Introduction

Harvester ants, which are granivores, employ foraging routes to harvest seeds on the soils surface, often from or near the parent plant1. Harvester ant species, tend to harvest near the trunk of their foraging trails which are, in turn, shaped by seed distribution and availability, disturbance frequency, and inter/intra-species interactions2. Harvester ants have seed preferences based on relative seed quantity, size/shape, and nutritional value of the seeds3. Harvester ant seed foraging is not limited to natural areas and may occur in agricultural area where seed preferences may benefit or harm crop production. Although harvester ants are known to remove weed seeds, their seed preferences may also include the consumption of crop seeds4,5,6,7,8. Ant removal of crop seeds and vegetation may cause economic loss, especially if the crop is situated within areas of high colony density9,10,11,12,13. Harvester ants, in particular, are frequently found in agricultural regions within and near to active crop fields and pasture, and may have a significant impact on the local plant community by the eradication of vegetation surrounding their nest entrance and seed collection11. Furthermore, seed predations behaviour affects the distribution and abundance of invasive exotic plants7. Harvester ants increase seed dispersal and seed protection, and provide nutrients that increase seedling survival of the desert plants. In addition, ants provide soil aeration through the creation of galleries and chambers, mix deep and upper layers of soil, and incorporate organic refuse into the soil14. Seed or seedling mortality in tropical forests induced by microbial pathogen infection8. Ants may modify the seed coat through mechanical and chemical means, making it more susceptible to water uptake and germination15. Furthermore, ants aerate the soil by creating tunnels and incorporating organic waste into the soil16. Ants play an important part in plant community dynamics as seed dispersers, predators, or both. These harvester ants seek for seeds in vegetation and plants before transporting them back to their colonies. A number of harvester ants collect the small seeds of many herbaceous plants as a source of food supply17. Harvester ants frequently store seeds in their nests for up to a year to meet the requirements of the colony18. The seeds are stored in granaries in underground nests in soil without losing their germination ability. It is possible that harvester ants, like fungus-cultivating ants, are protected from microbial damage by chemicals secreted by different glands in ants19. Metapleural glands (also called metasternal or metathoracic glands) are secretory glands that were considered unique to ants and basal in the evolutionary history of ants is responsible for production of antimicrobial compounds20,21, they are in charge of producing an antimicrobial fluid, which is stored in a reservoir on the ant’s backs22. These reservoirs, also known as bullas, vary in size depending on ant species and castes within the same species. Ants can groom secretions from the bulla onto the surface of their carapace22. This prevents the growth of bacterial and fungal spores on the ants and within their nests23. However, it is thought to be an essential component in an ant’s immunity against parasites. Recent research indicates that these ants may simply use other anti-parasite defenses like increased grooming and venom24. Since most male ants lack metapleural glands, they must rely on fluids provided by other workers24. Seed-cleaning behaviour is a fascinating aspect of ant-plant interactions, where ants remove elaiosome, seed husk, fungi, and other substances from seeds, enhancing their chances of germination and survival25,27. However ant seed-cleaning behaviour is important in seed germination and not all ant species have metapleural glands28. An elaiosome is a fleshy, oil-rich structure attached to the seeds of some plants, acting as a food source to attract ants, which then disperse the seeds by carrying them to their nests28. Elaiosome detachment and seed scarification have been linked to favourable25 negative30 and neutral impacts on seed germination31. The evolutionary adoption of seed-cleaning behaviour by myrmecochory eventually contributes to seed survival. In several plant species found in the Brazilian grassland, ants-induced mechanical and chemical variables facilitate seed survival and germination31. However, only a few studies have shown the seed manipulation behaviour and anti-microbial impact of the Trichomyrmex ant genus on seeds. The genus, Trichomyrmex Mayr, 1865 is a small genus included in the subfamily of Myrmicinae. The Palearctic, Afrotropical, Nearctic, Neotropical, Oriental, and Australian regions are home to the genus habitats32. Some species of Trichomyrmex nest in the ground, beneath stones, in decaying wood, and sometimes in termitaries. Trichomyrmex abyssinicus (Forel, 1894) and Trichomyrmex chobauti (Bingham, 1903) are frugivorous ant species and Trichomyrmex destructor (Jerdon, 1851), Trichomyrmex mayri (Forel, 1902) are the two granivorous from Trichomyrmex species21. Furthermore, Trichomyrmex scabriceps is important ants are commonly available in southern region of India. Based on the above facts in view and granivorous habit of the Trichomyrmex ants genus we examine the seed cleaning behaviour of Trichomyrmex scabriceps as well as role of exocrine gland secretion of worker ants acts as seed protectants through seed germination and microbial load test.

Results

Ethogram of worker ants of Trichomyrmex scabriceps

A systematic inventory listing and describing the behaviors shown by the worker ants described (Fig. 1). The behaviors are divided into two categories i.e., extra-nidal behaviors and intra-nidal behaviors. Extra-nidal behavior comprises behavior exhibited by the worker ants outside the nest. This includes scouting, harvesting of seeds and foraging. Initially, the scouting ants come out of the nest in search of the seeds and suitable foraging site of that particular day. This is followed by the worker ants foraging for seeds by climbing the plants and plucking the seeds from the ear head using their mandibles. The cleaning behaviors are carried out in several bounds and are interrupted by self-grooming, allo-grooming, and wall cleaning behavior. Apart from this behavior, they also exhibit trophallaxis and antennation inside the nest.

Fig. 1
figure 1

Ethogram of worker ants of Trichomyrmex scabriceps both in extra-nidal and intra-nidal behavior in the lab nest (Formicarium).

Full size image

Seed cleaning behavior of the worker ants of Trichomyrmex scabrices

The worker ants exhibit the seed cleaning behavior which was identified by constructing an ethogram. The distinct behaviors associated with seed cleaning. The ants also showed other behaviors while cleaning seeds in the granary such as abdominal waggling, self-grooming and allo-grooming etc. The frequency of seed-cleaning behavior exhibited by the worker ants was recorded for a sampling duration of one hour (Table 1). The prominent behavior exhibited by these worker ants was the storage of seeds inside the test tubes, which were placed inside the lab nest and the ants used them as granaries. The ants while working in the granary exhibited a high frequency of self-grooming. However, the antennation behavior that involves communication through antennal contact between ants was observed as the behavior with second highest frequency after self-grooming (Table 2). The seed cleaning behavior was the third most frequent behavior exhibited by the ants. This was followed by wall cleaning, allo-grooming and abdominal waggling. During the process of seed cleaning, ants use their mandibles to remove the husk of seeds and frequently clean the seeds whenever they come in contact with the seeds in the granary.

Table 1 Different seed cleaning and other behaviour exhibited by Trichomyrmex scabriceps.
Full size table
Table 2 Frequency of seed cleaning and other behaviour exhibited by Trichomyrmex scabriceps worker ants (One-way ANOVA with post-hoc Tukey’s test).
Full size table

Variation in seed morphology

The morphology of collected seeds was changed in the nest due to seed cleaning behaviour by ants. Documentation exists of seed morphology observations made both prior and after ant harvesting. It was found that the seeds’ husk was entirely removed by the ants prior to their storage in the granary (Fig. 2). The husk remnants were removed inside the nest and found deposited at the nest entrance by the non-foraging ants. It was observed that the seeds that were sampled from the ants during transportation to the nest had the husk intact.

Fig. 2
figure 2

Different seeds collected by ants in field and variation in morphology seeds before and after collected by the ants (a) Before collection (Eleusine coracana seeds) (b) After collection (Eleusine coracana seeds), (c) Before collection (Panicum sumatrense seeds) d. After collection (Panicum sumatrense seeds).

Full size image

To assess the role of seed cleaning and exocrine gland secretions

The role of seed cleaning and the chemical secretions of worker ants were investigated. This was assessed by using both germination and microbial load test.

Germination test

Among the four groups of seeds, the nest seeds that were collected by ants and stored in the nest granary alone exhibited a highest percentage of germination (71.67 ± 4.41). The plant seeds, seeds carried by the ants and ground seeds were (26.67 ± 1.66, 41.67 ± 1.67 and 23.33 ± 6.01) respectively (Fig. 3).

Fig. 3
figure 3

Mean (± SE) percent germination of different seeds; Different letters correspondence with the data of treatments indicates significant difference between the treatments (P < 0.05) by Duncan’s Multiple Range test (DMRT).

Full size image

Microbial load test

The purpose of the study was to quantify the quality and kind of microorganisms that are present on the surface of seeds. Remarkably, the microbial load (bacteria and fungi) was found in all treatments; nevertheless, nest seeds had a significantly reduced microbial growth than other seeds (control seeds excluded). The colony growth of nest seeds, plant seeds, ant seeds, ground seeds and control seeds were 12.33 ± 0.88, 25.00 ± 1.15, 16.33 ± 0.88, 17.33 ± 1.20 and 10.66 ± 0.88. The number of infected seeds were 6.67 ± 0.33, 10.00 ± 0.00, 9.33 ± 0.33, 10.00 ± 0.00 and 5.33 ± 0.33 respectively in nutrient agar (NA). However, microbial load and seeds infected were found to be roughly comparable potato dextrose agar (PDA) (Figs. 4 and 5).

Fig. 4
figure 4

Growth of microorganisms from different seeds. (i) Nutrient media (NA), (a) nest seeds, (b) plant seeds, (c) ant seeds, (d) ground seeds, (e) control (Bavistin), (ii) Potato dextrose agar media, (a) nest seeds, (b) plant seeds, (c) ant seeds, (d) ground seeds, (e) control (Bavistin).

Full size image
Fig. 5
figure 5

Microbial load test on different seeds in nutrient media agar (N = 20 seeds), and (Average number of colony count and infected seeds). Different letters correspondence with the data of treatments indicates significant difference between the treatments (P < 0.05) by Duncan’s Multiple Range test (DMRT).

Full size image

Discussion

The current study experimentally demonstrates the seed cleaning behaviour of T. scabriceps ant species and its positive effect on seed germination via antimicrobial chemicals produced by worker ants. However, ants exhibit various behaviours; it is beneficial to collect antimicrobial chemicals from the production region, and it will apply to seed coat while removing the seed husk, lipid-rich elaiosome from the seed, and untimely useful to seed by reduce the microbial diseases. The behaviour like abdominal wagging is short-term effects due to trophallactic exchanges between adults and larvae by modulating salivary secretion and performing a conspicuous abdominal oscillatory behavior, known as abdominal wagging. Self-grooming is a proactive behavior and is consequently stimulated by ants detecting the presence of individuals with microorganisms. However, ants having additional advantage with groom each other called as allo-grooming. It is an important defensive behaviour to the efficient mechanical removal of parasites such as mites, fleas, nematodes, fungal hyphae or fungal spores from their vicinity. However, once the seeds are removed from the plants, ants start carrying these seeds to their nest using mandibles. Intra-nidal behavior comprises behaviors exhibited by the worker ants inside the nest. The worker ants carry the harvested seeds to their nest, and store them in separate chambers in the nest (nest granary). Inside the nest, ants clean the seeds by removing husk, and dispose the husk from the nest to outside. Moreover, ants clean the seeds through by resorting to cleaning behavior, which includes the use of mandibles, legs and antennation of seeds.

Allo-grooming, on the other hand, is linked with cleaning the bodies of other ants; however, this is a behavior that is frequently observed and limited to social insects. Another behavior displayed by these ants in the nest is antennal contact, which aids in the propagation of information in the colony through ant-ant contact24. Abdominal waggling was discovered to be less common in the community. This could be done to decrease the cost of secretion. As a result, these ants only exhibit this costly behavior when they face a hazardous threat in the nest33. Allo-grooming behavior appears to be essential in removing conidia on the surface of Red imported fire ants (RIFA) individuals help ants exposed to fungus by performing intensive grooming behaviors, which either risk infecting themselves or immunize them as social immunity. They also proposed that interacting with fungus-exposed ants would reduce nest members’ susceptibility to the fungus. All of these examples show that RIFA profits from grooming behavior when fighting fungal pathogens. Seed cleaning behaviour by Mycocepurusg oeldii Ants (Attini) Facilitates Germination in Hymenaeac ourbaril (Caesalpiniaceae) confirmed by Oliveria et al., 1993 where study area observed that entirely cleaned seeds by the ants were germinating, on the other hand seeds from uncleaned open pods were totally covered by a fungus-infested35.

Ants produce a number of antimicrobial compounds against diseases, with metapleural glands being the principal source (MGs). Fernandez-Marn et al.36 investigated how ants actively use metapleural glands to combat microbial infections. They stated the insects groomed the MG aperture with their forelegs. MG grooming rates were low in the absence of infection, and worker ants groomed themselves after touching the MGs. After MG grooming, leaf-cutter ants cleaned their fungal gardens, substrata (leaves), queens, and nestmates. Furthermore, seed cleaning by ants can help seeds spread by mammals and birds37. However, little is known about how ant seed cleaning affects seed longevity. It is not clear why the seeds cleaned by ants are not or are less infected by pathogens. Workers in some ant species have special glands containing antimicrobial substances that safeguard the body surface and/or the colony from microorganisms. With this background information, the current research attempted to evaluate the microbial load on seeds that had been coming from seed cleaning behaviour of ants. Further, experimentally proved the hypothesis that decreased the microbial load of seeds due to the activity of antimicrobial compounds produced from the metapleural gland secretion21. Seeds were collected from nests and stored in the nest seeds gallery by these harvester ants were investigated secretions of antimicrobial compounds during the seed cleaning process by T. scabriceps worker ants. Separately, two independent experiments were carried out to evaluate this hypothesis: a seed germination experiment and a microbial load test. In the seed germination experiment, nest seeds alone resulted in a substantial increase in the amount of germination, whereas other groups showed a statistically significant difference with the nest seeds. These findings suggest that ant secretion may have been coated on the seeds, assisting the nest seeds to escape from infection and thus germination. It is thus contradictory to the germination test, because microbial growth was observed in all groups, including nest seeds, but relatively less than in other seeds except control seeds. The use of nutrient-rich media (nutrient agar and potato dextrose agar) in the case of the microbial load test as opposed to blotting paper with adequate moisture but no nutrients in the case of the seed germination test could explain this contradiction. According to the current results, removing the aerial parts of the seeds as well as the fruits has a positive impact37,38,39,40 Cleaning of seeds and fruits is a typical behavior in most ant species41,42,43. Additionally, insects can clean seeds and fruits42,43. Furthermore, seed cleaning may be beneficial because it frequently aids in the defense against pathogens, which in turn aids in seed germination38,39.

Many times, ants cannot spread large seeds dispersed by birds or mammals, but they do provide facilities for their survival due to evolutionary behavioral effects. Interestingly, in Southeast Brazil, Hymenaea courbaril L. fruits are consumed by capuchin monkeys and other frugivorous animals, but worker ants, Mycocepurus goeldii (Forel, 1983), show the greatest germination compared to uncleaned fruit seed. If seeds are not cleaned by worker ants, fungi attack them severely35. Cupania vernalis Cambess seeds were cleaned by 13 ant species in a highland forest in southeast Brazil, and the cleaned seeds germinated at a higher rate than uncleaned ones37. The survival rate of bird distributed plant seed, Dysoxylum alliaceum (Blume) Blume, Bijdr. (1825) increased after the seed coat was removed by the ant species Dorylus laevigatus (Smith, 1857) and Pheidologeton affinis (Jerdon, 1851)45. The average germination rate (62%) was observed in the control treatment, whereas manual sarcotesta removal and sarcotesta removal by ants were observed (78% and 90%, respectively) in Guarea guidonia (L.)46. Seed-cleaning behavior by seed harvester ants is a remarkable example of the complex interactions between ants and plants, highlighting the importance of mutualisms and co-evolutionary relationships in shaping ecosystem dynamics.

Conclusions

The current study shows T. scabriceps harvester ant behaviour both in inside and outside the nest. Self-grooming was discovered to be the most prevalent behaviour in the nest, followed by antennation, seed cleaning, nest wall cleaning, allo-grooming, and abdominal wagging. Furthermore, the nest seeds outperformed the other seeds in terms of germination rate and microbial load. The study convincingly reveals that seeds cleaned by ants boost germination due to antimicrobial chemicals secreted by the insects’ metapleural glands. Overall, seed cleaning behaviour helps with seed dispersal and promoting seed germination in a range of plant species. This is an essential evolutionary occurrence in ants that helps with the distribution and regulation of plant species in a specific ecosystem.

Materials and methods

Study site

The current study was conducted to explore the seed cleaning and seed protectant behaviour of T. scabriceps worker ants. All the study trails were performed over two years (2017-18), at sites located in the University of Agricultural Sciences, Gandhi Krishi Vignana Kendra (GKVK), Bengaluru (12º 57˝ N 72º 35˝ E, 930 m mean sea level), Karnataka, India. The annual rainfall is 915.8 mm and with annual temperature of 24.1 oc. The climate falls within semi-arid area of India.

Trichomyrmex scabriceps ants collection

In order to study how ants clean the seeds, it is essential to establish an ant population. Ethogram construction is required to conduct the experiment under laboratory conditions because field conditions will make it difficult to observe ant behaviour in a suitable manner, which will impede research. The worker ants of T. scabriceps were collected from the field nests during the foraging activity and by excavation of nest in a field located at GKVK, Bengaluru, Karnataka, India, by following methods21. The taxonomic identification of the collected ants was determined using taxonomic keys47 and the voucher specimens of ant species used in this study are deposited in the reference collection of the Department of Agricultural Entomology, College of Agriculture, University of Agricultural Sciences, Bengaluru.

Aspirator method

The worker ants were collected from the nest location and from foraging trails by using an aspirator. The far end of the aspirator tube was placed on the ant, which leads to aspirator chamber by drawing air from other end of the other tube resulting in worker ants being drawn into the aspirator chamber. Ants collected in the aspirator chamber were then transferred to another collection box and brought to the laboratory for further analysis.

Hand-picking by using a brush

Worker ants of T. scabriceps were also collected in the field by placing the brush near the trial march and nest, the worker ants climbed and held the brush with their mandibles, and then these were transferred to vials.

Collection of ants from the nest

This method was employed for collecting worker ants of the T. scabriceps whenever there was no activity of ants in the field or no trail of foraging ants could be seen in the field. In this method ants were collected directly from the nest by digging the nest entrance and tracing the nest chambers where ants were to be found and the worker ants coming out of the nests were aspirated into the vials.

Establishment of the queenless ant nests under lab condition

We collected around 10 mature nests of T. scabriceps randomly in the study area during the rainy season. All nests were kept for 12 months inside a laboratory room under a controlled temperature of 25 ºC and LD 12:12 h for good growth. The queenless ant, T. scabriceps was established in the lab condition to study seed cleaning behavior. Queenless ant nest was established due to getting the reproductive ants from the nest is very difficult. Though these worker ants perform foraging their behaviour in the absence of a queen also were studied. To establishing nest with a transparent glass chamber was custom-made and the sterile sand was spread at the bottom. A total of six sterile/autoclaved test tubes filled with ¼ (one fourth) volume of water were placed inside the chamber to create and maintain moisture conditions in the artificial nest. Later, T. scabriceps worker ants were introduced into the lab nest and were given the 50% honey solution. The seeds Ragi or finger millet (Eleusine coracana L.) and little millet (Panicum sumatrense Roth. ex. Roem. and Schultz) collected from the nearby agricultural field area which were located at study location and then collected seeds that were placed properly for getting the observation from T. scabriceps (Fig. 6).

Fig. 6
figure 6

Collection of worker ants of Trichomyrmex scabriceps from the field and establishment of formicarium.

Full size image

Seed cleaning behavior of Trichomyrmex scabrices

The behavioural observations were carried out first by building an ethogram. The behavioral observations of T. scabriceps colonies was recorded to study the seed cleaning behavior of worker ants in the naturally constructed nest in the ethogram, the diversity of seeds harvested and seed cleaning during foraging trips. The diversity of seeds harvested was assessed by examining the granaries of the harvester ant T. scabriceps. A description of changes in the morphology of seeds before and after processing by worker ants was obtained. The viability of seeds from ant granaries was evaluated by in vitro germination studies.

Seed cleaning behavior of worker ants in laboratory condition

The behavioral parameters associated with the seed cleaning behavior were identified after the construction of a complete ethogram of intra-nodal tasks performed by worker ants. The different behaviours as mentioned were observed and recorded. These behaviours of ants associated with ‘granaries’ was quantified based on behavioral sampling following time-contingent rules of starting and stopping of observations. This were done by marking of individual ants using non-toxic paints of different color to the different individual ants. The individually marked ants were observed for five minutes followed by an interval of two minute before recording all occurrences in the next sampling bout. A total 10 ants were selected for mark with different colors. Individually collected the ants and placed them in the fridge until they were unable to move any further. The chemical paints were disrupted the behaviour of the ants, and we were unable to obtain accurate observation. Therefore, harmless posca paint pen was utilized to mark each worker ant. Later, the ants let it to dry properly; otherwise, other ants would clean and interfere with the study observation. The observations were carried out over several days by dividing the day into four different sessions of three duration each. A total of one hour’s observations were recorded for each of the ten ants. The observations of seed cleaning (SC), abdominal wagging (AW), self-grooming (SG), allo-grooming (AG), wall cleaning (WC) and Antennation (A) were carried out using an electronic stopwatch and video recordings were also made in the laboratory condition. For each observation, the number of times a particular behaviour was performed was recorded, and the total frequency/parameter for each ant was considered for statistical analysis.

Role of exocrine gland secretions of the worker ants as seed protectants

The different experiments were conducted to prove the exocrine gland secretion of ants acts as a seed protectant from the following tests.

Germination test

The germination test was used to know the seed cleaning behaviour effect on enhance the seed germination. Seeds were collected from the nest, by excavating nests, where seeds were stored inside the special galleries. We sequentially performed a germination test under a controlled condition of temperature (25 ºC) and LD12:12 h using seeds with four treatments, (i) seeds from the nest (n = 20), (ii) seeds transported by ants (n = 20), (iii) seeds from the ground (n = 20), and (iv) seeds from the plant (n = 20). These four seed sets were tested for germination in triplicates. For each treatment the blotting paper were humidified with distil water. The known quantity of seeds was placed into per dish (90 × 15 mm, with all of the essential conditions provided. We analysed and randomly check each petri dish in treatment wise on a daily basis to determine the germination percentage, dead seeds, abnormal seedlings. The observations were continued until all seeds germinated or showed signs of decomposition. The number of germinated seeds was determined using a formula. The remaining seeds, which were contaminated with various microbes but did not germinate, were numbered and documented.

Germination Percentage (GP):

GP = Number of Total Germinated Seeds/Total Number of Seeds Tested*100.

Microbial load test

This test was conducted to determine the microbial infection of all four groups of seeds (n = 20) and four groups of seeds were tested in triplicates. These seeds were designated as nest seeds (T1), seeds collected directly from the plants, on which these ants were foraging. These seeds were designated as plant seeds (T2), seeds collected from the ants while they were foraging. These seeds were considered as ant-carried seeds (T3), and seeds collected from the vicinity of the nest ground were labelled as ground seeds (T4) and the seeds were treated with fungicide (Bavistin; 0.2%) and bactericide (Tetracycline) were labelled as control seeds (T5). For testing the microbial infection, the nutrient agar (NA) and potato dextrose agar (PDA) media were prepared at 200 mL quantity. The medium was poured onto the irradiation-sterilized disposable petri plates and allowed for the solidification. The twenty selected seeds were placed onto the Petri dishes in triplicates. The experiment was performed in a growth chamber with constant temperature at 30 °C with incubation of 48 h to determine the growth of microbial load in each treatment and compared with the control.

Statistical analysis

The frequency of seed cleaning and behavior of worker ants were tested by One-way ANOVA with post-hoc Tukey’s test. The difference between the treatments was tested by one-way analysis of variance (ANOVA) at 1% and 5% levels of significance by using Duncan’s multiple range test (DMRT)47. The statistical analysis was performed through WASP (version 2.0) statistical package.