The stingless bees (Hymenoptera: Apidae: Meliponini): a review of the current threats to their survival

Pollination by insects is an essential service for the conservation of plant biodiversity, since it increases the rate of fertilization of flowers and therefore increases the quality and quantity of seeds and fruits. In tropical regions, native plants depend heavily on stingless bees as pollinators since before the introduction of the European honey bee Apis mellifera. However, like for many other insect species worldwide, anthropogenic actions represent a true threat to stingless bee populations. In this article, we review the works that evaluated the potential negative impacts of human activities on populations of stingless bees. The studies constitute a strong body of evidence that stingless bee populations are at risk of decline around the world due to threats including habitat loss, agrochemicals, competition for resources, climate change, and the introduction of exotic species, including pathogens.


INTRODUCTION
Worldwide, bees play a fundamental role in the environment by facilitating the pollination of plant species in both natural ecosystems and agricultural crops, which has a direct impact on the yields of fruits and seeds (Nicholls and Altieri 2013). The approximately 20,000 species of bees belong to the Apidae family of the order Hymenoptera. Bees are one of the main pollinator taxa, responsible for the pollination of about 73% of cultivated plants (Santos et al. 2014), which can result in up to 50% higher production of seeds and fruits (Klein et al. 2007). The Apidae family contains several tribes: Apini, Bombini, Euglossini, Meliponini, Xylocopini, and other, lesser known tribes (Arnold et al. 2018). Advanced eusocial bee taxa are found mainly in two groups: highly eusocial bees (including the european honeybee, Apis mellifera Linnaeus Apini and the stingless bees, Meliponini) and the primitively eusocial bumblebees (Bombini).
The stingless bees are considered the most highly eusocial among bees native to the tropics and subtropics, including the American tropics (Neotropics), sub-Saharan African (Afrotropical), and the Indo-Australian (Australasian) regions. About 500 species of Meliponini have been described worldwide, and 80% of them occur in the western hemisphere from Mexico to Argentina, including the Caribbean and Pacific islands (Ayala et al. 2013). They are found in different ecosystems, including rainforest, cloud forest, and arid lands, and at a variety of altitudes, from sea level to 4000 m above sea level in the Andes. Their continental distribution denotes the importance of their ecological role as pollinators of native wild plants and crops (Ayala et al. 2013;Giannini et al. 2015a).
The stingless bees are distinguished from other bees by morphological features of their wings (weak or lacking submarginal crossveins and second recurrent vein in the forewing; Michener 2013). All stingless bees live in colonies that consist of tens to hundreds of thousands of workers, and usually a single queen. At any given time, the number of males present in the colony may range from zero to dozen; males and workers are usually similar in size and appearance, while queens are morphologically distinct (Michener 2013). Stingless bees feed on nectar and pollen collected by foraging workers. The honey made from the nectar is placed in separate pots from pollen.
Stingless bees were the main pollinators of plants on the American continent before the introduction honeybees from Europe. In addition to pollination services, stingless bees provide benefits to humans from their honey and geopropolis, which have demonstrated antioxidant, antimicrobial, anti-inflammatory, wound healing, and anticancer activities, among other properties (Seabrooks and Hu 2017;Jalil et al. 2017; Alvarez-Suarez et al. 2018). However, the stingless bees, like many insects groups worldwide, are threatened by habitat devastation, global climate change, and resource competition by the introduction of others bees such as A. mellifera (Becker et al. 2018;Jaffé et al. 2019;Hung et al. 2019). These factors, together with the loss of local ancestral ecological knowledge of meliponiculture, are the main drivers of decreases in the native stingless bee populations Reyes-González et al. 2020). Studies have also shown that anthropogenic actions such as the inadequate use of pesticides are among the most important factor threatening insect biodiversity in general (Dudley and Alexander 2017). This decline in populations of pollinator species is of concern because it could lead to a parallel decline in plant species (Biesmeijer et al. 2006;Lima et al. 2016). Other threats that are likely just as important and are increasingly clearly linked to pollinator decline include the spillover of pests and diseases from A. mellifera to stingless bees (Purkiss and Lach 2019;Tapia-González et al. 2019;Teixeira et al. 2020). Although more studies are needed, it is important to increase awareness to protect these bees. We therefore review each major category of factors threatening stingless bees: chemical pesticides, transgenic crops, deforestation, biotic factors, and climate change. We close the review with some perspectives for better understanding these threats and options to address them.

Insecticides
The application of pesticides to crops kills not only the target pest, but may also have negative effects on beneficial insect which include the pest's predators and parasitoids, as well as pollinators such as beetles, butterflies, and many different species of bees (Apidae) (DiBartolomeis et al. 2019; Sánchez-Bayo and Wyckhuys 2019). Pesticides represent a serious threat to the survival of stingless bees, and there is currently strong scientific evidence demonstrating the toxicological effects of pesticides on Meliponini species (Lima et al. 2016;Tomé et al. 2017). There are still few studies demonstrating stingless bees' level of exposure to pesticides in their habitat; 1 3 however, some pioneering work has been done in a meliponary where foragers of Scaptotrigona aff. depilis were shown to contain neonicotinoid pesticides (Rosa et al. 2015). This led the authors to conclude that S. aff. depilis is a potential indicator of environmental pesticide contamination. Furthermore, a meta-analysis demonstrated that stingless bees are more sensitive to pesticides than bumblebees, solitary bees, or honey bees (Arena and Sgolastra 2014). Over the last decade, much information has been generated in this respect, by assessing the toxicity of organophosphorus, neonicotinoid, pyrethroid, and other classes of pesticides: lethal effects have often been observed when used according to their respective labels, as well as tissue and organ malformations and changes in social behavior at sublethal doses (Table I). Crops including orange, lemon, apple, and papaya often use toxic bait formulations to manage the economically important fruit fly pest species, but it was recently demonstrated that toxic baits with the active ingredient malathion (Success®, Gelsura®) lead to mortality of worker bees of stingless bee species (Padilha et al. 2019) (Table I).
Most studies of pesticides' lethal effects on stingless bees have focused on adults, but there is experimental evidence of damage to larval health after ingestion of pollen and nectar contaminated with neonicotinoids, organophosphates, and other pesticides. The exposure of larvae of the stingless bee Scaptotrigona bipunctata to different doses of chlorpyrifos (Organophosphorus) showed reduced survival probability, likely because they developed into lighter, smaller, and deformed adult workers (Dorneles et al. 2021). Exposure to neonicotinoid products during the larval stage has also been shown to cause brain alterations (Miotelo et al. 2021). A study of Melipona quadrifasciata anthidioides exposed to the neonicotinoid imidacloprid from larval stages through adulthood showed a survival rate just above 50%; furthermore, imidacloprid exposure impaired the walking behavior of surviving adults, which was attributed to negative effects on the development of mushroom bodies (MBs) in the brain (Tomé et al. 2012). Similar findings of harmful effects, including the alteration of caste differentiation, impairment of social behavior, and increased asymmetry in individuals exposed to organophosphorus, carbamate, and pyrethroid pesticides during the larval stage, have been reported in different stingless bee species (Dos Santos et al. 2016a;Rosa et al. 2016;Prado-Silva et al. 2018;Boff et al. 2018).
Currently, several agrochemical products are marketed as safe for non-target insects or as reduced-risk insecticides, but toxicological trials using these pesticides and biopesticides have not always supported these designations. For example, when the toxicity of chlorantraniliprole, a pesticide labeled as a reduced-risk insecticide at recommended label concentrations, was assessed in two species of stingless bees, it impaired flight take-off in worker bees of both species (Partamona helleri and Scaptotrigona xanthotrica), which could compromise foraging activity and potentially reduce colony survival (Tomé et al. 2015a). This result challenges the claims of nontarget safety of such insecticides, particularly for native pollinator species. Further studies, using up-to-date protocols, are needed to better understand the toxicological effects of these chemical pesticides on populations of wild bees (Dorigo et al. 2019;Botina et al. 2020).

Herbicides, fungicides, and fertilizers
Herbicides were developed to eliminate undesirable grasses and other pest plants in crops or in water bodies, and although these compounds are designed to interfere with the metabolic processes of plants, the active compounds may also be toxic to bees. Changes in the expression of isoenzymes were assessed in T. angustula and T. fiebrigi under oral exposure to the herbicides Sanson (R) 40SC (nicosulfuron) and 200 Gramoxone (R) (paraquat) (Fermino et al. 2011). Nicosulfuron at 75% led to partial inhibition of esterases, and Paraquat led to 10% and 1% complete inhibition of esterases in the two species, respectively, while the superoxide dismutase isoenzyme increased in relative activity in both bee species at 10% of Paraquat herbicide, making these enzymes potentially useful bioindicators  of these herbicides. Such enzymatic alterations might contribute to explain the physiological and behavioral impairments observed upon exposure to herbicides, including the controversial glyphosate as well as mesotrione and atrazine (Table II). Fungicides are another agrochemical product that in recent research have demonstrated toxicological effects against stingless bees, causing alterations in the midgut and morphological signs of cell death as well as reduced survival rate (Table II). However, there are still few studies on the toxicological effects of fungicides and heavy metal-containing leaf fertilizers, so more investigation is needed to assess the impact of these chemicals on wild bees. Fertilizers are important agrochemicals used in extensive agriculture and orchards to improve plant growth and increase yields, but like others agrochemicals, can have adverse impacts on the environment. Works focused on the toxic effects of fertilizers on stingless bee species have shown reduced respiration rate upon oral exposure, compromising their survival (Table II).

Biopesticides
There is currently a global drive to decrease the environmental impact of synthetic chemicals, which has led to the increased use of products based on microorganisms or metabolites over the past few decades. This trend is particularly strong in organic agriculture, because such products are assumed to be more selective against the target pest or disease, innocuous to human health, and  (Table III). Work related to the effects of botanical pesticides was performed by Xavier et al. (2010) on adult Tetragonisca angustula and Nannotrigona testaceicornis. Their findings showed that the most toxic botanical insecticide to T. angustula was citronella oil and the least toxic was andiroba oil. Eucalyptus oil and neem oil presented intermediate toxicity to T. angustula, and garlic extract and rotenone were non-toxic to T. angustula. None of these botanical insecticides were toxic to N. testaceicornis. Another botanical bioinsecticide on the market is based on the bioactive molecule azadirachtin, a metabolite derived from the Indian neem tree (Azadirachta indica) which has shown high efficacy against arthropod pest species ( Barbosa et al. 2015). In 2015, initial studies of azadirachtin toxicity to important native pollinators including Partamona helleri and Scaptotrigona xanthotrica showed harmful effect when used as recommended on the label. Individual flight take-off of worker bees was impaired (Tomé et al. 2015a), and P. helleri adult workers and queens showed behavioral alterations and deformations (Bernardes et al. 2017. In the last decade, some biological pesticides formulated from bacterial metabolites are recognized as biopesticides or reduced-risk pesticides. One the most popular insecticides used today is Spinosad, a bioinsecticide made from spinosyn metabolites produced by the actinobacteria Saccharopolyspora spinosa. It is an agonist of the nicotinic acethylcholine receptors and interferes with aminobutiric acid (GABA) receptors in the nervous system (Tomé et al. 2015b). Previous studies of Spinosad product GF-120, a toxic commercial bait used in the management of fruit flies, evaluated the toxic effects on the stingless bee Plebeia moureana (Sánchez et al. 2012). This preliminary research showed that using the recommended label concentration mixed with food did not discourage engaged foragers, and foraging behavior was not significantly impaired over time by the continuous collection of contaminated food. However, although spinosad was initially considered a safe bioinsecticide, later studies have shown otherwise in others species. In 2014, a study showed that Trigona fulviventris and Scaptotrigona mexicana clearly avoid food sources baited with GF-120 and. S. mexicana colony sizes decline (Gómez-Escobar et al. 2014, 2018. Years later, other works have shown elevated toxicity of spinosad against these stingless bee species, causing morphological and behavioral alterations, among other harmful effects (Table III).
Another biomolecule with potential insecticidal properties called Abamectin is a metabolite of Streptomyces avermitilis, belonging to the Avermectin class. This biochemical compound is extremely toxic to humans (0.002 mg/kg) and other vertebrates as well as invertebrates (Prado et al. 2020). A study of abamectin verified its bioaccumulation in tissues of adult forager bees of Melipona scutellaris after oral and topical exposure, and mortality was higher from oral than topical exposure (Table III) These studies challenge the common perception of non-target safety of these bioinsecticides, particularly for stingless bee species (Table III).

TRANSGENIC CROPS
The demand for food has increased with the expanding global population, increasing the need to produce crops more efficiently, and with improved quality traits such as disease resistance, prolonged shelf-life, and drought resistance. This has given rise to genetically modified plants, a technology that purports to help solve this growing issue (Kamle et al. 2017). However, like many technological tools, it has limitations. For example, the expression of Bt toxins against pests requires applications of herbicides such as glyphosate (Hardee et al. 2001). One of the most common traits in transgenic maize and cotton is the expression of the Cry1F and Cry2Aa proteins from the bacterium Bacillus thuringiensis, which are toxic against Lepidopteran, Hemipteran, and Dipteran insect pests (Seide et al. 2018). Currently, there are few works evaluating the risks posed by transgenic crops harboring these insecticidal proteins on stingless bee species. Lima et al. (2013) assessed the effects of ingestion of the Cry1Ac protein on larvae of Trigona spinipes. They found a slight increase in larval mortality, and exposure did not cause impaired development of worker larvae. They mention that according to their results, Cry1Ac toxin is unlikely to be harmful to stingless bee larvae under field conditions. However, a later study conducted by feeding larvae of M. quadrifasciata with Cry1F and Cry2Aa contaminated food demonstrated delayed development of the bee larvae (Seide et al. 2018). Still, further investigations are needed since there are not enough studies to conclude on the toxicity of Cry toxins in stingless bees.

DEFORESTATION
Habitat loss by conversion to intensive agriculture is a main driver of declining entomofauna (Ceballos et al. 2017). Disturbance and habitat fragmentation caused by human activities had demonstrated negative effects on nest occurrence and species diversity of stingless bee species in Kenya (Kiatoko et al. 2017). It was recently shown that land use change increases the amount of toxic metals and metalloids that are bioaccumulated by T. angustula (Barbosa et al. 2021). A study by Brown and Albrecht (2001) demonstrated that the species richness of stingless bees, which ranged from 1 to 5 species, is inversely related to distance to the forest and directly related to percent forest cover. Their study showed that in tropical forest sites, bee community composition in even the smallest forest patches contained a much larger proportion of the local apifauna compared to deforested countryside just a few hundred meters from the forest. In another set of studies, stingless bee abundance decreased with increasing distance to the nearest forest patch, and their species richness was positively correlated with the proportion of forest cover (Brosi et al. 2007(Brosi et al. , 2008. Negative effects of landscape deforestation and anthropization on richness and abundance of native bees have been demonstrated in countries including Malaysia, Brazil, Costa Rica and Mexico (Samejima et al. 2004;Venturieri 2009;Brosi 2009;Cairns et al. 2005). For example, Samejima et al. 2004 showed that since stingless bees nest only in large trees, nest density was positively correlated with the density of large trees. In Mexico, communities with the greatest anthropogenic disturbance had lower overall species richness of stingless bees and a higher degree of dominance of the Africanized honeybee (A. mellifera) (Cairns et al. 2005). It is now known that the larger body size of honeybees provides them a strong advantage over the far smaller stingless bees, as maximum foraging and recruitment distances are significantly greater for larger foragers (Araújo et al. 2004;Kuhn-Neto et al. 2009). In addition, stingless bee species with larger body size were found in areas with higher degrees of deforestation, while smaller bee species were found in areas with less deforestation, supporting the hypotheses that small stingless bee species might be more negatively affected by deforestation (Mayes et al. 2019). It was also recently demonstrated that thermal stress increases with flight distance in M. subnitida nectar foragers, with their body temperatures approaching their critical thermal maxima, which might constrain their activity and survival (Souza-Junior et al. 2020). A study in Costa Rica showed that stingless bee species richness and abundance are strongly related to forest cover; species with the widest dietary niche breadth are less likely to persist in sites with less forest (Brosi 2009).). In this sense, subordinant wild bee species that are displaced by more dominant species may be particularly dependent on the larger or more diversified resource pool of sites with more forest cover, such that the loss of forest cover may affect the meliponine community composition (Brosi, 2009).Similar work was developed in Brazil, and showed that the presence of stingless bee species has strong positive correlation with forest and a strong negative correlation with pastures (Arena et al. 2018). Lichtenberg et al. (2017) suggests that land use change may disassemble bee communities through different mechanisms in temperate and tropical areas, and that community 1 3 processes such as competition can be important regulators of community disassembly under land use change. Studies performed on populations of stingless bees in the Brazilian Amazon reinforce the conclusion that anthropogenic activities such as deforestation lead to the decline or absence of Melipona species in degraded areas. (Venturieri 2009;Brown and de Oliveira 2014;Pioker-Hara et al. 2014).

DISEASES AND PESTS
In the 1980s, the phorid fly Melaloncha sinistra was first reported to parasitize workers of Nannotrigona postica. Phorid flies were found at the nest entrance, where they laid eggs on foraging bees, and examining samples showed that up to 37% of workers were parasitized (Simões et al. 1980). Male stingless bees often aggregate outside of nests in order to inseminate emerging virgin queens (Brown 1997), and phorid flies take advantage of this behavior to parasitize their host (Table IV). Another insect parasite of the stingless bees is the mantispid Plega hagenella. This pest was found extensively infesting two nests of Melipona subnitida. Mantispid larvae of P. hagenella develop inside closed brood combs, where they fed on bee larvae or pupae; when they emerge as adults, they are removed from the nest unharmed by the bees (Maia-Silva et al. 2013). Aethina tumida (Coleoptera), also known as the small hive beetle, is a parasite and scavenger of honey bee colonies (Apis mellifera); some scientific notes report the infestation of stingless bee hives by the small hive beetle, such that this beetle may represent a real threat for them (Table IV).
In addition to insect parasites, there are several arachnids that parasitize bees, including several different genera of mites that have been reported infesting stingless bee hives (Table IV). The first report of parasitic mites in stingless bee colonies in Mexico was in a colony of Melipona colimana, and the mite was identified as Pyemotes tritici (Acari: Pyemotidae); the researchers hypothesized that the mites were likely introduced into the colony by infested passalid beetles or ants (Macías-Macías and Otero-Colina 2004).
The introduction of managed bee species into other habitats has allowed the arrival of pests or diseases that spill over into native species, causing harmful effects on populations and putting endemic organisms at risk (Graystock et al. 2016). The wax moth Achroia grisella, a pest of Apis mellifera, was found infesting weak colonies of Melipona bicolor and M. quadrifasciata anthidioides (Cepeda-Aponte et al. 2002). Moth larvae of all developmental stages were observed in the nests, and were extracted from the hives using light to attract them and liquid paraffin placed over the colony to trap them (Cepeda-Aponte et al. 2002). This was the first report of a pest of A. mellifera affecting stingless bees.
Viral diseases are another important health problem for bee species. In particular, a few years ago, a first case of viral infection was reported with the identification of the acute bee paralysis virus in Melipona scutellaris (Ueira-Vieira et al. 2015); this virus originally affects honeybees. The route or mechanism of infection by different pathogens is through shared foraging areas between stingless bee species and A. mellifera (Guimarães-Cestaro et al. 2020). Currently, some surveys have shown other viruses affecting different species of stingless bees (Table IV). To elucidate the impact of viruses on stingless bee population, high-throughput sequencing was used to characterize the virome of Melipona quadrifasciata using healthy bees (symptomless) and unhealthy bees (symptoms such as tremors and paralysis). The results showed that unhealthy bees had higher viral load of viruses of the Dicistroviridae family, which includes common pathogenic honeybee viruses, as well as Parvoviridae and Circoviridae, which have never been reported as entomopathogenic (Caesar et al. 2019).
Given the interest in identifying pathogens that could affect populations of stingless bees, Nunes-Silva et al. (2016) carried out at screening for microsporidian and protozoan parasites (Nosematidae, Leishmaniinae and Neogregarinorida) that have been previously reported as pathogenic 1 3 to honeybees. Their findings showed that of the six different species tested, none was positive for Leishmaniinae or Nosematidae parasites, but the protozoan Apicystis bombi (Neogregarinorida) was found in two stingless bee species, Plebeia emerina and Tetragonisca fiebrigi. Although Leishmaniinae and Nosematidae parasites were not found, years later the microsporidium Nosema ceranae (Nosematidae) was found parasitizing species of stingless bees (Table IV). Presently, there are only two works related to brood bacterial infection causing mortality of stingless bee larvae (Table IV). The mosquito entomopathogenic bacterium Lysinibacillus sphaericus was found to cause brood losses in the Australian stingless bee Tetragonula carbonaria, confirming the disease's causal agent by pathogenicity experiments (Shanks et al. 2017), even though bioassays performed decades ago showed that L. sphaericus is not pathogenic to A. mellifera (Davidson et al. 1977).  Brown (1997) Finally, with respect to fungal diseases affecting populations of the stingless bees, there is no published evidence of natural infection. By evaluating the effect of three commercial formulations of entomopathogenic fungi (Isaria fumosorosea, Beauveria bassiana, and Metarhizium anisopliae) on three species of stingless bees (Tetragonisca angustula, Melipona beecheii, and Scaptotrigona mexicana), Toledo-Hernández et al. (2016) found moderate to high susceptibility (38.9-94.2% mortality) to entomopathogenic fungi; however, field surveys are necessary to support this finding. Even though the aforementioned pathogens have been detected in workers or in the beehive, most of the available data concern only the occurrence of the pathogens and did not examine in detail their effects on bees. Thus, there is currently a near complete lack of data on the detrimental effects on colony development or mortality in populations of stingless bees.

COMPETITION FOR RESOURCES
Species of stingless bees are known to compete for food resources such as pollen and nectar flowers, and studies of aggressive behavior of stingless bees have been published since the last century. For example, Johnson and Hubbell (1974) observed the behavior of different species of stingless bees on flowers and artificial baits that varied in their sucrose concentration. The larger Trigona silvestriana (10 mm) used aggressive behavior to displace other smaller species of stingless bees T. corvina, T. testacea, T. mexicana (7 mm), and the T. latitarsus (4 to 5 mm), especially from the bait with the highest sucrose concentration (2.4 M), as long as no more than 50 to 60 bees were on the bait. Levels of aggression increased in intensity when the species T. silvestriana and T. corvina visited the baits and flowers at the same time, which sometimes led to lethal conflicts over the resource. When T. corvina encountered conspecifics from other colonies on the resource, battles lasting up to 2 days were recorded, with up to 1800 dead bees found. They mention that the observed aggression is not influenced by baits, since similar behavior and levels of aggression were observed on natural food sources, especially at nectar-and pollenrich inflorescences. Years later, the same authors reported similar aggressive behavior during competition over bait among species of the genus Trigona; the species that showed the strongest interspecific aggression by means of threats, chases, or bumps was T. fuscipennis, while T. fulviventris had the highest conspecific aggression among workers (Hubbell and Johnson 1978).
While strongly aggressive competition for pollen and nectar resources has been demonstrated among native stingless bees, human activity has introduced other, larger competitors that may leave stingless bees at a disadvantage. Pioneering works related to the food competition between African and neotropical bees were performed by Roubik (1978), who found that after the introduction of Africanized honey bee (Apis mellifera) hives near flower patches, native Melipona species became less abundant or harvested less resources such as pollen, which may lead into their population decline. Roubik (1980) demonstrated that the combined advantages of large forager size and large colony size provide Africanized honeybee colonies superior competitive ability over the stingless bees at rich and compact resources. Later, in 1983, the same author found that Africanized honey bee colonies had no significant effect on native bee colonies; however, they conclude that the short duration of the study (1 month) and small African populations (100,000 to 150,000 bees in total in the area) were insufficient to evidence the perturbation of native bee communities (Roubik 1983). Later, Ramalho et al. (1990) would report that the stingless bees use the same pollen and nectar resources as Africanized honeybees (A. mellifera) in neotropical regions including Brazil, Mexico, Panama, and other countries. Their findings concerning bee food sources showed that 50 plant species provide resources to both stingless bees and honeybees; this leads to competition for foraging sources. Due to their smaller body size and smaller colonies, stingless bees are at a disadvantage, and scarcity of food such as pollen significantly reduces brood production in stingless bee colonies (Maia-Silva et al.

3
2016). Competition for resources such as pollen and nectar of the Myrtaceae, Melastomataceae, and Asteraceae families of plants were evidenced between A. mellifera and stingless bee species in studies by Wilms et al. (1996) and Wilms and Wiechers (1997). In addition, the past century of human activity has introduced other species of bumblebees (Bombus sp.), the alfalfa leafcutter bee Megachile rotundata, and various other solitary species (Goulson 2003). Research in past decades has called for experimental studies to determine whether the presence of exotic bee species reduces the populations of local wild bees in the long term (Goulson 2003). It should be noted that this competition could be more difficult to detect in floristically rich habitats because the native bees changes their foraging time or floral species, compensating behaviorally for competition (Roubik 2009). Now, there is ever increasing evidence that honey bee density has detrimental effects on wild pollinators, such as decreasing their flower visitation, reproductive success, abundance, and species richness (Garibaldi et al. 2017;Angelella et al. 2021).
In conclusion, it is overall clear that there are potential negative impacts of non-native bees, both in terms of disease spillover and in competition with native bees for nutritive resources and nesting sites (Russo 2016;Mallinger et al. 2017;Agüero et al. 2018;Hung et al. 2019).

CLIMATE CHANGE
Climate change is an important abiotic factor associated with the decline of worldwide bee populations (Soroye et al. 2020;Giannini et al. 2020). There are some studies of the effects of temperature on the winter diapause of the stingless bee. The diapause phenomenon is based on a temporary interruption of brood cell building by workers and egg-laying by the queen, and some species of the Plebeia genus are known to exhibit reproductive diapause (Dos Santos et al. 2015). Experimental works carried out by Dos Santos found that Plebeia droryana enter into diapause between 8 and 10 °C in a region with a median winter temperature of 10.1 °C. However, according to environmental prediction models, the temperature may increase in the coming decades, making climate change an important factor that could disrupt the diapause cycle of P. droryana at temperatures between 16 and 22 °C (Dos Santos et al. 2015Santos et al. , 2016b. These findings have important implications for ecosystem services. Climate change has already been observed or predicted to modify the phenology and behavior of insect pollinators of agricultural importance, including bees, and bees may be displaced from suitable habitat areas, resulting in reduced pollination success in wild plants and economically important crops (Giannini et al. 2012). In 2018, experimental bioassays revealed a strong effect of external temperature on immature stingless bees between 28 and 30 °C, causing an increase in mortality of the Melipona interrupta brood (Becker et al. 2018). On the other hand, a recent landscape genomics study (Jaffé et al. 2019) was carried out to determine the influence of habitat degradation and climate change on the genetic diversity of populations of M. sunitida distributed across northeastern Brazil, finding that the populations showed a clinal change in genetic structure across the distribution range of sampled area, with four identifiable genetic clusters. In addition, there were genomic signatures of adaptation to temperature, precipitation, and forest cover, as well as latitudinal and altitudinal patterns in the spatial distribution of adaptive genetic variation. According to these findings, the authors concluded that the knowledge obtained could help guide future conservation actions, such as preserving highland and lowland populations separately and conserving or restoring foothill and mountain forests in order to prevent the extinction of this important pollinator (Jaffé et al. 2019). In addition to the works mentioned, there are different studies that relate environmental parameters such as temperature and precipitation to the foraging behavior and honey production of species of the genus Melipona, whose flight activity decreases below 21 °C and above 28 °C, and which have lower honey production in warmer years (Souza et al. 2006;Koffler et al. 2015).

CONCLUSIONS AND PERSPECTIVES
The studies on stingless bee populations constitute a strong body of evidence that anthropogenic activities are the main drivers that threaten the survival of these wild bee species. Habitat loss and conversion to intensive agriculture, agrochemical pollution, biological factors such as competition for resources and pathogens linked to the introduction of exotic species, and climate change all compromise their survival and constitute an extinction threat to providers of important ecosystem services. Pesticides are the most used in the conventional agricultural, and although the lower doses evaluated in some studies did not cause direct mortality of wild bees, they did have harmful sublethal effects on both adults and larvae, such as physiological, morphological, and behavioral impairments, which may ultimately decrease their fitness. The use of botanical and synthetic pesticides in crop fields alters the flower visitation rates of pollinator bees, which could significantly affect yield (Brittain et al. 2010;Tschoeke et al. 2019). However, there are currently few studies on the levels of pesticides to which stingless bees are exposed to in their habitats; further work in this area will help to better understand the risk caused by pesticides on stingless bee population.
In order to prevent the extinction of native pollinators such as the stingless bees, governments need to develop initiatives to establish biological corridors. Such actions could be crucial to conserving ecosystems that provide services by buffering the joint effects of habitat loss and climate change. According to different studies, protecting conserved areas and restoring disturbed areas could improve habitat connectivity and safeguard ecosystem services, which could facilitate the dispersal and establishment of wild bees during distribution shifts (Giannini et al. 2015b). Promoting urban gardens could be an option to offer food resources for species that inhabit urban areas. This technique has been shown to promote foraging by the Australian stingless bee Tetragonula carbonaria even more than natural habitats and plantings, increasing resource availability and potentially enhancing bee fitness (Kaluza et al. 2016). If we are to protect wild bee populations from extinction, it will be essential to develop policies and research related to the protection of the ecological areas of these important pollinators around the world.

AUTHOR CONTRIBUTION
ETH conceived this review article and wrote the manuscript draft; GPC, VMHV, CCL, JTJ, YRR, and RLR participated in the revision, read, and approved the final manuscript.

DATA AVAILABILITY
The datasets analyzed during the current study are available from the corresponding author on reasonable request.

CODE AVAILABILITY
Not applicable.

DECLARATIONS
Ethics approval Not applicable.
Consent to participate All authors are agree to participate.
Consent for publication All authors are agree for publication of this article.

Conflict of interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.