcsA playing critical roles (22).A separate study in chickens showed that csgA and bcsA have significant roles in biofilm formation and without them, biofilm formation as well as virulence was reduced (23).In the context of invertebrate hosts, biofilm formation by another enteric pathogen, Vibrio cholerae, facilitates selective colonization of the fly intestinal tract, extending the role of biofilms across distant phyla (24).The same is true for S. Typhimurium in Caenorhabditis elegans.In this invertebrate, the transcriptional regulator SsrB is responsible for the switch between a virulent and chronic infection state.Specifically, unphosphorylated SsrB activates csgD, allowing for the transcription of biofilm formation components and promoting a persistent infection (25,26).
Relatedly, we previously showed that S. Typhimurium deficient in type 1 fimbriae (27) from fimA deletion were deficient in their ability to persist in the cockroach gut (11).However, the underlying mechanism behind this phenotype is unclear.Type 1 fimbriae are important for Salmonella formation of biofilms on gallstones in the gallbladders of mice and humans as a means of chronic persistence (28).Furthermore, a separate study showed that point mutations in fimA can affect the host tropism of Salmonella by altering affinity toward the ligand N-acetyl-D-glucosamine (29).
Given the above, we hypothesized that the establishment of aggregated populations of S. Typhimurium in the cockroach gut may involve fimA-mediated adhesion and serve to enhance infection of the vector.In the present study, we used a combination of microscopy, molecular biology, and bacterial culture assays to further characterize aggregate formation by S. Typhimurium in the cockroach gut, its underlying mecha nisms, and its effects on vector infection.

Melanized deposits are common in the cockroach gut and act as substrates for S. Typhimurium aggregation during infection
When we dissect guts from uninfected colony derived cockroaches, we frequently observe melanized deposits in the foregut (Fig. 1A and B).Respectively, these were noted by stereomicroscopy in 4/12 nymphs examined after a 1-day starvation period, in 6/12 adult males examined after a 1-day starvation period, and in 8/12 adult males examined after a 3-day starvation period (Fig. 1C).The deposits are consistent in appearance with melanization produced as an immune response to bacteria and tissue damage that occurs in other insects both systemically (13,(30)(31)(32) and in the gut (33,34).When adult male cockroaches were orally infected with wild-type S. Typhimurium, aggregates of bacteria were observed to form specifically around the melanized deposits in the foregut within the first 3 h of infection (Fig. 2B and C).This phenomenon was specific to S. Typhimurium and was seen in 100% of infected insects examined by fluorescence microscopy but did not occur when the cockroaches were infected with Escherichia coli (Fig. 2A and D, Fisher's exact test, P = 0.002).

Aggregate formation on melanized deposits by S. Typhimurium is dependent on fimA but not csgA or csgD
To examine the mechanism(s) behind formation of the aggregates, we orally infected cockroaches with 1 of 5 different strains of S. Typhimurium (wild-type, ΔfimA, ΔinvAspiB, ΔcsgA, and ΔcsgD) and examined foreguts 3 h post-infection for the presence of bacterial aggregates around the melanized deposits (Fig. 2D).The deletion mutants were chosen to determine if the respective functions of the deleted genes in biofilm formation, adhesion, and host manipulation are required for aggregate formation around melanized deposits in the cockroach gut.The gene fimA encodes the major structural subunit of type 1 fimbriae, invA and spiB encode essential components of type III secretion systems

Melanized deposits in the cockroach foregut are induced by the bacterial microbiota and demonstrate properties consistent with melanin
Based on their appearance, we hypothesized that the melanized deposits present in the cockroach foregut around which S. Typhimurium forms aggregates are composed of melanin produced as an immune response to commensal bacteria.We performed a series of four different experiments to test this hypothesis.First, we compared the density of the bacterial microbiota in guts from colony derived adult male cockroaches binned based on the presence or absence of melanized deposits in the foregut using a semi-quantitative PCR (qPCR) assay targeting a conserved region of the bacterial 16S rRNA gene (Fig. 3A).In this assay, cockroach guts that had visible melanized deposits had an average cycle threshold (CT) value of 17.38.This was significantly lower than the average CT value of guts that did not have visible melanized deposits, which was 19.22 (t-test, P = 0.0382, Fig. 3A).This difference indicates a 3.58-fold increase in bacterial density in guts with melanized deposits relative to those without.
To further investigate the link between the gut microbiota and the presence of melanized deposits, we next used an established method to generate gnotobiotic cockroaches by rearing them in a sterile environment lacking environmental bacteria (35), which comprise the gut microbiota of B. germanica (36)(37)(38).We then examined the guts of these insects for the prevalence of melanized deposits in the foregut and compared this to the prevalence in guts of control (colony derived) cockroaches (Fig. 3B  and D).Eight of nineteen control samples had melanized deposits in the gut, consistent with our initial observations (Fig. 1).On the other hand, 0/16 guts from gnotobiotic cockroaches had melanized deposits, which was a significantly lower proportion than was observed in the control group (Fisher's exact test, P = 0.0038, Fig. 3B and D).This pattern shows that the presence of melanized deposits in the gut is associated with the gut microbiota and supports the qPCR data linking melanized deposits to a higher density of bacteria in the gut.Additionally, both results support the hypothesis that the melanized deposits may originate from an immune response to increased levels of commensal bacteria in the gut.
We lastly carried out two complementary experiments to determine if the deposits observed in the foregut are indeed composed of melanin.Melanin is known to strongly absorb light at 405 nm (39,40).Direct absorbance measurements on dissolved gut tissues revealed higher absorbance at 405 nm in gut samples with melanized depos its relative to those without (Fig. 4A).This difference was small and may have been confounded by absorbance from other nitrophenol compounds in the gut but was nonetheless statistically significant (Mann-Whitney test, P = 0.041).Furthermore, when we fixed guts with melanized deposits in Carnoy's solution overnight, the deposits exhibited red-orange autofluorescence (Fig. 4B), which has been previously reported for melanin deposits in the insect gut (41).Both of the above observations are strongly indicative that the melanized granules present in the cockroach gut contain melanin, though they are not entirely specific.

Melanin deposits in the cockroach gut enhance Salmonella infection
To determine the effects of the microbiota-induced melanin deposits in infection of the cockroach vector by S. Typhimurium, we fed adult male cockroaches either the wild-type or ΔfimA strain.Twenty-four hours post-infection, we binned guts based on the presence or absence of melanin deposits and we compared the proportion of guts harboring detectable levels of S. Typhimurium (limit of detection = 500 CFU) by culture (Fig. 5).From these data, a significant decrease in the prevalence of infection with the wild-type strain was evident in guts without melanin deposits relative to those that had deposits (Fisher's exact test, P = 0.045).Viable S. Typhimurium remained detectable 24 h postinfection in 75% of guts with melanized deposits, while only 39% of the guts without melanized deposits retained S. Typhimurium at the same time point.Furthermore, we did not see a significant difference in the proportion of guts infected with the ΔfimA strain between those that had melanin deposits and those that did not (Fisher's exact test, P = 0.157), although a similar pattern to the wild-type strain was observed.These results generally indicate that the presence of microbiota-induced melanin deposits in the gut enhances infection of the vector.However, the observation of a similar trend for both the wild-type and ΔfimA strain suggests that the enhancement of infection by melanin deposits cannot be attributed solely to bacterial aggregation on these deposits.Rather, additional positive interactions between the microbiota, melanin deposits, and S. Typhimurium infection that are independent of fimA-mediated aggregation may exist.

DISCUSSION
Our current and prior findings demonstrate that distinct populations of S. Typhimurium exist during infection of the cockroach gut (Fig. 2) (11).Alongside planktonic bacteria in the gut lumen, aggregated populations frequently form using small granular melan ized deposits as substrates (Fig. 2B and C).Based on the new observations reported here, there appears to be some level of specificity to S. Typhimurium aggregation, as aggregates are not formed by E. coli in the cockroach gut (Fig. 2D).Furthermore, S. Typhimurium aggregates require type 1 fimbriae (fimA) to form but not curli fibers (csgA) or the master transcriptional regulator of biofilm formation (csgD) (Fig. 2D), revealing that the aggregates are not canonical biofilms.These findings extend the functional role of fimA in host colonization from mammals to an insect host, as fimA is involved in biofilm formation on gallstones in mice (28).However, the role of bacterial aggregation in colonization of the cockroach gut may not be restricted to S. Typhimurium, as it is still unclear if other types of fimbriae may enable other bacterial pathogens to behave similarly in the cockroach gut.
The melanized granules in the cockroach gut that act as substrates for S. Typhimu rium aggregation show several properties consistent with melanin that are not entirely specific.These include their microscopic appearance, autofluorescence, and absorbance at 405 nm (Fig. 1 and 4).The formation of melanin deposits, or melanization, is a well-studied phenomenon that has been documented in the gut tissue of numerous insect species (32-34, 41, 42), including in the foregut (13,(42)(43)(44).Melanization occurs as an immune response to microbes as well as in response to tissue damage.In our work, melanin deposits in the cockroach gut were not age or starvation dependent (Fig. 1) but were correlated with gut bacterial density and were abrogated in gnotobiotic insects lacking a gut microbiota (Fig. 3).These results together strongly indicate that the gut melanization we observed is an immune response to the gut microbiota, but tissue damage may also play a role that cannot be ruled out by our data alone.An additional limitation of our data is that they do not identify the specific bacteria linked to melanization, which could be further explored through 16S rRNA profiling of the gut microbiota.
It has been previously shown that the German cockroach microbiota confers colonization resistance against ingested E. coli (35).However, effects of the microbiota on different ingested bacteria may vary and may also be context dependent.German cockroaches acquire a highly diverse gut microbiota horizontally from the environment via their diet and conspecific coprophagy (36)(37)(38)45).As such, it is possible that the microbiota may have pleiotropic effects on infection through multiple mechanisms, simultaneously providing both positive and negative regulation.The revelation of aggregated and planktonic populations of S. Typhimurium further confounds under standing the microbiota's role in infection, as distinct commensal species may differently affect the two populations.Indeed, we show that melanin deposits, and by extension the gut microbiota, enhance S. Typhimurium infection of the cockroach gut (Fig. 5).One possible mechanism behind this observation is that melanin deposits act as substrates for fimA-mediated aggregation, and this is necessary for optimal infection.Nonetheless, in our infection assays, both the wild-type and ΔfimA mutant strains of S. Typhimurium showed similar deficiencies in colonization when melanin deposits were not present in the gut, although the deficiency was not statistically significant for the mutant strain.Thus, enhanced infection in guts with melanin deposits cannot be explained solely by bacterial aggregation on these substrates.The correlation between melanin deposits and a higher density of commensal bacteria in the gut points to additional positive interactions between some gut microbiota constituents and S. Typhimurium.For instance, some commensal bacteria that are linked to the formation of melanin deposits may enhance S. Typhimurium infection by producing beneficial metabolites, altering the microenviron ment, or modulating immune responses to the pathogen in the cockroach gut (46)(47)(48)(49).
In summary, higher densities of commensal bacteria in the cockroach gut correlate with the formation of melanized deposits, likely as an immune response.Upon entering the cockroach gut, S. Typhimurium forms aggregates on the surface of these melanized deposits, a process that is dependent on fimA.In the absence of melanized deposits, lower infection rates are observed (Fig. 6).These findings demonstrate a novel mecha nism by which S. Typhimurium interacts with various components of the cockroach gut environment to enhance infection of the vector and underscore the emerging but still underappreciated complexity of this insect-pathogen relationship.We emphasize that this mechanism likely does not operate in isolation.As discussed above, additional effects of different microbiota constituents on infection through mechanisms such as cooperation, competitive exclusion, and regulation of innate immunity should be considered and investigated to better understand the complex relationship between the cockroach gut, its microbiota, and S. Typhimurium infection.Moreover, S. Typhimurium can actively colonize several other insects, such as house flies and leafhoppers (50,51), and it is of interest to determine the extent to which mechanisms are conserved across FIG 6 Summary of results.Higher densities of commensal bacteria in the cockroach gut correlate with the formation of melanized deposits, likely as an immune response.Upon entering the cockroach gut, S. Typhimurium forms aggregates on the surface of melanized deposits, a process which is dependent on fimA.In the absence of melanized deposits, lower infection rates are observed.these hosts, as well as across other enteric pathogens that cockroaches may acquire in nature.

Cockroach colonies
The American Cyanamid Orlando laboratory strain of Blattella germanica was used for all experiments, as in our previous work on bacterial infections (11,12,35).Cockroach colonies were maintained in plastic enclosures at 25 ± 1°C and 40%-45% relative humidity on a 12:12 (L:D) hour photoperiod.The colonies were provided dog chow (Purina, St. Louis, MO, USA) and tap water ad libitum, and were given egg carton harborages for shelter.Mostly adult males were used in experiments in order to preserve females and nymphs for colony propagation and minimize physiological variation due to gonadotropic and developmental cycles.

Observation of melanized deposits in cockroach guts
After starvation of food and water for 1 or 3 d, whole guts were dissected from col ony derived adult males or nymphs and observed through either a stereomicroscope (M165FC with DFC310 FX camera, Leica, Wetzlar, Germany) or compound microscope (Primo Star with Axiocam 208, Zeiss, Oberkochen, Germany) under brightfield.The guts were visually scored for the presence or absence of melanized deposits in the foregut.

Observation of S. Typhimurium aggregates around melanized deposits in cockroach guts
S. Typhimurium was provisioned orally to groups of adult male cockroaches starved of food and water for 3 d to promote consistent experimental feeding as described in our previous work (11,12).In summary, bacterial cultures were grown overnight in LB at 37°C, diluted to OD 600 = 1, then provided to the cockroaches as a sole food source for 30 min.Insects that did not feed during the 30 min period were removed from enclosures and excluded from further analysis.S. Typhimurium strain 14028s expressing mCherry from the plasmid pFPV served as the wild-type control (26,52), as we previously showed it forms aggregates in the cockroach gut (11).In addition, several deletion mutants of the 14028s strain transformed with pFPV-mCherry were tested.These were as follows: ΔfimA, deficient in the major structural subunit of type 1 fimbriae, ΔinvAspiB, a type III secretion system 1/2 double mutant, ΔcsgD, deficient in a major transcriptional regulator of biofilm formation, and ΔcsgA, deficient in major curlin subunit, a critical component of biofilms.
Escherichia coli strain DH10B harboring pPFV-mCherry was also included as a control in the experiments.Three hours post-infection, guts were screened for detection of bacterial aggregates around existing melanized deposits.Whole guts were dissected and placed on micro scope slides for observation using an Axioskop 2 fluorescence microscope with an Axiocam 208 camera (Zeiss).Observations were confined to the foregut (the primary location of melanized deposits).Multiple planes were observed to confirm observed bacteria were coating the deposits as opposed to merely being above or below them and representative images were taken.Guts were scored as positive or negative for bacterial aggregates around melanized deposits and data were analyzed using Fisher's exact test to compare the prevalence of aggregate formation by the different strains.

Quantitation of gut microbiota density by real-time PCR
A semi-quantitative real-time PCR assay (qPCR) was carried out as in our previous work (53) to determine if guts from colony derived cockroaches with melanized deposits and those without harbor different densities of commensal bacteria.DNA was isolated from whole guts using the DNeasy blood and tissue kit (Qiagen, Germantown, MD, USA) according to the manufacturer's protocol.These guts were binned based on the presence or absence of melanized deposits as determined by stereomicroscopy after starvation for 3 d.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 DNA as template, and previously published primers targeting a conserved region of the bacterial 16S rRNA gene (331F/797R, F: 5′-TCCTACGGGAGGCAG CAGT-3′, R: 5′-GGACTACCAGGGTATCTAATCCTGTT-3′) (54).These primers cover 83.1% of bacterial taxa based on estimation using the SILVA TestPrime tool (54), 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 gut samples as a proxy for bacterial density and the data were analyzed by parametric two-tailed t-test as the assumption of normality was met.

Analysis of gnotobiotic cockroaches
German cockroaches acquire their diverse gut microbiota horizontally from the environment via their diet and conspecific coprophagy (36)(37)(38)45).To study the impact the microbiota has on the generation of melanized deposits in the gut, we developed gnotobiotic cockroaches which lack an environmentally acquired gut microbiota using a process established in our lab (35).The first step in the process involves gently removing and surface sterilizing mature oothecae from gravid females using 10% bleach and 70% ethanol.The sterilized oothecae and the nymphs derived from these are then maintained in a sterilized, air-tight container ventilated through a 0.2 µM filter with autoclaved water and dog chow (Purina).To verify our method, we periodically tested gnotobiotic cockroaches by culturing whole insect homogenates on LB agar without antibiotics to rule out contamination of the gnotobiotic colony.Once the gnotobiotic cockroaches molted into adults, these were screened for the presence of melanized deposits in the gut.To do so, we starved both gnotobiotic and control (colony derived) cockroaches of food and water for 3 d.After this, we dissected whole guts and scored them based on the presence or absence of melanized deposits in the foregut using stereomicroscopy as described above.Fisher's exact test was used to compare gnotobiotic and colony derived groups.

Quantitation of melanin in cockroach guts
To further investigate if the melanized deposits observed in the cockroach gut do in fact contain melanin, we carried out two additional experiments.In the first experiment, we adapted a protocol to quantify melanin concentration in cells based on absorbance of light at 405 nm (39,40).Colony derived adult male cockroaches were starved for 3 d.Whole guts were dissected and scored based on the presence or absence of melan ized deposits in the foregut by stereomicroscopy.The guts were then homogenized using a pestle in 500 µL of 1 N NaOH/10% DMSO solution.One hundred microliters of each sample in triplicate was loaded into a 96-well plate and absorbance at 405 nm was measured on a plate reader.Synthetic melanin (Sigma Aldrich, St. Louis, MO, USA) dissolved in the same solution was used as a positive control.A non-parametric one-tailed Mann-Whitney test was used to compare absorbance values in guts with or without melanin deposits as the assumption of normality was not met.
In a second experiment, we followed a process for detecting melanin autofluorescence by fixation in Carnoy's solution (41).A solution of ethanol-chloroform-acetic acid, 6:3:1 [vol/vol] was used to fix three individual whole guts from colony derived males overnight.Autofluorescence was then observed using an Axioskop 2 fluorescence microscope with an Axiocam 208 camera (Zeiss).

Quantitation of Salmonella infections
To examine the impact of melanin deposits in the gut on S. Typhimurium colonization, groups of adult male cockroaches were orally provisioned either wild-type or ΔfimA S. Typhimurium as described above.One day post-infection, insects were collected, whole guts were dissected, and the presence of foregut melanin deposits was noted by stereomicroscopy.Then, the whole guts were homogenized in sterile PBS and plated on LB agar containing 100 µg/mL ampicillin to quantify S. Typhimurium.Fisher's exact test was used to compare the proportion of guts with detectable S. Typhimurium in the presence or absence of melanin deposits.

FIG 2
FIG 2 Aggregate formation by S. Typhimurium on melanized deposits is dependent on fimA but not csgA or csgD.(A) Melanized deposits with no bacterial aggregates (asterisks) 3 h post-infection with mCherry-expressing E. coli.(B and C) Melanized deposits with large aggregates (arrows) of mCherry-expressing S. Typhimurium 3 h post-infection.The image in panel C was processed with background subtraction and planktonic bacteria are also visible.(D) Prevalence of aggregate formation by E. coli, wild-type S. Typhimurium (14028s), and various mutant strains of S. Typhimurium, as determined by fluorescence microscopy 3 h post-infection.Data were analyzed using Fisher's exact test.

FIG 3
FIG 3 Melanized deposits in the cockroach gut are linked to the bacterial microbiota.(A) Bacterial density in guts from control (colony derived) cockroaches that exhibit or do not exhibit melanized deposits as determined by semi-quantitative PCR.Data were analyzed using a two-tailed parametric t-test.(B) Proportion of guts with melanized deposits in control and gnotobiotic adult males as determined by stereomicroscopy.Data were analyzed by Fisher's exact test.(C) Representative micrograph of melanized deposits in the gut of a control (colony derived) cockroach with a normal gut microbiota.Note that the brightfield image from Fig. 1B, which is representative of the same experimental group, is used again in Fig. 3C for side-by-side comparison to the result shown in Fig. 3D.(D) Representative micrograph of a gut from a gnotobiotic cockroach without an environmentally acquired microbiota, which does not contain melanized deposits.Scale bars = 150 µm.

FIG 4 FIG 5
FIG 4 Melanized deposits in the cockroach gut exhibit properties consistent with melanin.(A) Absorbance at 405 nm in gut extracts from control (colony derived) cockroaches with and without melanized deposits.Data were analyzed by one-tailed non-parametric Mann-Whitney test.(B) Autofluorescence of melanized deposits (*) in the cockroach gut after fixation.Scale bar = 30 µm.