Cracking the code of seasonal seawater biofouling: enhanced biofouling control with quorum sensing inhibitor-functionalized membranes

Seawater reverse osmosis (SWRO) is widely deployed for desalination but suffers from membrane biofouling driven by bacterial attachment, proliferation, and EPS-rich biofilm formation, affecting ~70% of SWRO systems. Conventional chlorine-based pretreatment can damage RO polyamide layers, and biocides may select for resistant bacteria and leave dead biomass as nutrient sources, allowing regrowth from small surviving fractions. Quorum sensing (QS) coordinates bacterial behaviors, including biofilm formation, via signaling molecules; quorum sensing inhibitors (QSIs) disrupt these pathways, offering eco-friendly, low-toxicity, non-disinfectant mitigation of biofouling. Prior works showed that acylase (AC) and methyl anthranilate (MA) can inhibit AHL and PQS pathways and reduce Pseudomonas aeruginosa biofilms; immobilization of enzymes on membranes enhances stability and activity versus free dispersal and avoids secondary pollution. However, most biofouling studies used model strains, not reflecting complex seawater consortia, and seasonal shifts in seawater microbial communities and EPS composition can change fouling behavior and severity. This study asks how seasonal microbial dynamics influence SWRO biofouling and whether QSI-functionalized membranes (AC- and MA-modified) can regulate community assembly, EPS secretion, and biofilm formation to mitigate fouling.

The paper contextualizes that RO biofouling is a persistent limitation in desalination with prior studies documenting EPS roles in fouling, kinetic fouling behaviors, and shortcomings of chlorine/biocides (e.g., resistance, membrane damage). QS-based control has been explored: AC and MA reduce biofilm formation by targeting AHL and PQS systems; immobilized acylase shows ~90% higher enzymatic activity on membranes versus free enzyme. Most earlier investigations relied on single model bacteria, whereas full-scale plants exhibit diverse, seasonally varying communities with different biofilms across winter, summer, and autumn. Seasonal differences in EPS composition/amounts suggest season-dependent fouling propensities. Prior work by the authors showed AC and MA reduced EPS secretion and flux decline in RO/FO and that MA suppressed PQS and AHL synthesis gene expression in P. aeruginosa. The unresolved gaps are the roles of QSIs in real seawater microbial interspecies communication, community assembly, and seasonal biofilm development on SWRO membranes.

Membrane fabrication: SWRO polyamide membranes were modified by first coating with polydopamine (PDA; 2 g L−1 dopamine in 0.05 M Tris-HCl, pH 8.5, 1 h), then grafting glutaraldehyde (GA; 6 wt% in 50 mM PBS, pH 7, 2 h) to provide aldehyde groups, followed by covalent immobilization of inhibitors (AC: 1 mg mL−1 in PBS; MA: 3 mM; 12 h at 4 °C) via Schiff base chemistry, yielding AC_m and MA_m. Membranes were pre-soaked in ethanol and DI water and mounted with the active layer exposed. Characterization included FTIR/XPS (evidence of C=N formation and increased C–N), zeta potential, contact angle (hydrophilicity/wettability), and surface morphology (Supplementary Fig. 1, Table 1).

Biofouling experiments: Natural seawater (Bohai Bay, Tianjin, China; seasons: January-winter, May-spring, July-summer, October-fall) was prefiltered (0.45 µm; UF for feed conditioning) and physicochemically characterized (conductivity, temperature, pH 7.8–8.2, DOC 0.66–1.43 mg L−1). Microbial consortia were enriched from seawater (Supplementary Method 2). A crossflow SWRO unit (effective area 204 cm²) operated at 6.0 MPa, 25 ± 1 °C, tangential velocity 0.07–0.135 m s−1. Prior to tests, the system was cleaned/disinfected (NaClO, ethanol), and membranes compacted to stable flux. For each seasonal run, 100 mL of exponential-phase consortia (OD=1) were washed and dosed into 10 L pretreated NSW; 0.1% Bacto Marine Broth 2216 plus trace elements and vitamins were added to accelerate biofouling. Tested membranes included pristine (control) and QSI-modified (AC_m, MA_m) across WI, SP, SU, FA. Flux and salt rejection were monitored via data acquisition.

Foulant and biofilm analyses: EPS extraction from fouled coupons (NaOH 1 M, mild sonication; overnight extraction; 0.45 µm filtration). Quantification included proteins (BCA/Bradford), polysaccharides (phenol–H2SO4), eDNA (PicoGreen), TOC, ATP. Biofilm imaging and thickness via CLSM after LIVE/DEAD staining (SYTO9/PI), with ImageJ analysis for cell fluorescence and thickness.

Microbial community and functional analysis: 16S rRNA gene high-throughput sequencing of biofilm microbiota (Supplementary Method 3); alpha/beta diversity, LEfSe for biomarkers, Venn core OTUs, and phylogenetic molecular ecological networks (pMENs) with Spearman correlations to fouling factors (EPS components, biofilm thickness, TOC). Predictive functional profiling mapped to KEGG pathways (QS, biofilm formation, amino acid and carbohydrate metabolism), and key gene abundance differences (e.g., glmU, galU/galF, serB, ilvE, hom, ilvD) were evaluated.

• Membrane performance: Pristine SWRO membranes exhibited severe flux decline of 70–77% across seasons, worst in summer (77%). QSI-modified membranes alleviated decline: AC_m by 30–32% and MA_m by 18–22% relative to pristine. Salt rejection was slightly higher for QSI-modified membranes (≈98.5%) than pristine (≈97.5%).
• EPS and biofilm: On pristine membranes, EPS loads (µg cm−2) varied seasonally: winter 104, spring 110, summer 129, fall 118; polysaccharide fractions were higher in spring (44%), summer (42%), and fall (43%) than winter (23%). AC_m and MA_m significantly reduced total EPS by 65–69% and 55–59%, respectively. Component-wise decreases on AC_m: polysaccharides 42–62%, proteins 72–75%, eDNA 59–65%; on MA_m: polysaccharides 47–49%, proteins 60–66%, eDNA 50–57%. CLSM showed total cells reduced by 70–73% (AC_m) and 60–67% (MA_m); biofilm thickness decreased from ~60 µm (control) to ~17.5 µm (AC_m) and ~20 µm (MA_m); live/total ratios were similar across groups, indicating non-bactericidal growth delay.
• Community diversity and composition: Seawater Shannon diversity higher in warm seasons (FA > SU > SP > WI). Biofilm diversity on membranes was significantly reduced vs seawater (P<0.05), with winter biofilms lowest in diversity; QSI-modified membranes further reduced Shannon indices relative to controls. Proteobacteria dominated seawater (58–64%) and biofilms (44–59%), with Bacteroidetes as secondary (seawater 11–32%; biofilm 21–47%). On membranes, Gammaproteobacteria (14–48%) and Alphaproteobacteria (4–27%) dominated; Epsilonproteobacteria were low-abundance (3–8%) yet impactful. Seasonal shifts showed Cyanobacteria enrichment in winter seawater (~24%).
• Keystone taxa and correlations: Venn analysis identified 59 core OTUs implicated in biofouling. Dominant biofilm genera included Vibrio, Rhodobacteraceae, Flavobacterium, Shewanella, Lutibacter, Arcobacter, Acinetobacter, Tenacibaculum, Erythrobacter. pMENs and Spearman correlations revealed keystone taxa Rhodobacteraceae, Vibrio, Arcobacter strongly positively correlated (P≤0.01) with EPS and biofilm content. Alphaproteobacteria (Rhodobacteraceae) and Acinetobacter positively correlated with polysaccharides (P≤0.01), consistent with higher polysaccharide fractions in SP/SU/FA biofilms. Additional contributors included Cryomorphaceae and NS7 marine group. In summer, Erythrobacter strongly correlated with EPS/biofilm, aligning with the most severe fouling.
• QSI effects on community: AC_m significantly reduced Gammaproteobacteria by 72% (P<0.05); MA_m reduced Epsilonproteobacteria by 90% (P<0.05). At finer taxonomic levels, MA_m decreased Arcobacter by >90%; Rhodobacteraceae were nearly absent on AC_m. Both QSI membranes decreased overall Proteobacteria relative abundance (control 51.1% to 21.8% on AC_m and 29.7% on MA_m). QSI enhanced the abundance of QQ-related genera (e.g., Flavobacterium, Tenacibaculum, Glutamicibacter) associated with biofilm control.
• Functional impacts: Predictive functional profiling showed reduced abundance of QS, biofilm formation, and amino acid/carbohydrate metabolism functions on QSI membranes (KEGG). Key carbohydrate metabolism genes (e.g., glmU; galU/galF involved in UDP-N-acetylglucosamine and UDP-glucose pathways) were downregulated on AC_m and MA_m. Amino acid metabolism genes serB and ilvE were reduced by 38% and 35% on AC_m and by 36% and 33% on MA_m vs control; hom (homoserine) decreased significantly on MA_m; ilvD decreased significantly on AC_m. These disruptions align with decreased EPS (proteins/polysaccharides) biosynthesis and biofilm formation.

The study demonstrates that seasonal variability in seawater microbial communities translates into distinct biofouling behaviors on SWRO membranes, with higher diversity and EPS-positive taxa in warmer seasons driving more severe fouling (notably summer). Keystone taxa—Rhodobacteraceae, Vibrio, and Arcobacter—exhibit strong positive correlations with EPS components and biofilm metrics, directly linking community composition to fouling severity. QSI-functionalized membranes manipulate these communities by targeting QS pathways: AC preferentially suppresses AHL-related Proteobacteria (notably Rhodobacteraceae and Gammaproteobacteria), while MA more strongly suppresses Epsilonproteobacteria (e.g., Arcobacter), reshaping biofilm communities toward QQ-capable taxa. Importantly, QSI effects are non-bactericidal, reducing biofilm accumulation and EPS secretion while preserving live/dead ratios, thus minimizing selection for resistance. Functionally, QSIs downregulate carbohydrate and amino acid metabolism genes critical for EPS biosynthesis (e.g., UDP-sugar and amino acid pathways), providing a mechanistic basis for reduced EPS and improved hydraulic performance. The alignment of microbiome shifts, functional reductions, and performance outcomes across seasons underscores QS inhibition as an effective, season-robust anti-biofouling strategy for SWRO.

Seasonality significantly shapes seawater microbial diversity and, consequently, SWRO biofouling. Proteobacteria dominate biofilms, with Gamma- and Alphaproteobacteria prevalent and low-abundance Epsilonproteobacteria acting as keystones. Elevated polysaccharide-producing taxa during warm seasons underlie higher EPS fractions and severe fouling, particularly in summer. QSI-functionalized membranes (AC_m and MA_m) robustly mitigate biofouling across seasons by suppressing EPS-associated Proteobacteria (e.g., Arcobacter, Rhodobacteraceae), reducing QS, biofilm formation, and key amino acid/carbohydrate metabolism pathways (e.g., glycine/serine/threonine; amino/nucleotide sugar metabolism), thereby decreasing EPS secretion and biofilm thickness and improving salt rejection stability. These findings highlight QS inhibition as a promising, environmentally friendly approach for seasonally resilient biofouling control in SWRO and motivate further development of targeted QSI membrane functionalizations and field validations.