Insulin Affects Biological Behaviours of Pseudomonas aeruginosa

TL;DR

This study shows that insulin influences the behavior of Pseudomonas aeruginosa, affecting its growth, biofilm formation, and virulence.

Contribution

The study reveals that insulin modulates Pseudomonas aeruginosa virulence through multiple mechanisms.

Findings

All insulin concentrations increased bacterial growth.

Insulin reduced biofilm production, motility, and haemolytic activity.

200 µU/mL insulin decreased bacterial virulence in an animal model.

Abstract

Background: It is well known that host factors are capable of regulating microbial behaviours such as growth, metabolic pathways, virulence properties, and antimicrobial susceptibilities. In this respect, the present study aimed to investigate the relationship between insulin and various virulence properties of Pseudomonas aeruginosa ATCC 27853. Methods: Growth alterations, biofilm formation, motility, haemolytic activity, and pigment production (pyocyanin and pyoverdine) were determined in the presence/absence of three different insulin concentrations (10 µU/mL –20 µU/mL –200 µU/mL) under in vitro conditions. In addition, changes in bacterial virulence were evaluated in an in vivo animal (Caenorhabditis elegans) model. Alterations in growth, haemolytic activity, and pigment production were investigated spectrophotometrically. Biofilm formation was examined using the crystal violet well-plate assay. A soft agar plate method was used to detect swimming motility. Results: According to the results, all three insulin concentrations enhanced the bacterial growth. On the other hand, biofilm production, swimming motility, and haemolytic activity decreased in the presence of all insulin concentrations. Pyocyanin production was shown to be increased in the presence of only 10 µU/mL of insulin, but pyoverdine production did not change. In vivo animal survival rates showed that 200 µU/mL of insulin decreased bacterial virulence. Conclusions: This research demonstrates that P. aeruginosa can sense and respond to mammalian hormones (insulin), which can modulate microbial virulence through diverse mechanisms, providing new insights that may be relevant to infection dynamics.

Figures

Full text

1. Introduction

During infectious processes, many multidirectional and multifunctional interactions take place. These interactions occur not only between the microbe and its host but also in cell-to-cell signalling (quorum-sensing pathway, QS), which is crucial for the maintenance of pathogenesis [1,2]. According to the phenomenon of “microbial endocrinology”, both bacteria and the mammalian host manipulate each other’s behaviour through bidirectional inter-kingdom signalling pathways [3,4]. Mammalian hormones and bacterial QS molecules determine multicellular behaviour, which affects both the host immune system and bacterial virulence determinants, especially biofilm formation [5,6].

Since the early 1980s, it has been well recognised that hormones and hormone-binding proteins resembling those of mammals are widespread in bacteria and fungi. For instance, insulin has been shown to act as a signalling molecule, exhibit cross-kingdom activity, and regulate various virulence mechanisms in a taxonomically diverse range of organisms outside the kingdom Animalia, including Eubacteria (e.g., Escherichia coli), Fungi (Neurospora crassa), Protists (Tetrahymena spp.), and Plantae (Momordica charantia). Notably, E. coli has not only been shown to produce its own insulin-like molecules but also to respond to human insulin as a behavioural modifier [7,8,9,10,11,12].

Pseudomonas aeruginosa, an opportunistic, Gram-negative, rod-shaped, and non-fermentative bacterium, has been listed by the World Health Organization among globally critical pathogens. In addition, it is one of the most common causative agents of nosocomial and life-threatening infections, including ventilator-associated pneumonia, bloodstream infections, and soft tissue infections, particularly in patients with cystic fibrosis, diabetes, or burn injuries [1,2,13,14].

Flagella and type IV pili are described as major virulence determinants that interact with P. aeruginosa and the host during the initial stages of infection. The binding of flagella to host gangliosides also triggers the interaction between the host cell and bacterial lipopolysaccharide (LPS). In addition to attachment, other important biological properties include the type III secretion system (T3SS), proteases, pyocyanin, pyoverdine, and the ability to form biofilms [1,15,16]. The T3SS secretes cytotoxins into the host cell cytoplasm, while proteases facilitate the dissemination of the bacterium. The redox-active toxin pyocyanin exacerbates the infection, and pyoverdine acts as a siderophore (iron chelator), which is essential for the bacterium’s biofilm formation [1,2,15,17,18].

Due to the difficulties in eradication and persistence of critical infections, biofilm formation capacity and antibiotic resistance are major factors commonly responsible for life-threatening infections. Attachment, microcolony formation, maturation, and dispersion are the key steps in the biofilm formation of P. aeruginosa [19,20]. As a sugary slime matrix, biofilms consist of multi-layered bacteria and extracellular polysaccharides (EPS), DNA, proteins, and lipids [20,21].

Chronic P. aeruginosa infections occur nearly three times more often in diabetic patients [21,22]. Rising multidrug resistance has made these infections increasingly difficult to treat, and even the most effective antibiotics—such as ceftazidime, fluoroquinolones, amikacin, and piperacillin—provide successful outcomes in only 64–80% of cases. In diabetic foot wounds, treatment becomes particularly challenging due to impaired circulation, neuropathy, weakened immune responses, and poor glycaemic control, all of which contribute to enhanced biofilm formation by pathogenic bacteria. Paradoxically, insulin treatment has also been reported to increase biofilm formation and antibiotic tolerance under certain conditions [22,23,24].

Recent studies have demonstrated that host-derived metabolic signals, including insulin, can modulate bacterial physiology and virulence-associated traits. Insulin has been shown to influence bacterial growth, biofilm formation, and pathogenicity; however, these effects appear to be highly context-dependent and vary according to bacterial species, experimental model, and insulin concentration [12,14]. In this context, the interaction between insulin and P. aeruginosa is not completely clarified, with previous studies reporting divergent and sometimes conflicting outcomes [21,22,23,24], highlighting the need for an extensive evaluation of insulin-mediated effects on P. aeruginosa virulence.

Therefore, the aim of this study is to investigate the effect of insulin, at different concentrations ((10 µU/mL –20 µU/mL –200 µU/mL), on the growth and virulence characteristics (biofilm formation, motility, haemolytic activity, and pigment production) of P. aeruginosa ATCC 27853. In addition to in vitro alterations, the effects of interactions between P. aeruginosa and insulin were investigated in an in vivo animal model (Caenorhabditis elegans). We believe that the findings obtained from this study, considering the spread of virulent and resistant strains, will provide important preliminary data regarding the impact of insulin management as a host factor on the course of infection.

2. Results

2.1. The Alterations of Growth

The presence of insulin did not affect the growth of P. aeruginosa at 2, 4, and 6 h. However, a statistically significant increase was observed at 24 h (p ≤ 0.0006) (Figure 1).

2.2. The Alterations of Biofilm Formation

Biofilm formation was found to be suppressed in the presence of all insulin concentrations at 48 h incubation (p ≤ 0.0037) (Figure 2).

2.3. The Alterations in Motility

Bacterial motility was reduced in the presence of each insulin concentration (p = 0.0022) (Figure 3a,b).

2.4. The Alterations in Haemolytic Activity

Our results show that P. aeruginosa grown in the presence of insulin exhibited significantly decreased haemolysis when incubated with human red blood cells (p ˂ 0.0001) (Figure 4). Furthermore, no significant difference was found in the haemolysis zone diameters on blood agar media in the presence of insulin.

2.5. The Alterations in Pyocyanin and Pyoverdine Production

Pyocyanin production was significantly increased in the presence of 10 μU insulin (p ˂ 0.0001). The other two concentrations did not affect pyocyanin production (Figure 5a). Similarly, none of the insulin concentrations affected pyoverdine production (Figure 5b).

2.6. The Alterations in C. elegans Viability

P. aeruginosa infection significantly reduced the viability of C. elegans. In the presence of 10 and 20 µU/mL insulin, bacterial virulence was enhanced, resulting in complete worm mortality within two hours. In contrast, treatment with P. aeruginosa exposed to 200 µU/mL insulin-induced killing, leading to higher worm survival compared to the lower insulin concentrations (Figure 6).

3. Discussion

Severe trauma, burns, and sepsis often trigger simultaneous metabolic and immune disturbances, where hypermetabolism and dysregulated inflammation can lead to multiple organ dysfunction and death. These abnormalities are closely associated with hyperglycaemia driven by insulin resistance and diabetes, which also impairs immune cell functions [25]. Insulin therapy is widely used in intensive care to control hyperglycaemia and reduce morbidity and mortality. Studies show that insulin lowers mortality by about 4% and decreases septicaemia, ventilator dependence, and organ failure, particularly in paediatric patients [25,26,27,28].

Beyond its metabolic role, insulin modulates inflammation and enhances immune cell activity, contributing to improved infection control through anti-inflammatory and immunoregulatory effects [25,29,30]. With diabetes prevalence increasing globally—estimated to reach 8.3% by 2030—patients are at higher risk for complications including renal failure, myocardial infarction, blindness, stroke, and limb amputation [14,22]. Chronic foot ulcers are a major concern among diabetic patients, and in cases complicated by antibiotic resistance and high virulence, amputation may be required [14,21,22,31,32].

P. aeruginosa is a major pathogen in diabetic wound infections and is characterised by strong biofilm-forming ability [14,21]. Strains isolated from diabetic mice show enhanced biofilm formation and increased antibiotic tolerance compared with those from non-diabetic hosts [21]. Although low-dose insulin has been reported to reduce P. aeruginosa–induced sepsis in burn patients [14], other studies demonstrated that insulin can unexpectedly promote biofilm formation and antibiotic tolerance in vivo [25]. Insulin may facilitate the shift from planktonic to biofilm-associated phenotypes and influence host responses by decreasing pro-inflammatory cytokines (IL-1β, TNF-α) while increasing anti-inflammatory cytokines (IL-4, IL-10) [25]. Its anti-inflammatory effects have also been observed in both diabetic and non-diabetic human cells [33,34]. Overall, these findings suggest that insulin acts not only as a metabolic hormone but also as an immunomodulator that can alter P. aeruginosa pathogenicity in diabetic hosts.

In addition to in vivo investigations focusing on the effects of insulin on the clinical progression of critically ill patients, several in vitro studies are being conducted to elucidate microbe–insulin interactions and to better understand the potential benefits of insulin during treatment.

Gumus et al. (2017) [6] demonstrated that the growth of UPEC strains was not affected in the presence of 20 or 200µU/mL insulin, 0.1% glucose, or the combination of 200 µU/mL insulin with 0.1% glucose. However, the expression levels of several UPEC virulence genes (usp, sfa/foc, cnf-1) were reduced in the presence of insulin and/or glucose. In another study, the growth of Pseudomonas pseudomallei was shown to be inhibited in the presence of 10 µU/mL insulin under minimal medium conditions [11]. The growth of E. coli and S. aureus, on the other hand, was enhanced in the presence of high insulin concentrations (1 and 2 U/mL) [35]. Plotkin and Viselli (2000) evaluated the effects of various insulin concentrations (2, 20, 200, and 400 µU/mL) on the growth rates of E. coli, S. aureus, Enterococcus faecalis, and P. aeruginosa, demonstrating that these bacteria respond to insulin in a concentration-dependent manner [36].

Virulence factors have also been shown to be affected by physiological and supra-physiological concentrations of insulin. Most studies have focused on the relationship between insulin and biofilm formation in various bacteria. For example, biofilm formation by an E. coli strain was reported to increase in the presence of 20 and 200 µU/mL insulin under Mueller–Hinton broth conditions, whereas no effect was observed in a minimal nutrient medium. The authors also emphasised that the effects of insulin on colonisation are dependent on the presence of glucose [37]. To further clarify the hyperglycaemic stress response, Plotkin et al. (2023) [12] investigated the effects of insulin and/or glucose (supra-physiological concentrations) on the biofilm-forming capacities of various Gram-negative and Gram-positive bacteria. They found that insulin, in combination with high glucose concentrations (160–180 mg/dL), decreased biofilm formation in a concentration- and microbe-specific manner. In addition, the authors highlighted that the co-regulation of the insulin/glucose ratio plays a critical role in the clinical management of critically ill patients [12]. In another study conducted by the same authors, the effects of insulin (2, 20, and 200 µU/mL) and 13 different sugars on the adherence capacity of an E. coli strain to latex in a minimal nutritional medium were investigated. The highest attachment rates were observed in the presence of glucose, mannose, and 20–200 µU/mL insulin, suggesting that insulin enhances the interaction between E. coli and materials commonly used in medical settings in a nutrition-dependent manner [38].

Similar to previous studies the results of the present study also indicate that P. aeruginosa is highly responsive to insulin under both in vivo and in vitro conditions.

We found that growth was significantly enhanced over a 24 h period. In addition, biofilm formation capacity was reduced, which is in contrast with previous findings [21]. It is well known that bacterial motility facilitates surface colonisation and biofilm formation, as well as the production of bio-surfactants [39,40]. Moreover, haemoglobin release by haemolysis and proteolytic degradation of iron-binding proteins are important for bacterial heme-iron acquisition [41,42]. Considering these effects of motility and haemolysis in the pathogenesis, in our study, both motility and haemolytic activity were also shown to be significantly decreased in the presence of all insulin concentrations tested. As a distinction from these virulence mechanisms, pyocyanin has been shown to contribute to tissue damage, especially in lung infections. Furthermore, notable amounts have been identified in ear secretions, sputum, urine, and wound samples from patients with chronic infection [43,44]. In our study, pyoverdine production was not altered, and pyocyanin production was induced by the effects of only 10 µU/mL insulin.

Although increased bacterial growth is often associated with enhanced virulence, the expression of virulence factors such as haemolysin and pyocyanin is not only regulated via bacterial proliferation. Our findings suggest that insulin may differentially modulate bacterial growth and virulence-associated pathways, leading to enhanced growth while simultaneously suppressing certain virulence phenotypes at specific concentrations potentially through quorum-sensing interference or global regulatory mechanisms [45,46,47]. Pyocyanin production is also regulated not only by bacterial density but also by quorum-sensing networks and environmental cues, which are the main modulators. The lack of increased pyocyanin production at higher insulin concentrations may reflect an inhibitory or modulatory effect of insulin on quorum-sensing-mediated virulence regulation, rather than a direct metabolic stimulation. The absence of a parallel effect on pyoverdine production suggests that insulin-mediated signalling may preferentially influence redox-associated or quorum-sensing–linked virulence traits rather than iron-acquisition systems.

In vivo results showed that 10 and 20 µU/mL insulin significantly decreased the viability of C. elegans compared to infectious conditions without hormone at 1 and 2 h periods, but not after 3 h of incubation. However, consistent with previous studies [25,26,27], the supra-physiological insulin (200 µU/mL) seems to have a protective effect for the host over a three-hour period. C. elegans is a well-established in vivo model for studying P. aeruginosa virulence and host–pathogen interactions, particularly as a rapid and ethically feasible screening system. However, we acknowledge that this model lacks the physiological complexity of mammalian systems. On the other hand, previous studies showed that insulin or IGF-1 treatment modulated lifespan and physiological processes in C. elegans without inducing toxicity or increased mortality [48,49]. Therefore, future studies may address the direct impact of insulin on C. elegans physiology and immune responses; however, this was beyond the scope of the current work.

Consequently, our findings should be interpreted as preliminary and further studies using mammalian infection models are required to confirm the translational relevance of the observed effects.

4. Materials and Methods

Bacteria and insulin concentrations

P. aeruginosa ATCC 27853 was used to investigate the effect of insulin (10 µU/mL, 20 µU/mL, and 200 µU/mL) on bacterial growth and virulence properties. These insulin concentrations were selected based on normal blood levels as well as 2-fold and 20-fold higher than physiological levels [50]. In addition, these concentrations reflect physiologically and pathologically relevant ranges reported under metabolic stress, dysregulated glucose homeostasis, and potential localised tissue microenvironments during infection.

The detection of growth alterations

P. aeruginosa strain suspension (10^7^ CFU/mL) was prepared from overnight culture and was inoculated into tryptic soy broth (TSB) with or without insulin and incubated at 37 °C for 2, 4, 6 and 24 h. The growth alterations were determined spectrophotometrically (NanoDrop 2000 spectrophotometer, Thermo Fisher Scientific, Waltham, MA, USA) at 600 nm. All experiments were repeated three times independently and all conditions were analysed thrice.

The detection of biofilm formation

Biofilm formation was examined using a crystal violet well-plate assay [51]. P. aeruginosa strain was inoculated into TSB supplemented with 1% (v/v) glucose and diluted 1:50 in fresh TSB-glucose, yielding a final concentration of approximately 10^7^ CFU/mL. A 96-well plate containing 20 μL P. aeruginosa suspension, 180 μL TSB-glucose, and 20 μL of insulin at different concentrations was incubated at 37 °C for 48 h. The contents of wells were aspirated and gently washed three times with phosphate-buffered saline (PBS). Following washing, 200 μL of 99% methanol was added to fix the biofilm and incubated for 15 min. The wells were stained with 0.1% crystal violet for 5 min for determination of biofilm. Afterwards, the stain was rinsed off with water, and the plates were air-dried. After the washing stage, 95% ethanol was used to dissolve the stain. Biofilm formation was quantified by measuring absorbance at 450 nm using a spectrophotometer (NanoDrop 2000 spectrophotometer Thermo Fisher Scientific, Waltham, MA, USA). All experiments were performed three times.

The detection of motility

P. aeruginosa strain was grown in TSB in the presence/absence of insulin at different concentrations at 37 °C for 24 h. Approximately 10^8^ CFU/mL of bacterial suspension was inoculated into a soft agar plate [52,53] and incubated 30 °C for 16 h. After incubation, we measured the diameter of the migration zone to assess bacterial motility. The assay was repeated thrice.

The detection of haemolytic activity

Erythrocytes were used to detect haemolytic activity and were obtained from two donors with blood group 0, Rh (+). Erythrocytes were washed three times with PBS, which was used for resuspension to a final concentration of 2% erythrocytes at 4 °C. P. aeruginosa strain was cultured in TSB with or without insulin at different concentrations and incubated at 37 °C for 24 h. After incubation, bacterial culture was centrifuged at 8000 rpm, and the pellet was resuspended in PBS to achieve a turbidity equivalent to 2.0 McFarland standard. Equal volumes (1:1) of bacterial and erythrocyte suspensions were mixed and centrifuged at 1.000 rpm for 10 min. A 0.1% SDS solution was used as a positive control. Then, the mixtures were incubated at 100 rpm and 37 °C for 1 h. Following incubation, the tubes were centrifuged at 1.000 rpm for 10 min. Then, 200 μL of supernatant was transferred to a 96-well plate. The absorbance was measured spectrophotometrically at 540 nm (NanoDrop 2000 spectrophotometer, Thermo Scientific, Waltham, MA, USA). The results were evaluated according to the method described previously [54]. All experiments were performed three times.

Furthermore, we examined the degree of haemolysis (colony diameter, mm) using 5% sheep blood agar. For this purpose, P. aeruginosa was cultured in TSB with/without insulin at 37 °C for 24 h, and 2 μL of the suspension was inoculated onto blood agar. The results were interpreted based on the measured colony diameters. All experiments were repeated three times.

The detection of pyocyanin and pyoverdine

Both pigment productions were determined as previously described [55].

To detect pyocyanin production, P. aeruginosa (10^7^ CFU/mL) was inoculated into 3 mL of Mueller–Hinton Broth (MHB) with or without insulin at different concentrations and incubated at 300 rpm and 37 °C for 24 h. Following incubation, cultures were centrifuged at 3000 rpm for 10 min. Supernatants were collected and mixed with chloroform (3 mL). The mixture was vortexed until a greenish-blue colour appeared. The tubes were centrifuged again at 3000 rpm for 10 min. The blue-chloroform layer was collected and mixed with 1 mL of 0.02 N HCl until the mixture turned pink. The absorbance was measured at 520 nm. Pyocyanin concentration (μg/mL) was calculated by multiplying the absorbance at 520 nm by 17.072.

For pyoverdine quantification, P. aeruginosa (10^7^ CFU/mL) was inoculated into 3 mL of TSB with or without insulin at different concentrations and incubated at 300 rpm and 37 °C for 24 h. Bacterial growth was determined spectrophotometrically at 600 nm. Cultures were centrifuged at 7000 rpm for 15 min. To obtain cell-free supernatants, cultures were filtered through 0.4 μm filters, and the absorbance was measured at 405 nm. Relative pyoverdine production was determined by normalising OD405 values to OD600. All experiments were performed in triplicate.

The detection of Caenorhabditis elegans viability

To evaluate the changes in bacterial virulence in an in vivo model, the survival rates of C. elegans were determined [56]. For this purpose, the Bristol wild-type N2 strain of C. elegans (Caenorhabditis Genetics Center, University of Minnesota, Minneapolis, MN, USA) was routinely maintained on nematode growth medium (NGM) plates seeded with E. coli OP50 at 21 °C. The N2 strain was kindly provided by Assoc. Prof. Isak Demirel (Örebro University, School of Medical Sciences, Örebro, Sweden).

Until reaching the L4 stage, synchronised worms were transferred to NGM agar plates seeded with E. coli OP50 and incubated at 21 °C for 48 h. P. aeruginosa was grown in Mueller–Hinton Broth (MHB) with or without insulin at different concentrations and incubated at 37 °C for 24 h. The fourth larval-stage worms were washed with M9 buffer, and 10 worms were transferred into each well of a 96-well plate, after which P. aeruginosa (1:1 suspension in PBS) was added. C. elegans and P. aeruginosa were incubated together at 21 °C for 3 h. The viability of C. elegans was assessed every hour, and a worm was considered dead when it failed to respond to touch. All experiments were performed three times.

Statistical analysis

Statistical analysis was determined using GraphPad Prism 5. Growth alterations were analysed using two-way ANOVA followed by Dunnett’s multiple comparisons tests. Biofilm formation, pigment production and haemolytic activity were analysed using one-way ANOVA followed by Dunnett’s multiple comparisons tests. The motility alterations were analysed using one-way ANOVA followed by Tukey’s multiple comparisons tests. The survival ratio of C. elegans in the presence of P. aeruginosa was analysed using two-way ANOVA followed by Bonferroni post-test.

All measurements were compared to control conditions (TSB or MHB). The samples were tested in triplicate and each experiment was performed thrice. Multiple comparisons were made at a level of p < 0.05.

5. Conclusions

In conclusion, the present study highlights the multifaceted role of insulin, capable of altering microbial behaviour and infection outcomes. Our findings show that P. aeruginosa is highly responsive to insulin exposure under both in vivo and in vitro conditions. Insulin significantly enhanced bacterial growth while suppressing key virulence traits such as biofilm formation, motility, and haemolytic activity, although pigment production was differentially affected depending on the concentration. Furthermore, low insulin levels (10–20 µU/mL) reduced C. elegans survival at early time points, whereas high-level insulin (200 µU/mL) improved host viability, aligning with the clinical benefits previously reported in critically ill patients. It is possible to hypothesise that host-mediated factors such as innate immune responses, metabolic status, and stress signalling pathways may influence bacterial pathogenicity. Another explanation that should be considered is that the protective effect observed at 200 µU/mL insulin in vivo may be attributed to host-related mechanisms rather than direct bacterial modulation alone.

Taken together, these results demonstrate preliminary insights into the complex and concentration-dependent effects of insulin on P. aeruginosa growth, virulence, and host interactions. Our findings highlight the potential modulatory roles of insulin on microbial pathogenesis in a concentration-dependent and microbe-dependent manner. Because investigations focusing on P. aeruginosa and insulin interactions are limited, we consider that clarifying these interactions is important to determine proper insulin administration, particularly in patients at high risk of infection or impaired immune function. Furthermore, further mammalian infection models are required to fully elucidate the underlying pathways.