Influence of polyunsaturated fatty acids on Aeromonas hydrophila: sole carbon utilization, incorporation into membrane phospholipids, potentiation of cyclic lipopeptide activity, and other phenotypic impacts

Abstract

Aeromonas hydrophila is an adaptable and dangerous member of a genus capable of colonizing and infecting many hosts. Inhabitants of various aquatic environments, A. hydrophila can adversely affect aquaculture and cause human disease. The aim of this study was to determine the extent of polyunsaturated fatty acid (PUFA) handling in A. hydrophila with regard to incorporation into membrane phospholipids and effects on permeability, growth, antimicrobial resistance, biofilm formation, and swimming motility. Extracted phospholipids from PUFA-supplemented cultures were examined for lipid modification using both thin-layer chromatography and ultra performance liquid chromatography-mass spectrometry, demonstrating partial incorporation of all tested PUFAs. Phenotypic characteristics were evaluated using assays that tested for biofilm formation, membrane permeability, swimming motility, and antibiotic susceptibility. Temperature-dependent effects on biofilm formation and swimming motility were observed, and exposure to certain PUFAs resulted in significant (P < 0.005) increases to membrane permeability using a hydrophobic dye uptake assay. Importantly, several PUFAs caused substantial (up to 16-fold) lowering of minimum inhibitory concentrations (MICs) against the membrane-active cyclic peptides polymyxin B and colistin. These responses to exogenous PUFAs may predict cellular signaling capabilities that aid multi-host pathogenic bacteria in their diverse environmental interactions.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-025-04675-0.

Background

Aeromonas hydrophila is a versatile Gram-negative bacterial pathogen capable of surviving a wide range of aquatic environments, where it may cause contact- and consumption-mediated infection of warm and cold-blooded organisms [1]. Specifically, A. hydrophila causes motile Aeromonas septicemia and ulcerative infections in fish and has been attributed to die-offs and fish kills around the world [2, 3]. In predominantly immunocompromised humans, A. hydrophila causes a variety of illnesses, with often fatal gastrointestinal, soft tissue, and opportunistic wound infections comprising most cases [4]. Currently, many treatments for A. hydrophila infection are ineffective, or partially effective, due to the increased prevalence of multi-drug resistant strains [5].

Previous research demonstrates that exposure of gram-negative bacteria to polyunsaturated fatty acids (PUFAs) elicits changes to physiology and behavior, manifested in membrane permeability, antimicrobial susceptibility, growth patterns, biofilm formation, and motility [6–12]. Aeromonas species, akin to Vibrio species, are known to encode a surplus of genes involved in acquisition and utilization of exogenous fatty acids, including outer membrane transport proteins, (FadL), long-chain fatty acyl-CoA ligases (FadD), and phospholipid acyltransferase (plsB/C/X/Y) homologs [13]. Gaps in knowledge remain for the bacterial pathways governing maintenance and control of fatty acid handling, such as traversing the periplasm and capacity for signal transduction in regulatory processes. Furthermore, these genera are known for their capability to thrive in diverse aquatic environments exhibiting wide ranges in pH, salinity, and temperature [14, 15]. This environmental hardiness is accompanied by the possession of abundant methyl accepting chemotaxis proteins (MCPs), consisting of several classes involved in recognizing and responding to signaling molecules [16]. PUFAs represent candidate ligands for these pathways, especially in bacteria that oscillate between aquatic and human host lifestyles, where PUFA composition can define environmental location. ​The current study highlights the ability of A. hydrophila to utilize PUFAs for membrane phospholipid modification and phenotypic responses linked to virulence. The PUFAs chosen for this study encompass major and minor fatty acids found in both humans (gastrointestinal tract) and fish (tissue and epidermal mucus). Presumptively, PUFAs represent a collection of molecules that can not only be scavenged for beneficial catabolic processes, but can also be (i) assimilated into membrane phospholipids and (ii) detected by sensing pathways for control of important behaviors associated with survival and virulence. These findings may uncover a strategy for alleviating the severity of human and fish infections when they occur. Understanding the roles/impact of PUFAs on A. hydrophila holds potential for developing control, prevention, and treatment measures for antibiotic-resistant bacteria in our ever-changing world.​.

Methods

Bacterial strain and growth conditions

Aeromonas hydrophila subsp. hydrophila (Chester) Stanier ATCC 7966 (type strain) was used in this study. A. hydrophila was grown in a modified M9 media [17], henceforth termed CM9-HEPES, supplemented with 0.4% glucose and 0.4% casamino acids and buffered with 100mM HEPES (pH 7.4). Cultures were grown at 37 °C for all experiments unless described otherwise. Bacterial cultures were grown in the presence of 300 µM of each PUFA and compared to a control culture grown without fatty acid supplementation. Fatty acids used in this study were purchased from Cayman Chemicals [linoleic acid (18:2), a-linolenic acid (18:3a), g-linolenic acid (18:3 g), dihomo-g-linolenic acid (20:3), arachidonic acid (20:4), eicosapentaenoic acid (20:5), and docosahexaenoic acid (22:6)].

Growth curve analyses

Growth of A. hydrophila at 28 °C and 37 °C was assessed in cultures supplemented with and without 300 µM of a given fatty acid. For sole carbon growth curves, 1 mM fatty acid was supplemented and growth was monitored at 28 °C and 37 °C for 12 h with absorbance (OD_600nm) being measured every hour. A positive control containing 1 mM glucose and no fatty acids was also included. A negative control with equivalent volume of 100% ethanol (diluent for all PUFAs) was included to ensure that ethanol toxicity was not affecting growth.

Bacterial lipid extraction and thin-layer chromatography

Fourteen mL cultures were grown at 37 °C and lipids were extracted for analysis by thin-layer chromatography (TLC). Lipid extractions were performed according to the published method by Bligh and Dyer [18]. Phospholipids dried under nitrogen gas were applied to a silica gel 60 TLC plate. A solvent system of chloroform: methanol: acetic acid (65:25:10 v/v/v) was used to separate lipids in TLC glass chamber equilibrated for 2 h. A solution of 10% (v/v) sulfuric acid in ethanol was sprayed onto the plate, allowed to dry, and prior to development, and developed at 150 °C on a hot plate for approximately 1 min. The developed plate was scanned with a Canon CanoScan LiDE 210 and the image was adjusted only to provide more contrast for visualization of the lipid spots. Phospholipid controls (Figure S1) were purchased from Sigma (L-α-PE from egg yolk), Avanti Polar Lipids (L-α-PG from egg, chicken) and MedChemExpress (Cardiolipin from heart, bovine).

Ultra performance liquid chromatography/ESI-mass spectrometry (UPLC-MS)

Phospholipids were extracted from 20 mL cultures grown with or without fatty acid supplementation in CM9-HEPES (pH 7.4) minimal media at 37 °C. Lipids were extracted with Bond Elut Lipid Extraction SPE cartridges (Agilent) and dissolved in 95:5 methanol: chloroform. Samples were injected (5 uL) into ACQUITY UPLC H-Class (Waters) system for reversed-phase gradient elution using a BEH C-18 column (2.1 × 100 mm; 1.8 μm particles) heated to 40 C. Mobile phase A consisted of 30:70 1.25 mM ammonium acetate: methanol and mobile phase B was methanol. The 11 min gradient used a 0.4 mL/min flow rate and held 50%B for 2 min, ramped to 70% B, then to 100% B and held for 2 min before returning to initial conditions. Optima LC-MS grade solvents were used for sample preparation and mobile phase.

The LC effluent was directed into a Xevo TQD triple quadrupole mass spectrometer using negative ion electrospray ionization and operating in scan mode over the range m/z 200–850 with a scan time of 0.350 s. The capillary was held at 3 kV potential and the sample cone was held at 100 V. These conditions generate significant in-source cone fragmentation of eluting phospholipids to produce mass spectra with abundant [M-H]⁻ parent ion, in addition to substantial signal for the two most prominent cone fragments consisting of cleaved carboxylate ions [R₁COO⁻] and [R₂COO⁻]. Since the sn-2 fragment is generally more abundant than sn-1 [19, 20], this approach allows unambiguous determination of phospholipid composition by directly observing the masses of the deprotonated parent and the fatty acyl chains attached to sn-1 and sn-2. Resulting chromatograms and mass spectra were analyzed with MassLynx v4.2 software.

Crystal violet uptake assay

Cultures (3 mL) of A. hydrophila, unexposed or exposed to 300 µM of single PUFAs, were grown to OD₆₀₀ of approximately 0.90. These cells were centrifuged (2500 x g, 10 m), washed with phosphate buffered saline (PBS), and re-suspended to OD₆₀₀ 0.50 in 2 mL PBS. Crystal violet (5 µg/mL) was added and aliquots (650 uL) were removed every six minutes for 18 min during gentle agitation (120 rpm). Aliquots were pelleted (17,000 g, 1 m) and the supernatants were transferred to cuvettes for absorbance measurement (570 nm). Values were changed to percentage uptake by normalizing to a control containing CV, PBS, and no bacteria. Data is representative of two independent experiments.

Antimicrobial peptide susceptibility assay

A. hydrophila cultures were grown with 300 µM of each PUFA to OD₆₀₀ of approximately 0.8–0.9. After centrifugation, cultures were washed with CM9-HEPES and prepared as inoculum for the assay. Two-fold dilutions of polymyxin B and colistin were prepared in LoBind tubes in CM9-HEPES. Polypropylene microtiter plates were prepared by adding 30 µl of either CM9-HEPES or one of the antimicrobial peptide dilutions followed by 170 µl of prepared cultures to wells, such that each culture well contained 300 µM PUFA at the standard inoculum (5 × 10⁵ cfu ml^− 1) for microtiter broth dilution MIC determination [21]. Final concentrations covered the following concentration ranges, increasing twofold: polymyxin B and colistin ranged from 2.0 to 128 µg/ml. Plates were incubated at 37 °C in a microtiter plate shaker for 24 h. Each well was transferred to polystyrene microtiter plates and read in a BioTek Synergy microplate reader at 600 nm. Three biological replicates were performed in triplicate. Bacterial viability and death was further verified by sampling individual wells at each antimicrobial concentration for growth on Luria agar.

Beta lactam antibiotics susceptibility assay

These assays were performed in the same manner as described for the antimicrobial peptides. A. hydrophila inocula were prepared in the same manner described above. Ampicillin concentrations increased two-fold from 8 to 1024 µg/ml and imipenem concentrations increased twofold from 0.8 to 25.6 µg/ml.

Biofilm assay

Biofilm formation was analyzed using the glass tube method [22, 23]. Inoculum in CM9-HEPES (pH 7.4) was prepared from A. hydrophila cultures grown overnight in Luria broth. Individual PUFAs were administered at 300 µM in each borosilicate tube containing 1 mL of CM9-HEPES. Inoculum was added (OD₆₀₀ = 1.0) and tubes were incubated statically at 28 °C and 37 °C for 24 h. Established biofilms were gently washed three times with dH2O, and 1.1 mL of 0.1% CV solution was added. Tubes were incubated at room temperature for 20 min and CV solution was removed gently with dH₂O, and tubes were allowed to air dried. Stained biofilms were liberated by adding 1.1 ml of 30% acetic acid solution to each tube and incubating for 15 min at room temperature. Acetic acid-solubilized CV was transferred to cuvettes and read with a spectrophotometer at 570 nm.

Swimming motility soft agar assay

To assess swimming motility, soft agar was prepared (10 g L^− 1 tryptone, 10 g L^− 1 NaCl, 0.35% agar) as previously described [10]. 40 mL of molten agar was added to 50 mL conical tubes containing one of each PUFAs at a final concentration of 300 µM. The plates were inoculated in quadruplicate with 2 µL of a bacterial suspension at OD₆₀₀ of 1.0 in PBS. The plates were then incubated at 28–37 °C for 21 h. Bacterial motility was measured by averaging the diameter of the motility halo in each quadrant. Data represents three independent experiments performed in quadruplicate and statistical analysis was carried out by using student’s t-test (paired, two-tailed, p < 0.005).

Results

Growth characteristics of A. hydrophila in the presence of exogenous PUFAs

A. hydrophila is an ubiquitous inhabitant of aquatic sources around the world, where it populates the sediment, water column, and is readily isolated from shellfish [24, 25]. Acting as a primary or secondary fish pathogen, A. hydrophila not only wreaks havoc in aquaculture, but can also cause gastrointestinal and epidermal, dermal, and subcutaneous tissue infection in humans following ingestion and/or opportunistic infection following injury or trauma. We sought to examine the growth effects of PUFAs on A. hydrophila grown at two temperatures (28 °C and 37 °C) representing aquatic and host environments, while also testing the ability to utilize individual PUFAs as sole carbon sources. Overall, growth was slower at 28 °C, with 20:3 eliciting elevated growth during the first 4 h. At both temperatures, 4 PUFAs (18:3γ, 20:4, 20:5, and 22:6) slowed growth during at least the first 6 h. 18:3γ and 20:4 caused slowed growth for at least 10 h at 28 °C. (Fig. 1A). At 37 °C, the culture supplemented with 18:2 exhibited slower growth over the last 4 h. Strikingly, several fatty acids (18:2, 20:3, 20:5, and 22:6) allowed appreciable growth when serving as sole carbon sources (Fig. 1C & D).

Fig. 1.

Growth characteristics of Aeromonas hydrophila in the presence of exogenous fatty acids. A and B A. hydrophila was grown in CM9-HEPES (pH 7.4) media (starting OD = 0.05) for 12 h at 28 °C and 37 °C. C & D Exogenous fatty acids (1mM) were supplied in M9 minimal media (no glucose) as the sole carbon source and growth of both strains was monitored for 12 h. Data shown is average of at least two biological replicates. Standard deviations were all < 0.03 at each measured time

Phospholipid TLC profiles of A. hydrophila indicate exogenous PUFA-mediated structural alteration

The major phospholipids of A. hydrophila are phosphatidylethanolamine (PE), phosphatidylglycerol (PG) and cardiolipin (CL) [26]. Chromatographic separation of these phospholipids yielded observable migratory shifts in some, but not all, species. While noticeable shifts may correspond to more hydrophobic species (reflecting incorporation of the fatty acid containing increased carbon number and/or unsaturation), other lanes such as 18:3a and 20:5 lack upward mobility (Fig. 2). Possible explanations include (i) differential total lipid loads, which is based on culture volume, growth of bacteria, and efficiency of lipid extraction and (ii) variable incorporation of each PUFA into each phospholipid species. It is speculated that the doublets of CL (arrows) may reflect incorporation of more than one exogenously supplied PUFA, likely resulting from diphosphatidylglycerol synthesis from two singly modified PG species. Although unconfirmed, future studies are aimed at confirming structures of each species by performing preparative TLC to isolate and analyze each spot. These results qualitatively suggest incorporation of most of the supplemented fatty acids into phospholipids of A. hydrophila. Notably, the cultures grown with fatty acids of 20 carbon length yielded spots at the top of the plate, probably reflecting unassimilated free fatty acid. Thus, A. hydrophila likely does not incorporate the entirety of 300 µM fatty acid, an observation made previously [10] and the focus of ongoing experiments. A control TLC plate (Figure S1) shows migration of each major phospholipid, as well as linoleic acid to verify approximate location of the lipid species.

Fig. 2.

Thin-layer chromatography of phospholipids extracted from Aeromonas hydrophila grown in the presence of individual polyunsaturated fatty acids. Bacteria were grown to exponential phase (OD ≈ 0.8) in CM9-HEPES (PH 7.4) media at 37 °C with or without 300 µM of the indicated fatty acids (linoleic acid [18:2], alpha-linoleic acid [18:3α], gamma-linolenic acid [18:3γ], dihomo-gamma-linolenic acid [20:3], arachidonic acid [20:4], eicosapentaenoic acid [20:5] and docosahexaenoic acid [22:6]) prior to Bligh and Dyer extraction of phospholipids and separation by TLC in the solvent system chloroform/methanol/acetic acid (65:25:10 v/v/v). The plate was charred and scanned to produce the final image. Arrows indicate more hydrophobic phospholipid species (of cardiolipin) indicative of PUFA assimilation

UPLC/ESI-MS analyses confirm assimilation of exogenous PUFAs into A. hydrophila phospholipids

Phospholipid compositions were unambiguously identified using the novel mass spectrometry approach described herein. By using scan mode with a high cone voltage (100 V), without the use of collision gas and without collision energy, we performed untargeted lipidomics to get a complete lipid profile rather than having to rely on a priori knowledge of expected analytes and their retention time that is typically required for multiple reaction monitoring (MRM) analyses. The control culture (Fig. 3) shows the native PG and PE compositions are comprised of fatty acyl chains with ≤ 18 carbons and ≤ 1 double bond. Chromatographic signal intensities reflect the fact that 16:0 is the most abundant fatty acid in most life forms.

Fig. 3.

UPLC-MS results of isolated phospholipids from Aeromonas hydrophila. Base peak intensity chromatogram of the control culture of Aeromonas hydrophila showing the native phospholipid composition of phosphatidylglycerol and phosphatidylethanolamine. An example mass spectrum is shown for the 9.1 min peak; the parent ion (m/z 716.5) and cone fragments (m/z 255.3 and 281.3) establish the identity as PE 16:0/18:1. All other chromatographic peaks were similarly analyzed, and compositions are labeled above each peak. For both PG and PE, the native compositions are comprised of fatty acyl chains with ≤ 18 carbons and ≤ 1 double bond. This chromatogram was mass filtered for all PG and PE compositions that might contain the tested exogenously supplied polyunsaturated fatty acids, but only electronic noise was present

Phospholipid profiles of all samples – control and all tested fatty acids – were obtained by UPLC-MS in order to determine if supplemented fatty acids could be incorporated into membrane phospholipids. An example chromatogram and mass spectrum from the culture grown in the presence of 20:3 is shown in Fig. 4 – all others can be found in the Supplementary Figs. 2–6. Excellent chromatographic resolution was obtained with the optimized gradient such that phosphatidylglycerols were well-resolved from phosphatidylethanolamines (Fig. 4A). PGs, being slightly more polar than PEs, elute first on the reversed-phase gradient. Even species that differ by one double-bond (e.g. PG 18:0/20:3 and PG 18:1/20:3) are completely baseline resolved. Some co-elution does occur when the effective carbon numbers are very similar. Thus PG 16:0/20:3 and PG 18:1/20:3 nearly co-elute as the double bond on 18:1 provides enough polarity to almost counteract the increased hydrophobicity resulting from 2 extra carbons compared to 16:0.

Fig. 4.

UPLC-MS results of isolated phospholipids from Aeromonas hydrophila grown in the presence of 20:3. A. hydrophila was grown with or without 300 µM of a given fatty acid at 37 °C in CM9-HEPES (pH 7.4) to logarithmic phase (OD ≈ 0.8). Lipids were extracted with Bond Elut Lipid Extraction SPE cartridges (Agilent) and reconstituted in 95:5 MeOH: CHCl₃ for LC-MS analysis. Ions were detected by scanning quadrupole mass spectrometry following electrospray ionization. The culture grown in the presence of 20:3 is here used as an example; all tested cultures and the control were similarly analyzed (see Supplement). A Overlaid extracted ion chromatograms from the culture grown in the presence of 20:3. PLs that result from incorporation of the supplemented fatty acid are labeled. Both PG and PE are observed to be modified. The control culture was mass filtered for all of these signals, but only electronic noise was present. B Mass spectrum of the chromatographic peak that elutes at 8.75 min. The parent ion at m/z = 740.5 is consistent with many possible phosphatidylethanolamine species corresponding to PE 36:3 (www.lipidmaps.org). The observed cone fragments at m/z = 255.2 and 305.2, along with their relative intensities, establish the correct identity as PE 16:0/20:3. Similar analysis from all experimental cultures (see Supplement) establishes that A. hydrophila can modify its membrane phospholipids by recognizing, uptaking, and incorporating each of the tested exogenously supplied polyunsaturated fatty acids

Phospholipids are complex analytes that can vary in composition due to variable combinations of carbon number and degree of unsaturation of the attached fatty acyl chains. As a result, a single m/z value of a parent phospholipid can be consistent with many different compositions. If using a very mild (i.e. “soft”) set of ionization conditions, one may only observe the parent ion. The current method uses high cone voltage (100 V) to generate in-source fragmentation, and by directly observing RE: (Final) revision for manuscript the masses and intensities of the cone fragments along with the parent, unique compositions could be determined. For example, an m/z value of 740.6 (Fig. 4B) is consistent with PE 36:3 composition. This composition could be achieved with many combinations of fatty acid side chains. The direct observation of cone fragments matching 16:0 and 20:3 acyl chains establishes the correct identity as PE 16:0/20:3. A similar analysis of all chromatographic peaks from all experimental cultures (see Supplement) establishes that A. hydrophila is capable of recognizing, uptaking, and incorporating each of the supplemented PUFAs into the membrane phospholipids, which include phosphatidylglycerol and phosphatidylethanolamine. The incorporation is so extensive that even doubly-substituted phospholipids, such as PG 20:3/20:3 and PE 20:3/20:3, are observed (Fig. 4A). Supplementary data (Tables S1-S6) provides retention times, masses and corresponding phospholipid identities from PUFA-supplemented samples.

Exogenous PUFAs affect hydrophobic compound uptake in A. hydrophila

Given alteration to the membrane phospholipids, a hydrophobic compound uptake assay was used to assess permeability. Three PUFAs (18:3a, 20:4, and 22:6) caused significant (p < 0.001) permeabilization of the bacterial membrane (Fig. 5).

Fig. 5.

The effect of exogenous fatty acids on hydrophobic compound uptake in Aeromonas hydrophila. Bacteria were grown in CM9-HEPES (pH 7.4) with and without 300 µM of the indicated fatty acids to mid-log phase (OD = 0.8). Cultures were gently pelleted, washed with PBS and resuspended in an equal volume of PBS (OD₆₀₀ = 0.6). The amount of crystal violet in the supernatant following centrifugation was measured at regular intervals and expressed graphically as percentage of uptake. Results represent means of two independent determinations of hydrophobic compound uptake. All standard deviations were less than 5% (not graphed for visual clarity) and asterisks indicate significant difference (p < 0.001) as compared to control. Shown is one experiment with biological duplicates representative of two experiments

Exogenous PUFAs induce susceptibility to polymyxin B and colistin in A. hydrophila

The combination of phospholipid remodeling and permeability effects caused by PUFAs justified determination of the minimum inhibitory concentration of the membrane active cyclic peptides polymyxin B and colistin. When bacteria were grown with individual PUFAs and allowed continual access to the PUFA during the MIC assay, significant decreases in the MICs of both peptides were observed (Fig. 6). For polymyxin B, three PUFAs (20:3, 20:5, and 22:6) lowered the MIC by 4-fold, two PUFAs (18:2 and 18:3 g) lowered the MIC by 8-fold, and one PUFA (18:3a) caused a 16-fold decrease in MIC (Fig. 6C). Colistin resistance was lowered 16-fold by 18:3 g, 8-fold by 18:2 and 18:3a, 4-fold by 20:5, and 2-fold by 20:3 and 22:6 (Fig. 6D). We also tested two beta lactam antibiotics as controls, since these drugs are transported through membrane proteins and would not be expected to be affected by PUFA incorporation in phospholipids. No significant changes to MIC were observed for imipenem; however, two PUFAs (18:3a and 22:6) lowered the MIC to ampicillin (Fig. 6A and B). It is generally accepted that a 2-fold difference in broth dilution MIC assays is not considered a statistically significant finding.

Fig. 6.

The effect of exogenous fatty acids on imipenem, ampicillin, polymyxin B, and colistin resistance in Aeromonas hydrophila. Bacteria were grown at 37 °C in CM9-HEPES (pH 7.4) with and without 300 µM of the indicated fatty acids to mid-log phase (OD₆₀₀ = 0.8–0.9). Cultures were pelleted, washed with CM9 and resuspended in CM9 to an OD₆₀₀ of 0.1. Fatty acids were added to a final concentration of 300 µM. The bacterial suspension was distributed into microtiter plates and two-fold concentrations of (A) imipenem, (B) ampicillin, (C) polymyxin B or (D) colistin were added. After 20 h incubation, the optical density (600 nm) was read using a Biotek Synergy microplate reader. Experiments were conducted in triplicate, with each value representing the mean (standard deviations not graphed for clarity; all were < 0.04). Shown is one experiment with biological triplicates representative of three experiments. Symbols circled by dotted line indicate significant differences (p < 0.001) as compared to the control (no fatty acid) at the particular antimicrobial concentration

PUFAs impact biofilm formation in A. hydrophila

Biofilm formation in A. hydrophila is linked to its success in colonizing gastrointestinal epithelia and elevated virulence in the septicemic mouse model [27–29]. To determine the effects of PUFAs on biofilm formation, we analyzed biofilm formation on borosilicate glass tubes after 24 h at 28 °C and 37 °C. Elevated biofilm formation was observed for all cultures, except 22:6, when grown at 28 °C. Three PUFAs (18:2, 18:3 g, and 22:6) significantly (p < 0.005) lowered biofilm formation at 28 °C, while no significant changes were observed at 37 °C (Fig. 7).

Fig. 7.

Exogenous fatty acids influence temperature-dependent biofilm formation in Aeromonas hydrophila. Overnight cultures were grown at 37 °C and then pelleted, washed, resuspended in appropriate media and inoculated onto microtiter plates (starting OD ~0.1) in octuplet. Each culture was then grown at 28 °C and 37 °C in the presence of 300μM of the indicated fatty acids. After 24 h incubation, the biofilm assay by O’Toole was performed in CM9-HEPES (pH 7.4) minimal media at (A) 28 ºC and (B) 37 °C. The absorbance (OD₅₉₀) was measured using a Biotek Synergy microplate reader. Shown is one experiment with biological octuplets representative of three experiments. Significant differences in biofilm formation were determined by Student’s t-test and are indicated by asterisks (p < 0.005)

PUFAs impact swimming motility in A. hydrophila

Aeromonads possess the ability to alter motility in response to a variety of small molecules, including amino acids, proteins, and mucins [30]. Furthermore, the presence of lux gene homologs indicates quorum sensing capacity for A. hydrophila and flagella have been implicated in participating in virulence characteristics such as biofilm formation [31, 32]. We employed soft agar motility assays to examine the effect of PUFAs on swimming motility at 28 °C and 37 °C. Motility was higher overall at 28 °C, with 20:3 and 20:5 causing a significant decrease (p < 0.005) in swimming motility (Fig. 8A). At 37 °C, 18:2 increased swimming motility and 18:3 g decreased motility (Fig. 8B).

Fig. 8.

Exogenous fatty acids influence temperature-dependent swimming motility in Aeromonas hydrophila. Overnight cultures were grown at 37 °C and then pelleted, washed, resuspended in phosphate buffered saline and inoculated onto soft agar motility plates (2uL of culture prepared at OD_600nm = 1.0) in quadruplicate. Plates were incubated at 28–37 °C for 21 h. Bacterial motility was measured by averaging the diameter of the motility halo in each quadrant. Data represents three independent experiments and statistical analysis was carried out by using student’s t-test (paired, two-tailed, p < 0.005)

Discussion

A. hydrophila is a Gram-negative bacterium equipped with many virulence factors that enhance pathogenicity in its aquatic and human victims. Accordingly, substantial focus has centered on mechanisms defining key virulence traits such as biofilm formation and motility [33–36]. The efforts to mitigate financial loss from aquaculture hinge on combating A. hydrophila strains that are developing high levels of resistance to antimicrobials [37–40]. Thus, the need for aquaculture remediation and infection control of A. hydrophila persists. In the wake of numerous studies documenting PUFA-mediated responses in gram-negative bacteria [6–12], we investigated the ability of A. hydrophila to utilize PUFAs for phospholipid remodeling and control of phenotypes associated with virulence.

Growth curves of A. hydrophila were conducted at 28 °C and 37 °C in minimal media supplemented with glucose and casamino acids. Optimal growth temperatures for pathogenic A. hydrophila are reported to be between 25 °C and 37 °C, with little growth at temperatures as low as 5 °C with no growth seen above 55 °C [41]. The presence of four PUFAs (18:3γ, 20:4, 20:5, and 22:6) noticeably delayed growth, resulting in a longer (~ 2 h) lag phase. As observed by Statner et al. when they demonstrated that 28 °C is not necessarily the optimal temperature for motile aeromonads, our motile, pathogenic ATCC strain exhibited a shorter lag phase and higher rate of exponential phase growth at 37 °C [42]. A. hydrophila’s ability to utilize PUFAs as sole carbon sources surpasses observations in several of its aquatic relatives (A. salmonicida and Vibrio spp.) [7, 11, 12]. It may be worth noting that many fish species that can harbor A. hydrophila require PUFAs used in this study as essential fatty acids [43, 44].

To determine whether A. hydrophila was able to assimilate PUFAs into its membrane, phospholipids were extracted from A. hydrophila cultures grown at 37 °C in the presence or absence of 300 µM of each PUFA. While TLC offered qualitative indication of assimilation, ultra performance liquid chromatography-mass spectrometry (UPLC-MS) confirmed fatty acid incorporation of all tested PUFAs into each of the major species of membrane phospholipids – PG and PE. This indicates that the biochemical pathways leading to synthesis of phosphatidylglycerol and phosphatidylethanolamine are capable of recognizing, uptaking, and utilizing all of the supplied PUFAs.

Phenotypic interrogation included assays for permeability, biofilm formation, motility, and susceptibility to antimicrobials. The assimilation of PUFAs into phospholipids could have a variety of effects on membrane dynamics, altering barrier properties controlling resistance to acid, temperature, and antimicrobial stress [45, 46]. The hydrophobic compound uptake experiment indicated significantly (p < 0.001) heightened permeability when gamma-linolenic, arachidonic and docosahexaenoic acids are available. The increase in unsaturation, particularly in the cis-configuration, would be expected to increase permeability, and a similar observation was made in A. salmonicida [12]. Interestingly, the increase in permeability induced by these three PUFAs closely mirrors the pattern observed in the more pathogenic A-layer, as opposed to the non-A-layer, strain of A. salmonicida [12]. It remains unknown, although curious, how some PUFAs, despite similar carbon length and number of double bonds, exert diverse permeability effects.

Multi-drug resistance is a major concern for A. hydrophila, particularly in settings such as aquaculture and food/water consumption [47–51]. Antimicrobial peptides (AMPs) are attractive therapeutics due to their lower potential for resistance and their ability to enhance activity of conventional antibiotics [52, 53]. Every PUFA tested in this study caused increased vulnerability of A. hydrophila to both PMB and colistin. Several PUFAs lowered the MIC by at least 8-fold, underscoring the synergistic potential of PUFAs and AMPs in the battle against disease. There was no other rationale for PUFA effectiveness (e.g., omega-3 vs. omega-6), other than the permeabilizing effect of 18:3γ translating to highly significant susceptibility with both PMB and colistin.

The increased permeability caused by 18:3γ, 20:4, and 22:6 predicts increased vulnerability to AMPs which target the bacterial membrane. Both PMB and colistin permeabilize the outer and inner membranes [54, 55]. Indeed, these PUFAs lowered MICs to both cyclic peptides; however, since all PUFAs decreased the MIC, our findings did not correlate membrane permeability (as measured by CV) and change in MIC. As consistently observed in other Gram negative bacteria [6–12], PUFA incorporation does not affect susceptibility to cell wall-targeting antibiotics (ampicillin and imipenem) that utilize protein transport to reach their site of activity [56, 57]. A recent analogous study in our laboratory listed the homologs of the fatty acid handling machinery in Aeromonas spp., including A. hydrophila [12]. The identification of these homologs (for FadL, FadD, PlsB, PlsC, PlsX, and PlsY) further supports the expanded fatty acid assimilation/utilization capability for A. hydrophila. Furthermore, bioinformatic assessment of A. hydrophila ATCC 7966 using Integrated Microbial Genomes (https://img.jgi.doe.gov/cgi-bin/m/main.cgi) identified two enzymes that participate in phospholipid modifications (cyclopropane fatty acid synthase [AHA_1106] and delta-9 desaturase [AHA_0366]). These enzymes participate in modifications to fatty acid moieties on existing phospholipids, and thus, may be able to act on externally acquired PUFA constituents of phospholipids.

The ability to form biofilms confers an advantage for bacterial pathogens, allowing them to not only persist on/in a variety of locations, but also to create opportunity for developing resistance to stress, particularly to antibiotics. Biofilm formation in A. hydrophila has been associated with elevated risk to aquaponic produce, food processing equipment, and other food products [58–60]. Our temperature-dependent biofilm formation aligned with previous studies that observed elevated (approximately two-fold) biofilm formation of A. hydrophila at 28 °C versus 37 °C [61, 62]. The only PUFA-mediated effects on biofilm formation were observed at 28 C, with 18:2, 18:3 g, and 22:6 acting as biofilm suppressors. Biofilm assays were performed in microtiter plates and, while the relative amount of biofilm formation between temperatures was consistent, there was much lower overall detection of biofilm in addition to inconsistency within experimental replicates (data not shown). The data suggests a larger role of PUFA-mediated biofilm control in the aquatic environment compared to human infection; 18:2 is a major fatty acid in fish mucus, 18:3 g is rich in algae, and 22:6 is prevalent in fish tissue [63–65]. Since linoleic acid is the most abundant fatty acid in skin epithelia and the temperature range of human skin is 33–37 °C [66, 67], it may be beneficial to examine the interval to narrow the potential host niche impact. Additionally, there exists the prospect of PUFA application as pre- and probiotic supplements to control Aeromonas in aquaculture [68].

As a fish and human pathogen, A. hydrophila has likely developed PUFA detection and response pathways for the purposes of survival, persistence, and pathogenesis. Ongoing projects include exploring the effects of omega-3 and omega-6 fatty acids on A. hydrophila survival against antimicrobial peptides of human and fish origin.

Conclusions

The findings herein identify yet another gram negative bacteria of medical importance that utilizes polyunsaturated fatty acids for various purposes. Considering the diverse environments inhabited by A. hydrophila, the demonstrated physiological and behavioral responses to exogenous fatty acids suggests evolutionary adaptive capability worthy of further investigation. Temperature-dependent impacts on growth, biofilm formation and motility implicate PUFAs as ligands for undefined metabolic and regulatory pathways driving survival and pathogenesis. Importantly, numerous PUFAs significantly lowered the MICs of polymyxin B and colistin, alluding to potential therapeutic applications in both aquaculture and human disease.

Abbreviations

PUFA

Polyunsaturated fatty acid

PG

Phosphatidylglycerol

PE

Phosphatidylethanolamine

CL

Cardiolin (diphosphatidylglycerol)

Authors’ contributions

Author’s contributions – MAG, DKG, and SJS were major contributors in writing the manuscript. GLIM, KFC and DKG performed, analyzed and interpreted the thin-layer chromatography. MAG and DKG performed, analyzed and interpreted the antimicrobial peptide and beta lactam MIC assays. MAG, LDB, GLIM, and CRH performed the growth curve and motility assays. MAG, LDB, GLIM and DKG performed, analyzed and interpreted the biofilm and crystal violet uptake assays. SJS performed, analyzed and interpreted the UPLC/MS data. All authors have read and approved this manuscript.

Funding information

This work received no external funding source. This work was supported by various internal awards to MAG, GLIM, SJS and DKG.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.