1 Introduction

In 2012, Brazil produced more than 12 million tons of chicken meat, and exported approximately 4 million tons, being the world's largest exporter of this product (^{UBABEF, 2013}).

Increasing in production has been possible due to the modification of the environment, making it favorable to poultry production, but these modifications also favor pests and diseases associated with poultry, as the lesser meal worm (LMW), Alphitobius diaperinus (Panzer, 1997) (Coleoptera: Tenebrionidae) (^{O’Connor, 1987}). The large population of the LMW found in the poultries aviaries and natural poultry behavior lead the birds to feed on these insects and instead the chicken food, resulting in reduced development, weight loss and increased time to slaughter. It has been emphasized that LMW may play a potential role as vectors of bacteria, especially Salmonella spp.(Lign.), Campylobacter jejuni (Jones) and Escherichia coli (Theod.), bacteria that cause diseases in both poultry and humans (^{Despins and Axtell, 1995}; ^{Hazeleger et al., 2008}; ^{Chernaki-Leffer et al., 2010}).

Poultry house´s environment associated with the presence of the LMW, make aviaries an extremely favorable site for the presence and development of bacteria, which may be transmitted to humans through meat consumption and affect the meat commercialization and limit the exportation (^{Chernaki-Leffer et al., 2010}; ^{Oliveira et al., 2014}).

Thus, in order to meet the increased demand for chicken meat and maintain quality, measures are necessary to prevent and control insects and pathogens found in poultry houses. Currently, the most common way to LMW control are chemical insecticides, which, although immediately effective, have a low persistence, require constant reapplications that are stress poultry, may leave residues in meat, selecting insect resistant populations and eliminate natural enemies (^{Chernaki-Leffer et al., 2011}).

Beauveria bassiana (Bals) naturally occurs in poultry houses, causing high rates of mortality of LMW (^{Alves et al., 2005}), and laboratory studies have indicated the potential of this fungus as a control agent of the LMW (^{Chernaki-Leffer, 2004}; ^{Rohde et al., 2006).}

Bacteria control can be carried out preventively, with the use of disinfectants derived from ammonia, on poultry house surfaces before installing poultry (^{Bermudez and Stewart-Brown, 2003}). Bacterial control can also be carried out curatively, incorporating antibiotics into feed. It is noteworthy that antibiotics used for birds have the same active ingredients of some formulations used for humans and may cause selection for antimicrobial resistance (^{Padilha, 2004}; ^{Górniak and Spinosa, 2007}).

As alternatives for the control of pathogens in poultry, products of vegetable origin or based on organic acids have been shown to be effective under laboratory (^{Fruet, 2010}; ^{Jaenisch et al., 2010}; ^{De Bona, 2012}) and the use of this control strategy is potentially safer for the environment and less impactful to the conservation of natural enemies of insects and mites, including entomopathogenic fungi.

Thus, the present study aimed to assess the effect of plant extracts and a commercial disinfectant (based on an active organic acid) previously evaluated against some Salmonella spp. serovars on some biological parameters of the fungus B. bassiana.

2 Material and Methods

3 Results and Discussion

Other studies have shown that the hydroalcoholic extract of the jabuticaba plant (M. cauliflora) has significant antimicrobial activity against the bacteria Streptococcus spp. (Ros.) and the fungus Candida albicans (Ber.) (^{Macedo-Costa et al., 2009}; ^{Diniz et al., 2010}), not being found studies with other microorganisms taxonomically closer to the fungus studied here.

A study on the alcoholic extract of P. guajava (guava) at concentrations of 6.5, 12.5, 25 and 50% also demonstrated the inhibitory activity of mycelial growth of Colletotrichum gloeosporoides (Penz.). Although obtained with a fungus of another order, resembles those found in this study regarding the inhibitory action of P. guajava (^{Boneet et al., 2012}).

The extract of the jambolan plant (S. cumini) was the only one that showed no significant inhibitory activity on any of the parameters evaluated. Likewise, ^{Cock (2012)} also showed that methanol extracts of Syzygium australe (Ber) and S. luehmannii (F. Muell) had little fungicidal activity against Aspergillus niger (Tiegh). Although the study used extracts from different species, it is likely that classes of phytochemicals produced by these plants are close because they belong to the same genus (^{Rodrigues et al., 2010}).

The classes of compounds present in the alcoholic extract of S. cumini are generally similar to those extracted from M. cauliflora and P. guajava and include tannins, flavonoids, anthocyanins alkaloids and others, which have known antibacterial activity (^{Loguercio et al., 2005}; ^{Vargas-Alvarez et al., 2006}; ^{De Bona, 2012}). The differences in the inhibition of fungus viability may be related to the concentration and diversity of phytochemicals extracted.

Despite viability being a parameter present in the calculation of the biological index, its weight in the calculation is very small (only 10%), justifying the result of compatibility.

For CFU counts and colony diameter, there was no interaction of extracts. However, the presence of guava extract in the culture medium significantly stimulated conidia production. Despite all of these plants being in the Myrtaceae family, the variation in results may be directly linked to the nutritional status of the plant, as suggested by ^{Mamprim et al. (2013)}.

This variation in the interaction of products and biological parameters of B. bassiana has also been observed for extracts of eucalyptus (Eucalyptus citriodora (Hook)), chinaberry (Melia azedarach (Lin)) and laurel (Laurus nobilis (Lin)), which affected not only the viability and CFU but also significantly increased conidia production (^{Mamprim et al., 2013}).

Increased conidia production in the presence of the aqueous extract of guava may have two distinct explanations. Studies have demonstrated that conidia production can be increased by increasing the nutrient richness, the amount of carbon and nitrogen compounds in culture medium, or specific carbohydrates in the culture medium (^{Rombach, 1989}; ^{Leite et al., 2003}). To corroborate this hypothesis, ^{Mellinger (2006)} and ^{Scoparo (2011)} found polysaccharides in leaves extracts from Phyllanthus niruri (Lin) and Camellia sinensis (Lin).

Moreover, ^{Moino-Junior and Alves (1998)} suggest that fungi, when affected by certain substances, initiate a mechanism of physiological resistance and begin to metabolize compounds that can be used as nutrients, or when suffering from the toxic influence of the medium, initiate a physiological process that increases the production of conidia in a reproductive effort because conidia are considered resistant to adverse conditions. The latter seems quite consistent with the results in our study because the production of conidia was the most affected parameter in the guava extract treatment, and if the change was a result of the addition of nutrients to the medium, most likely other parameters would also be positively influenced.

Despite some parameters being negatively influenced, the Biological Index (^{Rossi-Zalaf et al., 2008}) showed that all aqueous extracts were compatible with the fungus B. bassiana Unioeste 4 (Table 1).

According to ^{Grezzi (2008)}, most chemicals disinfectants are used to control, prevent the growth or destroy microorganisms but are not necessarily effective against some resistant forms, such as spores. Both peracetic acid and hydrogen peroxide, which are the main components of the disinfectant formulation, have oxidizing action, cause protein denaturation and rupture of cell membranes.

It is worth noting that exposure time, temperature and concentration are key factors for the inactivation of microorganisms. Thus, more sensitive microorganisms are inactivated in 5 minutes at a concentration of 100 ppm of some compounds, whereas 500-10000 ppm are necessary for removal of spores, with exposure time of 15 seconds to 30 minutes (^{Rutala, 1997}). This explains the production of conidia in colonies with the 0.5 RC treatment because at this concentration the amount of peracetic acid were 2500 ppm (which is in 500-10000 ppm range). Additionally, the disinfectant was applied 48 hours after fungus inoculation and this period of time may have been sufficient for germination and early fungus development.

Different sensitivity levels of microorganisms to products are related to the concentration and to the substance: activity can change from inhibitory to a biocide with increasing concentration in the medium (^{Ostrosky et al., 2008}). This may explain the effect of the gradual increase in concentrations, the product changed from non-toxic (half the recommended dose (0.5 RC) to moderately toxic (recommended dose (CR)) and reaching toxic at twice the recommended concentration (RC 2).

For the parameters of viability and CFU, values ​​equal to zero were observed, and the interaction of compatibility was determined only by the colony diameter (vegetative growth) and conidia production; this explains why the disinfectant at 0.5 RC and RC resulted in a compatible and moderately toxic classifications, respectively.

Mortality observed in this study for the control (88.6% in pure fungus) is very close to ^{Rohde et al. (2006)}. In isolates selection to A. diaperinus obtained 86.7% to adults in the concentration of 1 × 10⁹ conidia/mL by the same fungus B. bassiana Unioeste 4.

In treatments where there was both reduction of and increased mortality, we suggest two alternative explanations: influence on time to fungal development or insect defense.

At first, we point that the treatments may have negatively influenced the metabolism of the fungus and its development, leading to the need for more time (in days) to colonize the insect; or the treatments may have affected the production or action of enzymes and toxins that act in colonization and death of the host, suggesting the need for a biochemical study to assess which enzymes and toxins may have been affected and how. This is based on ^{Xiao et al. (2012)}, who emphasized that some gene families of B. bassiana developed mechanisms to increase production of proteases and chitinase, mediators of cell growth, regulators of extracellular acidification, degrading cuticular fatty acids and an important class of hydrophobins associated with the ability to adhere to the insect. All of these mechanisms related to fungus colonization may have suffered some negative influence by the peracetic acid at the rate 0.5 RC.

Also, in addition to the mechanisms of colonization, mechanisms of virulence may also have been affected. According to ^{Von Döhren (2004)}, there are some factors related to fungal virulence that affects the immune system of the insect, including the production of a wide variety of compounds that act as immunosuppressant polyketide non-ribosomal, natural peptides, and toxins such as beauvericin and bassianolide. The latter also displays a high antibiotic activity, and its reduction may explain the non-confirmation of mortality by fungi in some cadavers; the action of the fungus may have weakened the immune system of the LMW and opportunistic bacteria may have caused the insect death.

The second hypothesis to explain reduced colonization and virulent is due to the immune system of the insect has had sufficient time to suppress fungal activity. ^{Chouvenc et al. (2009)} showed that the insects' immune system is quite efficient, verifying that in the hemolymph of termites, 24 h after exposure to the fungus Metarhizium anisopliae (Met), the amount of hemocytes had increased significantly. The authors noted the defense capacity of cell aggregation and melanization around the point of infection, in addition to the encapsulation and isolation of nodules, which, according to the authors, intensifies as the fungus takes longer to act.

To high action of the fungus, is it possible the aqueous extract may have enriched the culture medium providing a greater quantity of compounds to be assimilated by fungus and the fungus’ rapid metabolism increased the production of extracellular proteases, and therefore, unlike that observed with peracetic acid, the infectious process is accelerated and increased the percentage of mortality (^{Rombach, 1989}; ^{Leite et al., 2003}; ^{Ito et al., 2007}).

Comparing these hypotheses, it is possible to infer that the insecticidal activity of the fungus was influenced by the addition of the extracts to the medium. Some secondary compounds present in the extracts have insecticidal action, their residues may have remained with the fungus at the time of scraping and these compounds may have affected the immune system of the insect, leaving it weakened and assisting in the infectious processes of the fungus. The results obtained here suggest the need for a study characterizing how the enzymes and toxins are affected by the extracts and products.

Thus, all of the extracts and the Peroxitane at 0.5 RC are compatible with B. bassiana Unioeste 4, although extracts had a low impact on the inoculum potential of the fungus. Also, the insecticidal activity of the fungus to LMW showed that the mortality rates caused by the fungus from culture media with aqueous and alcoholic extracts were not adversely affected but increased when fungus were produced on culture media with aqueous extract of guava. Peracetic acid in the 0.5 RC had negatively influence on the fungus activity, demonstrating the negative action of the disinfectant on the insecticide potential of the fungus. As know, this negative effect here observed does not always mean that the same will occur in the field, but it shows a possibility for this to occur in poultry house (^{Alves et al., 1998a}).

The results of this study demonstrate the possibility of the combined use of plant extracts and B. bassiana in poultry houses. It is important to allow for a period of 48 hours after inoculation of the fungus and treatment with the extracts; this time is required for spore germination and host invasion by the fungus. In aviaries, applications to control bacteria are usually performed preventively, i.e., before poultry installation, and as B. bassiana does not offer any risk, to the poultry, it can be applied before poultry installation, which favors the use of this combined control strategy of bacteria and A. diaperinus.

4 Conclusions

Peracetic acid affect all the fungus parameter while vegetal extracts had none negative effect on the fungus. Even so, all the extracts and Peroxitane at RC 0.5 are compatible with the fungus B. bassiana Unioeste 04, however only the extracts had a low impact on inoculum potential.