All bacterial epibiotic predators are rich in secondary metabolites and most genera rich in secondary metabolites have demonstrable predatory abilities. Predation resistant species may benefit not only by escaping predation but also by utilizing nutrients released by lysis of prey cells by predatory bacteria. The resistant organisms may enjoy greater fitness benefits than the predator since they get the benefit without investing in the predation machinery.

In our experiment, a marine isolate of Streptomyces atrovirens showed good predatory activity on a range of species including Staphylococcus aureus and Proteus vulgaris. Escherichia coli was resistant to predation by this species. On slide culture with water agar when the predator, S. aureus and E. coli were grown together S. aureus population declined whereas the predation resistant E. coli increased their population as compared to controls.

However the growth of E. coli did not affect growth of the predator unfavorably. Thus, the experiment demonstrates that resistance to predation gives an added advantage of nutrition. This strengthens the possibility based on previously published evidences that evolution of antibiotic resistance not only gave a selective advantage of escaping predation, it also would have increased the fitness of the resistant organism by promoting growth on nutrients released from the prey cells lysed by the predator.

When the predator was grown with S. aureus and P. vulgaris as prey, S. aureus declined rapidly whereas P. vulgaris was spared. This suggests that the predator appears to show preference towards prey and in that case even a partial or relative resistance may give substantial advantage to a population.


Change in cell density of prey cells and the predator

Figure 1. Change in cell density of prey cells and the predator.

(A) Microscopic image of slide culture for three species experiment with prey and resistant cells.
(B) Change in absolute average cell count of S.aureus and E.coli along with controls over the incubation period.
(C) Change in cell count of S.aureus and E.coli.

(D) Change in cell count of the predator, S.atrovirens along with controls.
(E) Microscopic image of slide culture for three species experiment with two sensitive cells.
(F) Change in absolute average cell count of S.aureus and P. vulgaris along with controls over the incubation period.
(G) Change in cell count of S.aureus and P. vulgaris.

(H) Change in cell count of the predator, S.atrovirens along with controls.


Predation is defined as the act of consumption of a living organism by another living organism[1]. Predation is a major ecological force, shaping the structure of communities, driving diversity and evolution of life histories[2][3]. Since bacteria do not have phagocytic abilities[1] predatory bacteria lyse and degrade the prey cells using extracellular weapons including enzymes and secondary metabolites[4][5].

Involvement of antibiotics and other secondary metabolites in predation is inferred from the strong association between predatory activity and genomic richness of secondary metabolites[6]; differential expression of secondary metabolite genes when co-cultured with different prey species[7] and loss of predatory abilities after mutating the antibiotic genes[8]. Kumbhar and Watve[6] argued that antibiotics primarily evolved for predation, later diverging into mutualistic, signalling[9] and other functions.

Based on the association between antibiotics and predation Leisner et al[10] argued that antibiotic resistance primarily evolved to resist predation but further this ability allows the resistant organisms to cross feed on the nutrients released from the lysis of susceptible prey cells by the predator. Resistant organisms can thus hitch hike on the predator and they might enjoy a fitness advantage over the predator since the predator needs to invest in the predation machinery which the hitch hiker doesn’t have to.

Although this is a logical possibility, this has not been experimentally demonstrated so far.
Streptomyces sp. is known for producing an array of secondary metabolites[11] accounting for around 70–80% secondary metabolites within the actinomycetes[12]. Streptomyces spp. are non-obligatory predators which are suspected to utilize a diversity of diffusible secondary metabolites for predation[4][6][7]. The lysis of prey cells adjoining Streptomyces mycelium has been demonstrated microscopically using Differential Interference Contrast (DIC) microscopy[7].

In the present study we monitored the populations of susceptible and resistant prey species in presence of a predatory Streptomyces. We screened S. atrovirens, a predatory marine isolate against 14 lab strains and environmental isolates out of which 6 were susceptible to predation (S.aureus, P.vulgaris, Mycobacterium smegmatis, Serratia marcencesus, Micrococcus leuteus and Bacillus subtilis). After studying predatory activity against individual cultures we studied the responses of mixed populations of prey cells. The population response of a predation susceptible and resistant species, as well as two susceptible species co-cultured in presence of predator is reported here.


To experimentally demonstrate additional nutritional and growth benefit to a predation resistance organism in presence of a predator and prey population.

Results & Discussion

Since the medium contained no nutrients, in pure culture controls the prey cells did not show a significant trend in population during the incubation period. In prey co-culture there was no significant trend in the populations of S. aureus and E. coli however S. aureus and P. vulgaris populations showed a growing trend when co-culturing indicating some synergistic interaction. We did not investigate the nature of this interaction. In one to one predator-prey interactions, S. aureus and P. vulgaris populations showed a monotonic declining trend while the predator population increased. In the E. coli predator interaction, neither the population of the predator nor that of E. coli showed a time trend.

In the three species interaction experiment with S. aureus and E. coli along with the predator, the S. aureus population declined by 97% whereas the E. coli population increased by 60% (Fig. 1C). The predator population increased by 72% (Fig. 1D).

In the three species interaction experiment with S. aureus and P. vulgaris along with the predator by the second day, the S. aureus population declined by about 68% and remained low on the third day but P. vulgaris population did not show a significant change. However, during the last phase of growth the P. vulgaris population showed a decline by about 61% whereas S. aureus showed a small but significant increase (Fig. 1G). S. atrovirens population increased monotonically (Fig. 1H). This suggests that the predator might have selectively preferred S. aureus in the first phase and when its population declined below a threshold shifted to P. vulgaris allowing some comeback growth of S. aureus.

The results highlight the complexity of multi-species interactions in bacteria. Earlier studies have shown that in the absence of other soluble nutrients, some bacterial genera turn predatory and grow at the expense of surrounding cells. We show here that different bacteria show differential sensitivity to predation which is important in shaping the patterns of interaction. Since bacterial predation is an extracellular phenomenon, a resistant organism gets an added nutritional benefit when in the proximity of a predator and a susceptible organism.

This was predicted earlier[10] but our experiments have demonstrated this advantage of resistance empirically for the first time. The mechanism of predation resistance is not yet known. Secondary metabolites of the predator are suspected to have a role in predation[7], therefore it is likely that resistance to the secondary metabolites might be the primary mechanism of predation resistance. If this is true it gives an added dimension to the evolution of antibiotic resistance.

More complex and more interesting appear to be the interaction between two predation susceptible species. The predator that showed successful predation on both when tested separately appeared to exhibit preferential consumption of one species when challenged with both together. This appears to have spared the other species at least for some time. This demonstrates that even partial or relative resistance may give substantial selective advantages in natural multispecies settings.

Microbiology has largely progressed by pure culture studies. Natural ecosystems, on the other hand, have a variety of multispecies interactions about which our understanding is still quite primitive. The experiments suggest that studying multispecies interactions will reveal a variety of novel dimensions of bacterial life in nature.


Antagonistic associations between microbial species have been viewed as two species interactions. The cost-benefits in multi-species interactions can be substantially different, understanding of which can be crucial in understanding evolution of phenomena such as antibiotic resistance or susceptibility.


Escherichia coli (NCIM 2184), Staphylococcus aureus (NCIM 2121) and Proteus vulgaris (NCIM 2172) were inoculated in nutrient broth and incubated for 24 h at 37°C. Prey cells were centrifuged (Eppendorf centrifuge 5810R) at 6000 rpm for 10 min to concentrate cells followed by washing with sterile distilled water to remove traces of nutrients. After washing cells were suspended in sterile distilled water making thick slurry of cells. Prey cell number was standardized based upon a relationship between optical density of the washed suspension and the cell density obtained using Neubauer chamber (Rohem India BS 748).

To study three species interaction we used two different co-culture combinations namely E. coli and S. aureus in approximately 1:1 ratio and P. vulgaris and S. aureus as prey in the same ratio. In these combinations it was possible to identify the cells based on morphology alone. It was not possible to study the E. coli–P. vulgaris combination owing to the difficulty of differential identification microscopically. The prey population was spread over the surface of water agarose bed on a slide culture over which S. atrovirens the predator, was spot inoculated.

The agarose bed was covered with sterile coverslips so that same slide could be observed every day. The slides were incubated at 30°C for up to 5 days in moist chambers. Three controls were prepared similarly with (i) individual prey species in pure culture, (ii) individual prey species in presence of predator, and (iii) co-culture of two prey species without the predator.

Observations were made daily using DIC microscopy on Zeiss Axioimager M-1 upright microscope (100X/1.40 Oil DIC M27), alpha plan apochromat objective, with a digital camera, HAL 100 illuminator and quartz collector controlled by Axio software 4.8. DIC images were taken and resulting images were used for the quantification of cells. Images were taken for every slide every day from one end of the slide to the other to cover the entire spread of the population.

Images were labeled with field number. Those with substandard image quality were rejected and from the remaining, 10 images were selected by random number generation. Cells from an entire field were counted systematically using temporary grid-lines. The number of cells of each of the prey species per field was counted from the images. S. atrovirens being a mycelial species, we recorded in every image the number of times a diagonal crossed mycelial threads. The change in the population densities thus measured during the 5 day incubation period was studied.

Reproducibility runs

Experiment was repeated 3 times at different time points (data not shown here) and similar trends were observed in all the repetitions. Since the starting populations were substantially different in every run we did not pool the results. In all the 3 experiments the results were qualitatively the same in that in every run, S. aureus population declined while E. coli population and the predator population increased in three species experiment.

In case of three species experiment with two sensitive organisms S.aureus and P. vulgaris populations of both the organisms declined; S. aureus being first to show decline followed by decrease in P. vulgaris population. Predator, S. atrovirens increased in both the experiments.

Funding Statement

Funding for this research was provided by Rajiv Gandhi Science and Technology Commission under Maharashtra Gene Bank Programme.

Conflict Of Interest

The authors declare no conflicts of interest.

Ethics Statement

Not Applicable.

No fraudulence is committed in performing these experiments or during processing of the data. We understand that in the case of fraudulence, the study can be retracted by ScienceMatters.