Mechanism and Risk Control of Chlorine-Resistant Bacteria in Drinking Water Supply Systems: A Comprehensive Bibliometric Analysis

Abstract

Ensuring safe drinking water is a global priority, with pathogen control being an essential aspect. Chlorine disinfection is widely adopted for its affordability and potent antimicrobial effects. However, certain bacteria, known as chlorine-resistant bacteria (CRB), can still survive in water systems with residual chlorine, posing risks to water quality and distribution systems. Their emergence, ironically, can be partially attributed to the very application or increased dosage of chlorine disinfectants in certain cases, which unintentionally promoted the selection and adaptation of CRB in the environment. Despite their significance, research on CRB remains fragmented, with few systematic reviews or bibliometric analyses. Thus, this study addresses the gap by analyzing 1367 publications (1984–2025) regarding CRB in drinking water supply systems (DWSSs) using CiteSpace. Important aspects including typical species, potential risks, resistance mechanisms, and inactivation methods are reviewed. Contributions from key countries/institutions/journals/authors are also examined. More importantly, overlooked issues like CRB’s impact on taste and odor (T&O) issues in water and their molecular resistance mechanisms are also highlighted. The identification of these gaps in CRB research motivates further studies on their hazards, intrinsic mechanisms and control, which would hopefully help with the delivery of high-quality, safe drinking water worldwide.

1. Introduction

Ensuring drinking water safety, a global priority, relies heavily on the proper control of pathogenic microorganisms in water quality management processes. To safeguard public health and prevent waterborne diseases, microbial safety management in drinking water currently involves effective control, removal, or inactivation of pathogens through processes such as filtration, disinfection (e.g., chlorination, ozonation, UV irradiation, etc.), and maintenance of distribution systems. Among all the approaches used for disinfection, chlorination has been one of the most commonly used techniques due to its effectiveness, affordability, and ease of application [1,2]. Chlorine is highly efficient at inactivating a great variety of microorganisms, including health-threatening bacteria and viruses, ensuring water is safe for consumption. In addition, it helps maintain a residual chlorine presence within the water distribution system, which can substantially inhibit microbial regeneration and potential recontamination over time and distance, up to the point of use at the customer end [3]. Its long history of successful application has further solidified its role as a standard disinfection method worldwide. However, with the advancement of microbial monitoring technology, researchers have found that chlorine disinfection and the presence of residual chlorine cannot inactivate all microbes in the water supply system, and some of them may be opportunistic pathogens which can form biofilms in the distribution network [4,5]. Since these specific bacteria can survive in water treatment plants and distribution systems with a certain level of residual chlorine concentration, they are usually called “chlorine-resistant bacteria” (CRB) [1,6].

Currently, the general awareness about the hazards of CRB in water supply systems mainly focuses on their risks to public health, infrastructure, and aquatic environment. For example, certain waterborne diseases are linked to infections caused by Mycobacterium avium complex, which exhibits resistance to chlorine at specific concentrations and can proliferate within water distribution systems, posing a risk to vulnerable populations [7,8]. Meanwhile, these bacteria can also compromise water quality by forming biofilms, which accelerate pipe corrosion, reduce disinfection efficacy, and release harmful byproducts. Currently, most research related to this hazard are focused on biofilm formation and pipe corrosion [8,9,10], few research has recognized that CRB may also contribute to taste and odor (T&O) problems in drinking water [11]. For example, Actinomycetes, which are frequently found in biofilm, soft deposits and bulk water of drinking water distribution systems (DWDSs) [12], have long been linked to flavor issues in drinking water [12,13]. They are also found to be less susceptible to disinfectants than other bacteria due to their spore-forming ability [12,14]. Therefore, the possible hazard of CRB on drinking water T&O is highlighted and discussed in the prospect section of the study. Regarding environmental hazards, lots of studies have recognized that some CRB may also be multi-drug resistant and can spread resistance genes in aquatic ecosystems, increasing the risk of the proliferation of “superbugs” and global antimicrobial resistance [3,15].

In view of the various hazards of CRB, several physical, chemical or combined methods [16,17] were developed to combat them, which could roughly be categorized into UV-based [5,18,19] method, ozone-based [20,21] method and chlorine dioxide (ClO₂) disinfection [22,23]. It is worth noting that there have been few studies developing biological methods for CRB handling [24]. Meanwhile, since CRB can also cause or aggravate membrane biofouling in some cases [17,25], many studies have focused on developing novel reverse osmosis membranes [26] or nanofiltration membranes [27,28] with antibiofouling and chlorine resistant properties. However, most of these approaches are either costly, inadequately researched, or have not been widely tested or applied in large-scale practical projects. Therefore, there is a need for further exploration in innovative mitigation strategies for CRB, which would allow for more cost-effective, efficient, and easier implementation.

To develop effective control methods, a deeper and more comprehensive understanding of the resistance mechanisms of CRB is essential. At present, research in this field has mainly been focusing on changes of cell structure (i.e., morphology, membrane structure, cell hydrophobicity, efflux system), biofilm formation (i.e., aggregation ability), extracellular polymeric substances (EPS) secretion, enzyme activity (i.e., superoxide dismutase (SOD) activity) and energy metabolism (i.e., total adenosine triphosphate (ATP) content) [9,29,30,31]. There is no good understanding, at the molecular level, about the intrinsic chlorine-resistant molecular metabolic mechanism of CRB in DWDSs [32], and yet limited investigation has been done on this front to provide the much needed fundamental understanding of bacterial chlorine resistance [29].

Although the existence and risks of CRB have received widespread attention from researchers, there is currently no unified international standard for determining if strains can be categorized as CRB [1]. Current research employs a variety of methods to evaluate chlorine resistance and identify CRB, which include logarithmic removal rate, concentration-time (CT) values, minimum inhibitory concentrations (MICs), inhibition zone on plates, survival time in certain concentration of free chlorine, and reference strains method. These approaches have been thoroughly detailed and systematically reviewed in the recent work by Wang, et al. [10] and are therefore not further discussed here. It was also in the same work Wang, et al. combined the logarithmic removal rate with the reference strain E. coli and proposed it as an appropriate method. The proposal was based on the summary of existing evaluation approaches, which provides a standard range for chlorine resistance across different strains and facilitates the analysis of the presence of CRB in the environment. In addition, the World Health Organization (WHO) introduced a notable method in 2011 for assessing the chlorine resistance of pathogens. This method measures the survival time of pathogenic microorganisms when exposed to a standard disinfection dosage of 0.5 mg/L free chlorine. Chlorine resistance is categorized into three levels, depending on whether the survival time exceeds 1 min or 30 min, forming a systematic framework for classification [33].

The extensive use of chlorine disinfectants around the globe has, to some extent, facilitated the selection and domestication of CRB in the environment. In some cases, water utilities have resorted to using higher disinfectant doses during wastewater treatment. For instance, when treated wastewater is recycled for landscaping or irrigation use, up to 2000 mg-min/L of chlorine, a relatively high dosage, can be applied to ensure effective disinfection and reduce biofouling in irrigation systems [34]. However, higher disinfectant levels can lead to the formation of carcinogenic disinfection byproducts (DBPs) [24,35], while also contributing to the proliferation of bacteria with increased chlorine resistance [1,20]. Therefore, comprehensive research on CRB is essential for ensuring microbial safety and delivering high-quality water supply.

To the best of our knowledge, there is a lack of systematic bibliometric analyses on chlorine-resistant bacteria (CRB) in drinking water supply systems (DWSSs). Therefore, this study utilized CiteSpace to analyze research trends related to CRB from 1984 to 2025, covering key aspects such as typical species, potential risks, resistance mechanisms, and inactivation methods of CRB. Additionally, the often-overlooked issue of CRB contributing to taste and odor (T&O) problems in drinking water is also raised and highlighted, supported by preliminary laboratory findings. This research offers a comprehensive overview and valuable reference for future studies aimed at improving drinking water safety.

2. Research Data and Methods

2.1. Data Collection and Search Strategy

In this study, CiteSpace (Version 6.4.R1) was used to analyze information from chlorine-resistant bacteria (CRB) related literature, while the Web of Science (WoS) Core Collection database served as the data source. WoS is recognized internationally as a leading citation database that indexes high-impact journals [36], and it is also well supported by CiteSpace for data analysis. It provided us with a great starting point for examining and analyzing the work with the most influence in the field. Due to subscription limitations at the authors’ institution, only selected portions of WoS core collection were accessible for this research. More specifically, our analysis incorporated data from the following indices: Science Citation Index Expanded (SCI-EXPANDED, from 1984 to present), Social Sciences Citation Index (SSCI, from 1984 to present), Arts and Humani-ties Citation Index (AHCI, from 2012 to present), Conference Proceedings Citation Index–Science (CPCI-S, from 2001 to present, Emerging Sources Citation Index (ESCI, from 2019 to present), Current Chemical Reactions (CCR-EXPANDED, from 1985 to present), and Index Chemicus (IC, from 199 to present) [37].

To retrieve publications related to research on CRB in water systems, the following search terms were used: Topic = (“chlorine” OR “chlorination” OR “*chloramine” OR “chloramination” OR “hypochlorite” OR “hypochlorous acid” OR “Cl₂” OR “NH₂Cl” OR “NaClO” OR “HClO”) AND (“resistant*” OR “resistance” OR “tolerance”) AND “bacter *” AND “water”. The search was done on Tuesday, 28 January 2025, 10.22 a.m. Beijing time. The database, which began in 1984, yielded 1367 English publications up to 2025, including articles, review articles, proceeding papers and early access. Among these, 1244 were classified as “articles”, 108 as “review articles”, 53 as “proceeding papers” and 7 as “early access” records.

This study also has its limitations, as the results are certainly influenced by the selection of the data source. As mentioned earlier, the authors’ institution subscribed to a subset of the indices in WOS core collection, but it does not have access to Book Citation Index-Science (BKCI-S), Book Citation Index-Social Sciences and Humanities (BKCI-SSH). Additionally, despite WoS Core Collection being widely recognized for its comprehensive coverage, consistency, and high-quality data, it may not always offer the most recent or complete information, potentially introducing slight biases in the analysis. Nevertheless, considering the extensive timeframe and substantial number of publications reviewed, any identified trends or insights are unlikely to be significantly affected by these limitations [37,38].

2.2. Bibliometric Analysis Methods

CiteSpace is a software that enables visual analysis of collaboration networks among authors, institutions, and countries, through which research collaboration patterns at various levels can be uncovered. It identifies development trends and research hotspots through keyword co-occurrence analysis and reveals research field dynamics via literature co-citation analysis. In this study, node types including “institution”, “country”, “keyword” and “reference” were used, parameters were set to analyze annual slices, with k in the g-index at 30 for “reference” and 25 for the other node types, the top N at 50, and the top N% at 10. By integrating these analyses, a comprehensive review of publications was conducted to gain insights into the typical species, potential risks, resistance mechanisms, and inactivation methods of CRB in drinking water supply systems (DWSSs).

3. Results

3.1. Bibliometric Analysis

3.1.1. The Number of Published Articles

Analyzing the annual publication statistics of chlorine-resistant bacteria (CRB), as shown in Figure 1, the trend of literature growth can be roughly divided into two phases, the initial phase of gradual growth (phase 1) followed by a rapid expansion phase (phase 2), with 2010 serving as the dividing point. Prior to 2010, the annual publication counts on CRB remained below 40, showing a slow upward trend. However, after 2010, the rate of published papers increased significantly, indicating a rapid acceleration in research activity, especially in the last 5 years (from 2021). Since data collection extended up to January 2025, 2024 recorded the highest number of publications, reaching 134. This publication trend underscores the growing global interest in CRB, highlighting it as an emerging research hotspot over the past 15 years.

3.1.2. The Major Countries and Institutions

Relationship graphs of countries and institutions highlighted significant contributors in this field. Here, node type of “country” and “institution” were selected separately to do the analysis on the respective network maps, and the results were shown in Figure 2 (countries) and 3 (institutions). As can be seen from Figure 2, the 1367 papers in this field come from 93 countries, corresponding to the 93 nodes in the graph, out of which 7 countries have purple outlines, indicating their high centrality in the network. This means they are important in connecting different countries for cooperation. These include USA (betweenness centrality = 0.43), Peoples R China (0.28), Belgium (0.28), Spain (0.17), Denmark (0.13), Germany (0.12), and France (0.11). Meanwhile,

Table 1. Top 10 countries/institutions/journals/authors by number of published articles.

Rank	Countries	Number of Published Articles	Institutions	Number of Published Articles	Journals	Number of Published Articles	Authors	Number of Published Articles
1	PEOPLES R CHINA	421	Chinese Academy of Sciences	96	Water Research	123	Hu, Hongying	21
2	USA	346	Tsinghua University	46	Science of the Total Environment	71	Ye, Chengsong	21
3	JAPAN	70	University of Chinese Academy of Sciences (UCAS)	39	Applied and Environmental Microbiology	47	Wu, Yinhu	15
4	SPAIN	52	Tongji University	34	Journal of Hazardous Materials	41	Simões, Manuel	14
5	AUSTRALIA	51	Harbin Institute of Technology	33	Environmental Science Technology	40	Wang, Haibo	13
6	ENGLAND	51	Research Center for Eco Environmental Sciences (RCEES)	28	Chemosphere	36	Yuan, Yixing	12
7	SOUTH KOREA	51	United States Department of Agriculture (USDA)	28	Water Science and Technology	34	Feng, Mingbao	11
8	INDIA	50	Zhejiang University	27	Journal of Food Protection	30	Simões, Lúcia C	11
9	FRANCE	47	Xiamen University	23	Journal of Applied Microbiology	28	Peng, Shi	11
10	CANADA	44	University System of Georgia	22	Food Control	22	Yu, Xin	11

showed the top 10 countries, institutions, journals and authors by number of published articles. As can be seen in the Table 1, while China has a slightly lower betweenness centrality than the USA, it leads in the total number of publications. The dominance of the US and China when per country analysis is performed is hardly surprising, as the sizes of their respective research community and economy are much greater than others. The larger numbers of researchers in these countries lead to more academic outputs and there is also a higher probability that members of their research community will take part in international collaboration. On the other hand, smaller countries in regions with a high degree of economic and political integration can have high centrality as well. The researchers’ collaboration can easily extend beyond their respective national border, which is exemplified by the European countries in the list.

As for institutions, which can be seen in Figure 3 and Table 1, a total of 828 institutions were involved in research of CRB from 1984 to 2025. Most articles were published in the last 10 years, as the annual ring-shaped circles representing the temporal distribution of publications from each institution over different years are mainly orange to red in color. The top ten institutions in terms of the number of publications were Chinese Academy of Sciences (CAS) (96), Tsinghua University (46), University of Chinese Academy of Sciences (UCAS) (39), Tongji University (34), Harbin Institute of Technology (33), Research Center for Eco Environmental Sciences (RCEES) (28), United States Department of Agriculture (USDA) (28), Zhejiang University (27), Xiamen University (23) and University System of Georgia (22).

3.1.3. Literature Co-Citation Analysis

The literature co-citation analysis provides insights into the dynamics of a research field. The results of this study are presented in Figure 4 (cluster relationship graph), Figure 5 (timeline view of co-citation reference clusters), Figure 6 (top 25 references with the strongest citation bursts), and

Table 2. Top 25 references with high co-citation count.

Rank	Co-Citation Count	Author	Years	Publications
1	67	Jin M	2020	Jin M, 2020, ISME J, V14, P1847, DOI 10.1038/s41396-020-0656-9
2	61	Luo LW	2021	Luo LW, 2021, WATER RES, V188, P0, DOI 10.1016/j.watres.2020.116474
3	56	Liu SS	2018	Liu SS, 2018, WATER RES, V136, P131, DOI 10.1016/j.watres.2018.02.036
4	47	Zhang TY	2019	Zhang TY, 2019, CHEM ENG J, V358, P589, DOI 10.1016/j.cej.2018.09.218
5	39	He H	2019	He H, 2019, ENVIRON SCI TECHNOL, V53, P2013, DOI 10.1021/acs.est.8b04393
6	37	Wang LP	2021	Wang LP, 2021, ENVIRON SCI TECHNOL, V55, P9221, DOI 10.1021/acs.est.1c00645
7	37	Yoon Y	2017	Yoon Y, 2017, WATER RES, V123, P783, DOI 10.1016/j.watres.2017.06.056
8	36	Shi P	2013	Shi P, 2013, WATER RES, V47, P111, DOI 10.1016/j.watres.2012.09.046
9	36	Stange C	2019	Stange C, 2019, INT J HYG ENVIR HEAL, V222, P541, DOI 10.1016/j.ijheh.2019.02.002
10	34	Su HC	2018	Su HC, 2018, SCI TOTAL ENVIRON, V616, P453, DOI 10.1016/j.scitotenv.2017.10.318
11	33	Sanganyado E	2019	Sanganyado E, 2019, SCI TOTAL ENVIRON, V669, P785, DOI 10.1016/j.scitotenv.2019.03.162
12	31	Khan S	2016	Khan S, 2016, CHEMOSPHERE, V152, P132, DOI 10.1016/j.chemosphere.2016.02.086
13	31	Jia SY	2015	Jia SY, 2015, ENVIRON SCI TECHNOL, V49, P12271, DOI 10.1021/acs.est.5b03521
14	30	Liu LZ	2019	Liu LZ, 2019, CHEMOSPHERE, V219, P971, DOI 10.1016/j.chemosphere.2018.12.067
15	30	Zheng J	2017	Zheng J, 2017, CHEM ENG J, V317, P309, DOI 10.1016/j.cej.2017.02.076
16	30	Ding WQ	2019	Ding WQ, 2019, WATER RES, V160, P339, DOI 10.1016/j.watres.2019.05.014
17	30	Hou AM	2019	Hou AM, 2019, WATER RES, V156, P366, DOI 10.1016/j.watres.2019.03.035
18	28	Zhang HC	2019	Zhang HC, 2019, ENVIRON SCI TECHNOL, V53, P2141, DOI 10.1021/acs.est.8b05907
19	28	Xu LK	2016	Xu LK, 2016, ENVIRON POLLUT, V213, P119, DOI 10.1016/j.envpol.2016.02.013
20	28	Zhang YY,	2015	Zhang YY, 2015, SCI TOTAL ENVIRON, V512, P125, DOI 10.1016/j.scitotenv.2015.01.028
21	27	Wang HC	2020	Wang HC, 2020, WATER RES, V185, P0, DOI 10.1016/j.watres.2020.116290
22	27	Jia SY	2020	Jia SY, 2020, WATER RES, V176, P0, DOI 10.1016/j.watres.2020.115721
23	25	Zhang Y	2017	Zhang Y, 2017, ENVIRON SCI TECHNOL, V51, P570, DOI 10.1021/acs.est.6b03132
24	25	Chen S	2018	Chen S, 2018, WATER RES, V142, P279, DOI 10.1016/j.watres.2018.05.055
25	22	Pazda M	2019	Pazda M, 2019, SCI TOTAL ENVIRON, V697, P0, DOI 10.1016/j.scitotenv.2019.134023

(top 25 references with high co-citation counts). The co-citation cluster analysis generated eight clusters: #0 antibiotic-resistant bacteria, #1 public health, #2 reverse osmosis, #3 metagenomic analysis, #4 drinking water distribution system, #7 Aeromonas, #10 Escherichia coli O157:H7, and #11 peracetic acid, which are discussed in subsequent sections. Cluster numbering is non-continuous due to size-based assignments, dynamic formation, filtering, and the influence of time slicing and network modularity, which are to ensure only significant clusters are highlighted. References with a high co-citation count are those frequently cited alongside others over time, representing the foundational literature. In contrast, references with a high citation burst are those experiencing a rapid and significant increase in citations, indicating emerging or trending research. The top 25 references in both categories are shown in Figure 6 and Table 2.

3.1.4. Keywords Co-Occurrence Analysis

Keywords co-occurrence analysis helps identify key research hotspots in a field. The findings of this study are illustrated in Figure 7 (timeline view of strong citation bursts keywords and their clusters) and Figure 8 (top 39 keywords with strong citation bursts). As shown in the figures, keywords with strong citation bursts are primarily grouped into eight clusters: #0 antibiotic resistance gene, #1 antibiotic-resistant bacteria, #2 dual-species biofilm, #3 reverse osmosis membrane, #4 hydrogen peroxide, #5 chlorine disinfection by-products, #6 electrochemical disinfection, and #7 cooling water system. Additionally, the top 39 keywords with strong citation bursts include chlorine dioxide, opportunistic pathogens, drinking water distribution systems (DWDSs), reverse osmosis membranes, resistance genes, horizontal transfer, ultraviolet disinfection, microbial community, extracellular polymeric substances (EPS), and pipe materials. These keywords represent critical research trends and are discussed in detail in the subsequent sections.

3.2. Typical Species of CRB in Drinking Water Supply Systems

Although chlorine-resistant bacteria (CRB) have garnered significant attention from researchers in recent years, there is still no international standard for their identification or for quantifying their resistance, making it challenging to compare resistance levels across different strains in various studies [1]. Thus, in studies, CRB is usually used to refer to bacteria which are highly resistant to chlorine disinfection, or those that can survive or even regenerate in the presence of residual chlorine [21]. Common quantification of these traits include logarithmic removal rate, concentration-time (CT) values, minimum inhibitory concentrations (MICs), inhibition zone on plates, survival time in certain concentration of free chlorine, and reference strains method [1].

According to the literature survey, the CRB reported by researchers were mostly single bacteria strain screened and isolated from various environments, and tested for chlorine resistance [30,39,40]. As can be seen from

Table 3. Typical species, sources and chlorine resistance of isolated CRB in studies.

Phylum	Genus	Species	Source	Chlorine Resistance	Ref
Proteobacteria	Pseudomonas	Pseudomonas peli	Drinking water distribution systems (DWDS)	CT value method The CT value to achieve 99.9% inactivation of the P. peli was 51.26–90.36 mg min/L, inversely proportional to the free chlorine concentration.	[22]
		Pseudomonas aeruginosa	Chlorinated river water for drinking purpose	Treated with various doses of chlorine at room temperature for 24 h and 48 h, Colony count was significantly higher for the resistant strain at higher concentrations of chlorine.	[41]
		Pseudomonas aeruginosa	Hospital drinking water systems	WHO Survival time method P. aeruginosa isolates exhibited resistance to 0.5 mg/L chlorine for both 5- and 30-min exposure durations. When exposed to a higher chlorine concentration (1.5 mg/L), 80% of the isolates were able to survive after a 5-min exposure, with 40% remaining viable even after a 30-min exposure.	[39]
		Pseudomonas sp.	Drinking Water	WHO Survival time method Treated by 2 mg/L free chlorine for 30 min, the bacteria were not completely inactivated.	[40]
	Klebsiella	Klebsiella sp.
		Klebsiella pneumoniae	Cooling water system	MIC method 80% of K. pneumoniae survival at 2 mg/L chlorine for 30 min at 30 °C.	[42]
	Aeromonas	Aeromonas jandaei	Drinking water treatment plant (DWTP)	Isolated from DWTP water samples with bacterial growth could not be effectively controlled by an increase of sodium hypochlorite disinfectant. Also, B. alvei, B. cereus, and Lysinibacillus fusiformis exhibits spores with strong resistance to common disinfectants.	[20]
		Aeromonas sobria
	Vogesella	Vogesella perlucida
	Pelomonas	Pelomonas sp.
	Acinetobacter	Acinetobacter sp.	Water distribution system	WHO Survival time method Resistant to 20 mg/L chlorine for 30 min.	[43]
	Serratia	Serratia sp.
	Sphingomonas	Sphingomonas sp.	Model DWDS	4 mg/L chlorine with 240 mm retention time provided only approximately 5% viability reduction of the strain.	[44]
	Burkholderia	Burkholderia sp.	Tap water	Isolated bacteria as reference + Inhibition zone method Treated by 14.5% standard NaClO, the diameter of the inhibition zone is less than 20 mm.	[3]
	Acidovorax	Acidovorax sp.
	Halomonas	Halomonas boliviensis	Marine biofilm	Reference strain + Logarithmic removal rate method Using the chlorine-sensitive bacterium Pseudoalteromonas espejiana as a reference strain, H. boliviensis showed only a ≤1-fold reduction in viable count, even after prolonged chlorine exposure (4–8 h) at 8 mg/L residual chlorine.	[31]
	Phaeobacter	Phaeobacter caeruleus	Marine biofilm	The viable P. caeruleus cell numbers in chlorine-treated samples (0.4 mg Cl2/L for 30 min were higher than that in the control sample, showing the stimulation of microbial growth by chlorine.	[18]
Firmicutes	Bacillus	Bacillus alvei	DWTP	Three bacteria were found survived in and were isolated from a finished water under 0.3 mg/L residual chlorine for 30 min, thus operationally defined as chlorine-resistant bacteria.	[16]
		Bacillus cereus
		Bacillus alvei	DWTP	Isolated from DWTP water samples with bacterial growth could not be effectively controlled by an increase of sodium hypochlorite disinfectant. Also, B. alvei, B. cereus, and Lysinibacillus fusiformis exhibits spores with strong resistance to common disinfectants.	[20]
		Bacillus cereus
		Bacillus cereus	DWTP	Reference strain + CT value method The inactivation rate of B. cereus species was 2-log lower than that of Escherichia coli at 1 mg/L NaClO.	[5]
		Bacillus sp.	Tap water	Isolated bacteria as reference + Inhibition zone method Treated by 14.5% standard NaClO, the diameter of the inhibition zone is less than 20 mm.	[3]
		Bacillus sp.	Lake Water	Treated by 0.5 mg/L free chlorine for 30 min, the live-to-dead ratio of 8 strains was between 0.3–4.4.	[25]
	Lysinibacillus	Lysinibacillus fusiformis	DWTP	Isolated from DWTP water samples with bacterial growth could not be effectively controlled by an increase of sodium hypochlorite disinfectant. Also, B. alvei, B. cereus, and Lysinibacillus fusiformis exhibits spores with strong resistance to common disinfectants.	[20]
		Lysinibacillus fusiformis	DWTP	Three bacteria were found survived in and were isolated from a finished water under 0.3 mg/L residual chlorine for 30 min, thus operationally defined as chlorine-resistant bacteria.	[16]
	Paenibacillus	Paenibacillus sp.	Tap water	Isolated bacteria as reference + Inhibition zone method Treated by 14.5% standard NaClO, the diameter of the inhibition zone is less than 20 mm.	[3]
	Clostridium	Clostridium sp.	Drinking Water	WHO Survival time method Treated by 2 mg/L free chlorine for 30 min, the bacteria were not completely inactivated.	[40]
	Staphylococcus	Staphylococcus aureus
Actinobacteria	Mycobacterium	Mycobacterium fortuitum	Water distribution system	CT value method For a CT value of 60 mg·min/L, frequently used in water treatment lines, chlorine disinfection inactivates over 4 log units of M. gordonae and 1.5 log units of M. fortuitum or M. chelonae. CT values determined under similar conditions show that even the most susceptible species, M. aurum and M. gordonae, are 100 and 330 times more resistant to chlorine than Escherichia coli.	[45]
		M. chelonae
		M. gordonae
		M. aurum
		M. mucogenicum	Water distribution system	Treated by 2 mg/L free chlorine for 60 min, the inactivation rate is 3.2 log.	[8]
	Legionella	Legionella pneumophila	Spring Water	CT value method The Legionella with the strongest chlorine resistance (of 20 strains) has a CT99.9% of 0.62 mg·min/L.	[46]
		Legionella pneumophila	Cooling water	MIC method The MIC50 values of bacteria are between 256 and 1024 mg/L free chlorine.	[47]
	Gordonia	Gordonia	DWDS	CT value method Exhibited high tolerance to chlorine with a CT value of 120 mg min/L for 99.9% reduction.	[23]

, the main focus of current research is on drinking water supply systems (DWSSs), the main classification at the phylum level include Proteobacteria, Firmicutes and Actinobacteria, which can be further divided at the genus level as follows: Proteobacteria (Pseudomonas, Klebsiella, Aeromonas, Vogesella, Pelomonas, Acinetobacter, Serratia, Sphingomonas, Burkholderia, Acidovorax, Halomonas, Phaeobacter), Firmicutes (Bacillus, Lysinibacillus, Paenibacillus, Clostridium, Staphylococcus) and Actinobacteria (Mycobacterium, Legionella, Gordonia). Pseudomonas aeruginosa (citation bursts time 2018–2019) and Escherichia coli O157:H7 (citation bursts time 2002–2016), belonging to the Proteobacteria, were among the keywords with the strongest citation bursts in Figure 8, showing their trend during that time. Overall, there are still limited species of CRB that were isolated, further investigation is needed in the future.

Beyond individual bacterial strains, bacterial communities also represent a key research focus. However, studies in this area often fail to isolate bacterial populations that exhibit a relative increase under chlorine disinfection or systematically assess their chlorine resistance. Additionally, there is no standardized approach for identifying CRB in studies examining changes in bacterial community structure. Luo, et al. [1] proposed that bacterial communities with a 200% increase in relative abundance after disinfection could be classified as CRB. However, this criterion has not been consistently adopted in research. For example, Wang, et al. [9] examined the effects of varying chlorine residual concentrations on species diversity and bacterial community composition in both planktonic and biofilm samples. Their findings revealed that biofilm samples had greater microbial richness than planktonic ones. In planktonic communities, Proteobacteria and Actinobacteria remained dominant across all chlorine levels, whereas in biofilms, Proteobacteria were gradually replaced by Actinobacteria as chlorine concentration increased. Moreover, higher chlorine residuals were shown to have facilitated biofilm formation by Gram-positive bacteria, highlighting their role in biofilm adaptation to chlorine stress.

3.3. Risks of CRB in Drinking Water

At present, most of the research on the hazards of chlorine-resistant bacteria (CRB) focuses on the following three aspects: public health, infrastructure, and aquatic environment.

3.3.1. Influence on Public Health

CRB in water supply systems can threaten public health (the main cluster #1 public health shown in bibliometric analysis in Figure 5), as some are pathogenic or opportunistic pathogens (shown as keywords like “pathogenic bacteria”, “opportunistic pathogens” with high citation bursts in Figure 8). Furthermore, CRB can proliferate within drinking water distribution systems (DWDSs), leading to increased bacterial concentrations [48]. For example, Lu, et al. [23] isolated Gordonia, a novel CRB, from the drinking water supply system in Jinan City. It is a rare but emerging human pathogen which can cause infections on skin and soft tissue infections, bacteremia, and mastitis, in both immunocompromised and immunocompetent hosts. Gholipour, et al. [39] isolated Pseudomonas aeruginosa (chlorine-resistant) from hospital drinking water systems, which can cause various health issues ranging from postoperative wound infection to bacteremia and sepsis [1]. Meanwhile, from drinking water treatment facilities in West Bengal, Roy and Ghosh [40] isolated different species of CRB (including Staphylococcus aureus, Klebsiella, Pseudomonas and Clostrdial), which in most cases, are opportunistic pathogens. Both Staphylococcus aureus and Klebsiella can cause a variety of self-limiting to life-threatening diseases in humans, including pneumonia, sepsis, wound or surgical site infections, meningitis, etc. If these microorganisms are not effectively removed during water purification or inactivated by disinfection, their entry into the pipeline network can significantly elevate the risk of users contracting waterborne infectious diseases.

3.3.2. Influence on Water Supply Infrastructure

CRB pose significant challenges to water supply infrastructure, due to their ability to withstand disinfection by modifying their cell structure and protein composition. This resistance allows them to deplete residual chlorine, promoting bacterial regrowth and biofilm development in water supply systems, which further enhances their resistance. This cycle amplifies bacterial contamination in distribution networks, exacerbating biofilm accumulation, pipeline corrosion, harmful byproducts release, and dangerous pathogens spread [8,9,10]. This problem domain attracted the attention of the research community, as illustrated by the main clusters of #2 dual-species biofilm and #5 chlorine disinfection by-product in Figure 7, as well as keywords including “microbial community”, “extracellular polymeric substances”, “pipe materials” in Figure 8. For example, the research of Wu, et al. [48] showed that the regrowth potential of CRB in chloraminated water samples correlates with the biodegradable organic carbon in water, where increase in chloramine dosage can induce change in the community structure of bacteria. Du, et al. [49] revealed that chlorine, resource availability, and spatio-temporal variations can jointly affect bacterial movement and EPS secretion. Chlorination, coupled with nutrient levels, suppressed nutrient-driven cell motility and EPS production while enhancing interactions between bacterial cells and surfaces. This process facilitated bacterial attachment to surfaces, ultimately promoting biofilms development. Shan, et al. [50] investigated chlorine resistance and biofilm formation across three common domestic pipe materials: polypropylene random (PPR), stainless steel (SS), and copper, to understand their role in microbial pollution. PPR pipes exhibited the highest biofilm biomass, while SS pipes had the lowest biofilm and EPS levels. Copper pipes showed the highest EPS content despite lower biomass. Bacterial chlorine resistance varied slightly across pipe materials and was also a function of the physiological state of cells, such as EPS levels. Zhang, et al. [51] studied the impact of disinfection on reclaimed water using sodium hypochlorite (NaClO) and chlorine dioxide (ClO₂) in annular reactors containing new cast iron coupons over multiple 30-day periods. Which found that disinfection accelerated the corrosion process, with the effect being most pronounced during the first 30 days. In this initial stage, the strong oxidizing properties of disinfectants were the primary drivers of corrosion. Between 30 to 90 days, however, the bacterial community played a more significant role in influencing the corrosion process.

Despite the aforementioned research work, we have found few studies focusing on harmful byproducts released by CRB like taste and odor (T&O) substances [11]. The authors believe that this is a very important and currently ignored research direction and will further discuss it in the outlook section, based on the combination of literature research and the preliminary experiment results from the lab.

3.3.3. Influence on Aquatic Environment

Regarding environmental hazards, antibiotic resistance and antibiotic resistance genes (ARG) of CRB have been the research hotspots in recent years [52]. Antibiotic resistant bacteria showed up as #0 cluster in Figure 5 and #1 cluster in Figure 7, while antibiotic resistance gene is the #0 cluster in Figure 7. Related keywords such as “resistance genes”, “horizontal transfer”, “pharmaceuticals” can also be seen in Figure 8, as multi-drug resistant CRB may propagate resistance genes through horizontal gene transfers in aquatic ecosystems, increasing the risk of the spread of “superbugs” and global antimicrobial resistance [3,15]. For example, Jia, et al. [15] investigated the patterns of ARGs and bacterial community changing in a drinking water treatment and distribution system. The results showed that there is a significant increase in ARGs total relative abundance and a reduction of diversity in the opportunistic bacteria. The key factor driving the shift was the residual chlorine. Pseudomonas and Acidovorax, both resistant to chlorine, carried multidrug resistance genes and bacitracin resistance gene bacA. The relative abundance of these bacteria increased as a result of chlorination, and this change in bacterial community structure consequently led to an increase of ARGs abundance. Ma, et al. [53] employed metagenomic analysis alongside culture methods to investigate the impact of chlorination on ARGs and their bacterial hosts, including total microbes and E. coli. By simulating chlorination dosages in environments relevant to human health, such as drinking water and swimming pools, it was observed that although the number of ARGs decreased after chlorination, the remaining ones were mainly carried by microorganisms. Khan, et al. [3] examined 148 surviving bacteria from chlorinated water systems and identified correlations between chlorine tolerance and resistance to antibiotics such as tetracycline, sulfamethoxazole, and amoxicillin. Also, in the presence of free chlorine, antibiotic-resistant bacteria were found to survive longer than their antibiotic-sensitive counterparts. Additionally, spore-forming bacteria showed greater tolerance to disinfectants.

In addition, the horizontal transfer of ARG among the surviving bacteria post disinfection deserves attention. Shi, et al. [54] examined how microbial antibiotic resistance is influenced by chlorination in a drinking water treatment plant. Their findings revealed that chlorination increased the proportion of surviving bacteria resistant to antibiotics such as chloramphenicol, trimethoprim, and cephalothin. Metagenomic analysis further confirmed that chlorination could concentrate various ARGs along with plasmids, insertion sequences, and integrons involved in their horizontal transfer. Jin, et al. [55] explored the horizontal transfer of ARGs during chlorination, emphasizing the role of ARGs released from dead antibiotic-resistant bacteria (ARB) and their uptake by culturable chlorine-injured bacteria. E. coli, Salmonella aberdeen, Pseudomonas aeruginosa, and Enterococcus faecalis exhibited different levels of resistance to sodium hypochlorite. When these bacteria were chlorine-injured, with enhanced oxidative stress response and membrane permeability, they demonstrated up to 550 times higher transformation frequency of plasmid RP4 compared to untreated counterparts. The findings revealed that chlorination promotes horizontal ARG transfer via natural transformation, leading to new ARB emergence and the conversion of non-ARB into ARB, posing significant public health risks. Zhong, et al. [52] identified two chlorine-resistant bacterial strains with varying antibiotic resistance after mixed culture experiments, revealing that antibiotic resistance can spread horizontally between different bacterial species. It is thus possible for non-pathogenic bacteria to act as carriers, transferring resistance to pathogens and ultimately posing a threat to human health. In water supply networks, interactions between antibiotics, ARGs, and chlorine disinfectants can lead to bacterial adaptations, resulting in contamination and pollution of drinking water.

3.4. Resistance Mechanism of CRB

Each disinfectant employs its own unique mechanism to inactivate bacteria, and different microorganisms can exhibit varying resistance mechanisms to the same disinfectant. Current research on the chlorine resistance mechanisms of CRB primarily focuses on two aspects, external chlorine resistance related to cell permeability barriers or chlorine-consuming substances, and internal resistance determined by genetic materials.

For external chlorine resistance, the main mechanisms include changes in biofilm formation (e.g., aggregation ability), extracellular polymeric substances (EPS) secretion and cell structure (e.g., morphology, hydrophobicity, membrane composition, efflux systems, and spore coats) [9,29,30,31], which are typically observed simultaneously.

Biofilm formation is one key strategy for chlorine resistance (dual-species biofilm is the #2 main cluster in Figure 5). Bacteria aggregate on surfaces, creating protective layers that shield them from chlorine exposure. These biofilms also trap organic material, reducing chlorine’s penetration and effectiveness. For example, Schwering, et al. [56] investigated chlorine tolerance in multi-species biofilms formed from environmental drinking water. It was observed that the pathogens within multi-species biofilms had greater chlorine tolerance, 50–300 times higher than those in single-species biofilms, demonstrating that co-colonization in multi-species biofilms significantly enhances chlorine resistance. Zhu, et al. [57] reported that dual-species biofilms involving synergistic interactions with Sphingomonas sp. exhibited higher EPS production compared to other dual-species biofilms. Factors such as Interspecific interactions, biomass, EPS production, biofilm thickness, and compactness were all associated with chlorine resistance. Specifically, synergistic biofilms are typically denser and thicker, with increased EPS secretion, which reduced disinfectant penetration into the biofilm matrix. Consequently, these synergistic dual-species biofilms frequently display enhanced resistance to chlorine. Shan, et al. [58] showed that pipe materials can also affect biofilm characteristics, with chlorine resistance differing among biofilms due to variations in their intrinsic community properties. Zhu, et al. [59] investigated chlorine resistance and biofilm development in multi-species biofilms on two outdoor pipe materials, polyethylene (PE) and cast iron. The findings revealed that EPS and corrosion products shielded bacteria from chlorination. While intrinsic properties of the microbial community contributed to chlorine resistance, the same community could exhibit varying resistance depending on the pipe material.

EPS (one of the keywords with high citation bursts in Figure 8) secretion is closely related to biofilm formation as well. Studies have shown that EPS can account for 50–80% of the organic matter of biofilms, sometimes even over 60% of the total biomass in certain bacterial biofilms [60]. For example, Liu, et al. [61] investigated how chlorination affects EPS production and its role in early-stage biofilm formation. The results showed that EPS, especially proteins, is important and can promote biofilm growth. Removing cell-attached proteins and polysaccharides inhibited biofilm formation, and chlorination, by altering microbial protein production and cell surface charge, also influenced biofilm formation. Research has also indicated that EPS serves as the outermost protective layer of bacteria, which is capable of consuming free chlorine and shielding cells from chlorine attack, thus improving chlorine resistance [57,60,62]. For example, Luo, et al. [60] found a significant positive correlation between chlorine resistance of the tested strains and their membrane fouling potential. In biofouling layers formed by highly chlorine resistant strains, extracellular substances were found to cause serious fouling. EPS amount per cell was identified as the quantity which links chlorine resistance to the fouling potential. Furthermore, the study indicated that EPS was the main contributor to chlorine resistance for bacterial strains exposed to a chlorine concentration of 0.5 mg/L. Xue, et al. [62] performed a comprehensive investigation of capsular EPS, comparing whole-cell and extracted EPS. Their findings showed that capsular EPS present on bacterial cell surfaces could consume residual disinfectants, thereby diminishing their effectiveness in inactivating bacteria. Additionally, capsular EPS provides protective effects by reducing membrane permeabilization caused by oxidative disinfectants, as evidenced by alterations in functional groups. Although disinfectants decreased the intensity of functional moieties on bacterial cells and EPS, complete removal did not occur, suggesting minimal membrane damage at low chlorine concentrations typical in water distribution systems. Overall, these results implied that capsular EPS enhanced bacterial survival either by neutralizing disinfectants or by restricting their interaction with the cell membrane.

Additionally, some bacteria possess cellular adaptations enhancing chlorine resistance. For instance, changes in cell wall structure can reduce chlorine permeability. Parvin, et al. [63] found that Staphylococcus aureus biofilms resist both antibiotics and disinfectants due to cell wall protection. Biofilm-forming bacteria showed higher activity in cell wall synthesis proteins than planktonic cells. Furthermore, biofilm culture duration also influenced cell wall structure, with increased cell wall thickness and peptidoglycan production observed over time. Correspondingly, disinfectant tolerance was highest in 12-day dry surface biofilms, followed by 12-day hydrated biofilms, 3-day biofilms, and lowest in planktonic cells, highlighting the role of cell wall modifications in biocide resistance. Efflux pumps, known for removing harmful substances such as chlorine, play a critical role in bacterial disinfectant resistance. For instance, Wang, et al. [64] examined differences in viability, morphology, enzyme activity, and gene expression patterns between chlorine-resistant and chlorine-sensitive strains. Their findings indicated that, upon chlorine exposure, the sensitive strain experienced more pronounced membrane damage and greater depletion of intracellular ATP compared to the resistant strain. Transcriptomic analysis revealed substantial up-regulation of efflux pump systems (proV/W, OmpF, marA), oxidative stress response (trxC, yhhP, soxS) and DNA repair (dnaQ, polB, yaiV) in the chlorine resistant strain. Similarly, Wang, et al. [9] also highlighted three primary mechanisms responsible for bacterial chlorine resistance: enhancement of efflux pump activity, activation of cellular repair pathways, and increased nutrient absorption capabilities.

In addition, due to their unique structure and chemical composition, spores exhibited greater chlorine resistance compared to actively growing cells [5,16]. Some Gram-negative bacteria, notably Bacillus and Clostridium, form spores when exposed to unfavorable environments, and spores are one of the most resistant of all forms of bacteria present [3].

For internal chlorine resistance determined by genetic materials, current research mainly focuses on regulation of enzyme activities (e.g., superoxide dismutase (SOD) activity) or energy metabolisms (e.g., ATP content). For example, Rajeev, et al. [31] examined the chlorine resistance mechanisms of bacterium Halomonas boliviensis by analyzing cellular morphology, membrane damage, EPS, aggregation capability, and SOD activity. The results showed that compared to P. espejiana, H. boliviensis exhibited higher capsular EPS production, enhanced aggregation ability, and increased SOD levels, which likely contributed to its chlorine resistance. Zhou, et al. [65] studied the microbial community and function in polyethylene (PE) and ductile cast iron (DI) pipes with presence of both residual disinfectants and antibiotics. The results showed that in addition to increasing biofilm EPS secretion, disinfectant (i.e., chlorine) exposure also enhanced the bacterial activity (ATP content). The oxidative stress-related enzymes in microbial communities were also more active under the synergistic influence of antibiotics and disinfectants. Jathar, et al. [29] examined eight chlorine-resistant isolates (three of the Serratia sp. and five of the Acinetobacter) for their antioxidant profile after chlorine exposure. Due to the difference in both profiles between and within species, it was concluded that antioxidant enzymes (for example, superoxide dismutase (SOD) and guaiacol peroxidase (GPX)) could be only one of the mechanisms for bacterial self-preservation against oxidative stress from chlorination.

At present, the multitude of protective mechanisms in chlorine-resistant bacteria are still not thoroughly understood, which necessitate further in-depth studies in the future. These mechanisms have made them a significant challenge for water treatment, necessitating alternative disinfection approaches and robust system monitoring.

3.5. Control Methods of CRB

Currently, research on inactivation methods for CRB can be roughly classified into physical, chemical, biological and combined methods [16,17].

3.5.1. Physical Method

Ultraviolet (UV) light disinfection (high citation keywords “ultraviolet disinfection” in Figure 8) is a widely used method for ensuring the safety of drinking water. Its effectiveness lies in its mechanism of inactivating microorganisms. By photochemically damaging their genetic material, specifically DNA or RNA, UV light disrupts transcription and replication processes, rendering the microbes unable to reproduce or function. Thus, UV disinfection has been evaluated for its ability to inactivate CRB in literature. For example, Cho, et al. [18] isolated a biofilm-forming CRB, Phaeobacter caeruleus, and used it as a representative to test disinfection efficiency of UV. The results showed that UV significantly reduced the viable count of P. caeruleus by up to 99.8%, demonstrating that UV irradiation can effectively inactivate CRB. Jing, et al. [19] investigated the efficacy of different UV light wavelengths in inactivating various CRB and examined the resulting microbial cell structure changes. Their findings indicated exposure to UV-LED at 265 nm, UV-LED at 285 nm, and low-pressure ultraviolet (LPUV) effectively inactivated CRB primarily through DNA damage and reduction of intracellular ATP, with UV-LED at 265 nm being the most efficient wavelength. In contrast, the mechanisms underlying bacterial inactivation at 222 nm wavelength differed considerably, primarily involving substantial production of reactive oxygen species (ROS). This resulted in pronounced damage to bacterial cell membranes, direct reduction in ATP levels, and limited DNA impairments. Additionally, multispectral medium-pressure ultraviolet (MPUV) triggered all the aforementioned mechanisms, although DNA damage played the predominant role in microbial inactivation. Xu, et al. [30] explored CRB controlling methods, including chloramine, ozone, and ultraviolet (UV) disinfection, against eight bacterial strains with different chlorine resistance. Among all the approaches, UV disinfection was the most capable as measured by the average and median log inactivation rates. It also has the additional advantage that there was no observable correlation between UV resistance of the bacteria and EPS secretion. Compared to chlorine and ozone, which tend to select bacteria with high EPS secretion and then consumed by the EPS, UV irradiation is the best method for controlling CRB of this nature. While UV has shown promising results in inactivating CRB, its large-scale implementation in water supply systems is influenced by several limitations. These include its inability to provide a residual disinfectant effect, its vulnerability to water turbidity which significantly reduces disinfection efficiency, and the photoreactivation of bacteria post treatment [66].

In addition to UV, membrane filtration is another widely used physical method for removing microbes from water, with its efficiency largely determined by the membrane’s pore size. However, the presence of CRB can contribute to or exacerbate membrane biofouling in certain cases [17,25]. Thus, several studies have focused on developing novel reverse osmosis membranes [26] or nanofiltration membranes [27,28] with antibiofouling and chlorine resistant properties. For example, Wang, et al. [26] utilized layer-by-layer interfacial polymerization (LbL-IP) to graft sulfonamide monomers, 4-aminobenzene sulfonamide (4-ABSA) and 2-aminoethanesulfonamide (2-AESA), onto polyamide reverse osmosis (PA RO) membranes. This modification notably improved water flux, antifouling properties, and chlorine resistance. Abundance of work in this area can be seen in Figure 5 and Figure 7, where #2 reverse osmosis, and #3 reverse osmosis membrane were shown as the main clusters. Keywords such as “reverse osmosis membranes” also showed high citation bursts, as can be seen in Figure 8.

3.5.2. Chemical Method

Chemical disinfection used for CRB inactivation mainly includes approaches such as ozone-based method and chlorine dioxide (ClO₂) (high citation bursts keywords in Figure 8). For example, Ding, et al. [20] found that CRB were more easily inactivated by ozone disinfection than spores at low ozone concentrations. However, Bacillus cereus spores showed similar inactivation patterns across different ozone levels, with inactivation exceeding 3 log as ozone concentration and treatment time increased. Their stronger cell-surface hydrophobicity likely contributed to higher disinfection resistance. Jia, et al. [22] isolated Pseudomonas peli from a water supply network and tested ClO₂ and UV for inactivation. ClO₂ was more effective than free chlorine, disrupting the cell membrane’s integrity and permeability. P. peli was also highly sensitive to UV, with a 40 mJ/cm² dose achieving 99.99% inactivation.

In addition to these common methods, some new chemical sterilization methods are currently emerging. For example, Li, et al. [17] combined electrochemical devices (ED) (sequential oxidation/reduction (Ox/Red), Red-Ox and Ox-Red-Ox processes) with chlorination (AC) to evaluate their synergistic effects on CRB in drinking water (The design of the reactor from the study is shown in Figure 9a). The ED/AC approach achieved a 3.0 log synergistic effect and reduced energy consumption by 2.5 times for inactivating various bacteria in DI and tap water. The Ox-Red-Ox process showed the best performance, achieving over 6.7 log inactivation of Bacillus cereus at 4.0 V and 0.5 mg/L Cl₂, significantly outperforming ED alone (1.25 log) and AC disinfection (2.57 log). The Ti₄O₇-based anode facilitated free chlorine radical and H⁺ ion generation, enhancing disinfection efficiency. Lu, et al. [67] developed a synergistic disinfection method combining nanowire-assisted electroporation (EP) with chlorine (Cl₂), utilizing a locally enhanced electric field to create cell pores, allowing chlorine to penetrate and inactivate bacteria. This EP/Cl₂ approach demonstrated enhanced effectiveness against chlorine-resistant Bacillus cereus and Aeromonas media as well as chlorine-sensitive E. coli, which are commonly found in drinking water systems (The design of the reactor from the study is shown in Figure 9b). The approach resulted in more than 6-log inactivation of B. cereus, rendering viable bacteria undetectable, at 1.5 V-EP and 0.9 mg/L Cl₂, significantly outperforming individual disinfection methods, with EP alone achieving 1.11 log and Cl₂ alone achieving 1.13 log inactivation. Zhou, et al. [68] developed a core-sheath nanostructured material, Cu/Cu₂O-ZnO-Fe₃O₄, and evaluated its effectiveness in inactivating CRB and degrading trichloroacetic acid (TCAA) via photodegradation. The material demonstrated strong antibacterial activity with low cytotoxicity, attributed to its nanowire structure, ion release, and reactive oxygen species (ROS) generation. These results highlight its potential for tap water treatment applications. Deng, et al. [69] used liquid ground-electrode dielectric barrier discharge (lgDBD) for synergistic sterilization via electroporation and reactive species oxidation. At 12 kV, Bacillus subtilis (chlorine-resistant) was completely inactivated within 6 min, respectively, achieving over 7.0 log reduction. The lgDBD process maintained high disinfection efficiency across different pH levels and water types.

3.5.3. Biological Method

Until now, few studies have focused on the development of novel biological methods to combat CRB. Khan and Joshi [42] isolated a highly chlorine-resistant, biofilm-forming Klebsiella pneumoniae from the cooling water system of a nuclear power plant. Instead of being controlled, K. pneumoniae enhanced biofilm formation with increasing chlorine concentrations, highlighting the limitations of chlorination. As a remedy, chlorine-resistant bacteriophages specific to K. pneumoniae were isolated and effectively inhibited planktonic growth, biofilm formation, and removed preformed biofilms. Additionally, Epigallocatechin gallate (EGCG), a plant polyphenol, shows potential for drinking water disinfection due to its sustained antibacterial effects and health benefits. In Feng, et al. [70], inactivation rates of Bacillus subtilis by EGCG were tested under varying concentrations, contact times, pH levels, and temperatures. The study found that inactivation rates increased with higher EGCG concentrations, particularly below 800 mg/L and under acidic conditions. Also, Feng, et al. [24] explored EGCG’s effectiveness against B. subtilis (chlorine-resistant) and its spores. EGCG demonstrated the ability to continuously inhibit CRB, improving biological stability in water distribution systems. However, EGCG was found to be less effective for raw water with high spore content, making it more suitable as a complementary disinfectant when used alongside spore removal techniques like ozone, ultraviolet, or ultrafiltration. These findings highlighted EGCG’s potential as a targeted solution for enhancing water quality in specific treatment processes.

3.5.4. Combined Method

Traditional disinfection methods, relying on a single mechanism of action, have encountered major challenges, including health and ecological risks from disinfection-resistant bacteria. Therefore, there is an urgent need to develop advanced disinfection processes that incorporate multiple mechanisms. For example, Cai, et al. [16] explored a control strategy for spore-forming bacteria (SFB) using pre-oxidation, coagulation sedimentation, and UV-based advanced oxidation processes (UV-AOPs) in a lab-scale water treatment setup. A 5-min pre-oxidation with Cl₂ or ClO₂ facilitated spore transformation, enabling efficient removal through coagulation sedimentation with a 3.15-log reduction. UV inactivation of the most chlorine-resistant SFB strain was enhanced by adding 0.1 mM H₂O₂, Cl₂, or peroxymonosulfate (PMS), achieving a consistent 2-log inactivation at 40 mJ/cm², UV/H₂O₂, UV/Cl₂, and UV/PMS increased inactivation rates by 1.20, 1.36 and 1.91 times, respectively, compared to UV alone. This approach presents a promising method for managing chlorine-resistant SFB in water treatment systems (the design of the reactor from the study is shown in Figure 9c). Zeng, et al. [5] evaluated ultraviolet (UV) irradiation and two UV-AOPs (UV/H₂O₂ and UV/PMS) for Bacillus cereus inactivation (The design of the reactor from the study is shown in Figure 9d). A linear relationship was found between UV dosage and inactivation rate, with over 3-log inactivation at 180 mJ/cm². The addition of 20 mg/L H₂O₂ and PMS lowered the required UV dose to 140 and 120 mJ/cm², respectively. UV-AOPs prevented bacterial regrowth within 24 h, while flow cytometry and electron microscopy confirmed membrane and cytoplasmic damage, leading to intracellular material release (The images from flow cytometry and electron microscopy are shown in Figure 10). Among the treatments, UV/PMS was the most effective, followed by UV/H₂O₂ and UV alone, emphasizing UV-AOPs as promising methods for controlling chlorine-resistant bacteria in drinking water systems. The characteristics of different CRB control methods are shown in

Table 4. Characteristics of different CRB control methods.

CRB	Control Methods	Removal Efficiency	Mechanism	Ref
Phaeobacter caeruleus	Physical method 254 nm of UV light at a dose of 50 mJ/cm2	UV disinfection reduced viable P. caeruleus by 99.8% (3.3  ±  0.6 × 101 CFU/mL vs. 1.4  ±  0.1 × 10⁴ CFU/mL in the control).	DNA damage	[18]
Pseudomonas aeruginosa, Bacillus subtilis, Mycobacterium fortuitum, Pantoea spp., and Stenotrophomonas spp.	Physical method Low-pressure UV (LPUV), medium-pressure UV (MPUV), UV-LEDs (265 and 285 nm), and 222 nm KrCl excilamp irradiation. UV: 0, 5, 10, 20, 40, and 80 mJ/cm2.	Inactivation efficacy of the UV: UV-LED 265 nm > LPUV ≈ MPUV ≈ 222 nm > UV-LED 285 nm. Bacterial resistance to UV: P. aeruginosa < Stenotrophomonas spp. < M. Fortuitum ≈ B. subtilis < Pantoea spp.	CRB inactivation varied by UV type: UV-LED (265/285 nm) and LPUV caused DNA damage and ATP decline, 222 nm induced ROS production, membrane damage, ATP loss, and DNA lesions, while MPUV primarily targeted DNA but also triggered both mechanisms.	[19]
Bacillus cereus CR19	Physical method 40 mJ/cm2 UV Chemical method 2 mg-Cl2/L chlorine, 2 mg-Cl2/L chloramine, and 2 mg/L ozone.	Inactivation efficiency: 40 mJ/cm2 UV (1.90 log), 2 mg-Cl2/L chlorine (0.67 log), 2 mg-Cl2/L chloramine (1.68 log), and 2 mg/L ozone (0.19 log).	EPS primarily contributes to bacterial resistance against chlorine and ozone but not UV or chloramine. Carbon source metabolism is linked to multidrug resistance.	[30]
bacteria & spores Bacillus alvei, Lysinibacillus fusiformis, and Bacillus cereus	Chemical method 1.5 mg/L ozone concentration for 1 min	Bacterial inactivation exceeded 3 log, significantly higher than spores. B. alvei, L. fusiformis, and B. cereus spores showed log reductions of 2.33, 2.10, and 1.97, respectively. Over 99.9% of B. cereus spores were inactivated with 3 mg/L ozone in 1 min.	Both cell structures and gene fragments were damaged by ozone disinfection.	[20]
Pseudomonas peli 083992	Physical method UV: 40 mJ/cm2 Chemical method ClO2: 0.38 mg/L	ClO2 (0.38 mg/L, 5 min) inactivated P. peli by 3.6 log. UV (40 mJ/cm2) achieved over 4 log (99.99%) inactivation, with near-total inactivation at 60 mJ/cm2.	Free chlorine and chlorine dioxide inactivated P. peli primarily by disrupting the integrity and permeability of the cell membrane.	[22]
Bacillus cereus	Chemical method Ox-Red-Ox: 4.0 V and Cl2: 0.5 mg/L	Achieved over 6.7 log removal of B. cereus at 4.0 V and 0.5 mg/L Cl2	Electrochemical oxidation at high voltage generated HClO and free chlorine radicals, enhancing oxidative damage to bacterial cell structures.	[17]
Bacillus cereus	Chemical method EP (electroporation): 1.5 V and Cl2: 0.9 mg/L	EP/Cl2 achieved > 6 log B. cereus inactivation at 1.5 V EP and 0.9 mg/L Cl2, far exceeding EP (1.11 log) or Cl2 (1.13 log) alone.	EP/Cl2 disinfection created reversible pores for chlorine penetration, overcoming the EPS barrier. Chlorine oxidation then expanded these pores, enhancing bacterial inactivation by cell structure destruction.	[67]
P. aeruginosa, S. aureus	Chemical method Cu/Cu2O-ZnO-Fe3O4	Cu/Cu2O-ZnO-Fe3O4 eliminated 106 CFU/mL P. aeruginosa and S. aureus in 30 min at 10 mg/L, 20 min at 25.5 mg/L, and 10 min at 255 mg/L.	These materials took advantage of their nanostructure, ion release, and ROS effects to change and damage the cell wall and membrane, penetrate cells, and trigger apoptosis.	[68]
Pseudomonas fluorescens and Bacillus subtilis	Chemical method dielectric barrier discharge (lgDBD): 12 kV	A 12 kV discharge inactivated B. subtilis and P. fluorescens by over 7 log in 6 and 8 min, respectively.	During lgDBD treatment, streamer propagation creates a strong electric field, inducing bacterial electroporation. This disrupts membrane integrity, causing intracellular leakage and allowing ROS/RNS penetration. These reactive species then damage proteins and DNA, preventing bacterial repair.	[69]
B. subtilis	Biological method EGCG (800 mg/L)	High concentration of EGCG (800 mg/L) exhibited a significant inactivation effect on B. subtilis vegetative cells (1.3 log).	EGCG disrupts the morphology and energy metabolism of B. subtilis in a concentration-dependent manner. It inhibits multiple gene expressions, impairing material synthesis, energy metabolism, and the antioxidant system.	[24]
B. subtilis	Biological method EGCG (400 mg/L) EGCG (800 mg/L)	A 30-min disinfection with 400 mg/L EGCG reduced B. subtilis by 1.07 log, while 800 mg/L achieved a 1.32 log reduction.	EGCG inactivates B. subtilis by disrupting its structure, energy metabolism, and antioxidant system. It lowers SOD, CAT, and GSH levels, weakening defenses and impairing respiration and ATP synthesis.	[70]
Bacillus alvei, Bacillus cereus, and Lysinibacillus fusiformis	Physical method UV (40 mJ/cm2) Combined method coupling pre-oxidation (Cl2 or ClO2), coagulation sedimentation (20 mg/L PAC and 0.08 mg/L PAM), and UV-AOPs inactivation	5 min long Cl2 (0.9 mg/L) or ClO2 (0.5 mg/L) pre-oxidation induced apparent spore transformation (>75%). Coagulation sedimentation can efficiently remove the formed spores up to 3.15-lg. UV-AOPs substantially enhanced SFB inactivation with >2-lg at 40 mJ/cm2 dosage.	UV inactivated bacteria by damaging DNA without breaking the cell structure. UV-AOPs caused intracellular leakage, morphological damage, and structural disruption, leading to bacterial death.	[16]
Bacillus cereus spores	Physical method UV Combined method UV/H2O2 or UV/peroxymonosulfate (PMS)	The B. cereus spores showed high chlorine resistance. UV inactivation reaching over 3 log at 180 mJ/cm2. Adding 20 mg/L H2O2 or PMS reduced the required UV dose to 140 mJ/cm2 and 120 mJ/cm2, respectively.	Flow cytometry and SEM showed that UV/H2O2 and UV/PMS disrupted spore structure, damaging membranes and cytoplasm, leading to intracellular leakage.	[5]

.

4. Future Research Directions

4.1. Overlooked Source of T&O in Drinking Water Caused by CRB

With improved living standards, public expectations for drinking water quality have risen, particularly regarding aesthetic qualities like taste and odor (T&O). Consumers often associate unusual tastes or smells in tap water with poor quality or health risks, making T&O issues a major source of complaints [11,71]. T&O issues can stem from raw water contaminants or occur during distribution. Common sources include odors leached from plastic pipes [71,72,73], metallic ions from corrosion [11], reactions between residual disinfectants and organic matter [74], and biofilms formed by chlorine-resistant bacteria [11,74]. While the first three causes have been extensively studied and mitigated through measures such as material upgrades and corrosion control, T&O issues caused by biofilms remain an unaddressed challenge.

Generally, chlorine-resistant bacteria can deplete residual chlorine, promoting bacterial regrowth and biofilm development, which under certain conditions may release odor-causing compounds [11,74]. Among odor-producing microorganisms, Actinobacteria (e.g., Streptomyces, Nocardia, Actinomadura and Thermoactinomyces) are significant contributors [75]. Streptomyces, for example, produces volatile organic compounds like geosmin (GSM), 2-methylisoborneol (2-MIB), pyrazines, and sulfur-based compounds (e.g., dimethyl disulfide, dimethyl trisulfide). These bacteria often inhabit sediments and biofilms in water pipelines, where their spore-forming abilities make them more resistant to disinfectants [76]. Therefore, effectively managing biofilms and chlorine-resistant bacteria is essential to addressing persistent T&O challenges in drinking water systems and ensuring customer satisfaction.

Also, the authors have recently done a project on monitoring bacterial community structure at different residual chlorine points in an actual DWDS in southern China. The preliminary results showed that at the phylum level, Proteobacteria, Actinobacteria and Cyanobacteria are dominant ones in raw water, and proportion of Proteobacteria increased after chlorination, which demonstrated its relatively good chlorine resistance. Actinobacteria and Cyanobacteria, accounting for a relatively large proportion in raw water, experienced initial decrease after chlorination. but subsequently increased at points of DWDS with lower residual chlorine levels. indicating their possible regrowth and certain level of chlorine resistance in DWDS. As an example, Rhodococcus, a genus of Actinobacteria, increased by roughly 200% after chlorine disinfection, which is a clear sign of its chlorine resistance. Moreover, some genus of Actinobacteria and Cyanobacteria are known to generate odor-causing compounds, contributing to T&O issues in drinking water, though the effects and mechanisms remain unclear [75,77].

Based on the above, the authors believe that the relationship between chlorine-resistant bacteria and T&O substances in DWDS and their effective control methods are directions worthy of in-depth research.

4.2. Internal Intrinsic Chlorine Resistance Determined by Genetic Materials

Though various external or internal factors for chlorine resistance of CRB have been reported in the literature, limited investigation has been done on the internal intrinsic chlorine-resistant mechanisms, which are fundamentally determined by genetic materials and their expression. These include functional genes expressions related to oxidative stress, DNA repair, antioxidant enzyme secretion, porin regulation, and cell wall repair etc. [29,32]. Fortunately, the research community has begun to pay attention. For example, Miao, et al. [32] investigated microbial interactions and metabolic pathways influencing bacterial chlorine resistance in chloraminated DWSSs. Dominant CRB (Bdellovibrio, Bradyrhizobium, Peredibacter, Sphingomonas, and Hydrogenophaga) interacted to sustain basic metabolism. Among bacterial pathways, 4.21% were CRB-specific, with glutaminyl-tRNA biosynthesis being the most prominent. After chloramine disinfection, the relative abundance of glutamate-tRNA ligase (GlnRS) and related genes increased by over 10%, and a GlnRS overexpression strain exhibited higher growth and lower inactivation rates. These findings suggest that glutaminyl-tRNA biosynthesis plays a key role in enhancing bacterial chlorine resistance in DWSSs.

Additionally, determining whether intracellular molecular reactions contribute more to overall resistance than cell permeability barriers or chlorine-neutralizing substances remains challenging. Thus, more experimental and analytical work would need to be done in this field for us to have a clearer understanding at a more fundamental level.

4.3. Novel and Effective Control Method for CRB Risks

Despite many existing research articles looking at control methods for CRB, most known approaches are not deployable at scale. They are often too costly, inadequately studied or lack extensive testing in large-scale practical applications. An ideal control method would need to balance cost and effectiveness while minimizing harmful side-effects. There are no established methodologies to explore these trade-offs and a proper framework for evaluating various metrics in vitro and in situ can be very useful in driving the development of innovative, cost-effective, and practical methods for CRB control.

From a technical perspective, since EPS production and biofilm formation are key factors in bacterial chlorine resistance, controlling CRB could be achieved by adjusting environmental parameters such as total organic carbon (TOC), pH, and temperature. These changes can influence cell membrane structure and reduce EPS secretion, thereby weakening bacterial resistance. This indirect approach to reducing resistance offers a practical alternative to increasing disinfectant use, helping to minimize harmful disinfection by-products while effectively targeting chlorine-resistant bacteria. As a sustainable and efficient strategy for enhancing water treatment, further research in this area holds significant potential and deserves attention from the scientific community.

4.4. Monitoring and Operational Management of CRB in Water Networks

To apply CRB control methods in real water networks, an operational framework that combines technologies from multiple disciplines is also needed. Recent progress in bio-sensing research [78,79] can potentially enable real time monitoring of microbial growth. When deployed strategically, the sensors can generate comprehensive views of how CRB are distributed spatially and temporally in water networks. More efficient biofilm sampling and analysis [80,81] can also contribute to faster response to the deterioration of water quality. Combined with data analytics, these technologies make it possible to have a more accurate and timely understanding of problems caused by CRB in water systems. The availability of this information can lead to better formulation of management strategies. Known techniques like chlorine alteration and boosting [82] can be applied in strategic sites to suppress the proliferation of CRB. Oxidant combinations can be applied at the appropriate time to disrupt biofilms and residual bacteria. It is also possible to correlate water quality issues with infrastructural deficiencies such as leakage, stagnation, corrosion or deposit. An appropriate process can be set up to address these deficiencies in a timely fashion or even preventatively as the monitoring technologies become better. Ultimately, the integration of sensing, data collection and analysis into the regulatory as well as the operational framework can result in much more efficient decision making and handling of CRB related issues in water management.

5. Conclusions

In summary, chlorine-resistant bacteria (CRB) in water supply systems have become a pressing issue in the field of drinking water safety. Bibliometric analysis has revealed increasing attention to this topic over the past few decades. Several common CRB species have been identified, whose resistance mechanisms primarily involve biofilm formation, matrix permeability barriers, chlorine consumption, and intracellular molecular reactions. These bacteria not only pose risks as potential pathogens but also compromise chlorine disinfection, leading to failures in water treatment and related issues in the distribution network. Furthermore, through horizontal transfer, their resistance could be acquired by other harmful microbes which pose more direct threat to human health. Existing control methods, such as chemical disinfection with ClO₂ and ozone, physical approaches like membrane filtration and UV light, and combined disinfection techniques, have shown effectiveness but also face certain limitations. Future research should prioritize uncovering the molecular mechanisms behind chlorine resistance and further exploring the link between CRB and water odor problems. These efforts will aid in developing more effective and targeted strategies which can be applied at scale to ensure drinking water safety and protect public health.