Biological safety of hot water is important, and it is affected by pipeline material to a certain degree. Polypropylene random (PPR), polyvinyl chloride (PVC) and stainless steel (SS) are the common materials for pipelines in domestic hot water systems (DHWS), and biofilm growth characteristics, and biofilm microbial communities and biological diversity on the walls of pipelines are affected by the pipeline materials to a certain extent. In this paper, the effects of different materials on the growth characteristics and diversity of microbial communities were studied. The results showed that after about 60 days, the bacteria of the biofilm on the wall of pipelines completed a microbial growth cycle. Compared with PPR and SS, a greater amount of the total number of bacteria, Escherichia coli and heterotrophic plate count (HPC) attached to the PVC pipeline. Although the types of bacteria on the pipelines were similar, the proportions of species were different. Proteobacteria were the dominant bacteria at the phylum level on all the walls of the PPR, PVC and SS pipelines, and the dominant bacteria at the genus level changed before and after the exfoliation of biofilm. Some potential pathogens, such as Pseudomonas and Legionella, were detected in biofilm, so effective biofilm disinfection should be considered to ensure biological safety in DHWS.
Biofilm is quite abundant in domestic hot water systems (DHWS), indicating that there is a large quantity of bacteria. According to the statistics, about 72% of the bacteria, in the DHWS, exist on the pipeline surface, with 26% in the water phase and only 2% in the sludge of the hot water tank (Bagh et al. 2004). Biofilm is considered to be a microbial plant, where specific processes, such as pipeline corrosion, residual disinfectant decay and accumulation of inorganic materials, can take place (Douterelo et al. 2017). Although the biofilm on the pipelines is composed of bacteria from drinking water (Flemming et al. 2002), once the potential pathogens in DHWS, such as Pseudomonas aeruginosa and Legionella pneumophila, enter the hot water, the biological safety of using hot water will be affected (Moritz et al. 2010). Therefore, it is necessary to determine the growth of biofilm, bacterial species and the diversity of the microbial community, so as to effectively control the growth of biofilm on the surface of the pipelines and ensure the biological safety of DHWS.
Generally, external factors such as the flow rate of water, temperature, and value of pH in the pipelines affect the diversity of microbial species in biofilm and the interaction between bacteria (Douterelo et al. 2014a, 2014b, 2017; Lee et al. 2014). Some research has been done on the biofilm growth and the microbial community diversity of the pipeline materials. Kilb et al. (2003) found that plastic pipeline easily causes microbial attachment and growth due to its characteristics of releasing biodegradable compounds. Douterelo et al. (2014b) found that the material of the pipeline was the main factor affecting the structure of the biofilm community, and plastic pipeline samples had a higher bacterial diversity and richness than cast iron samples. Niquette et al. (2000) found that the microbial density, growing on plastic materials such as PE (Polyethylene) and polyvinyl chloride (PVC), was lower than that on steel or iron pipelines. However, there has been little research on the effects of different materials on the growth of the biofilm in DHWS pipelines.
In this paper, the effects of polypropylene random (PPR), PVC and stainless steel (SS) on the growth of biofilm on pipeline surfaces, and the biological safety of DHWS were studied by simulating the hot water system, and the growth characteristics of the biofilm and the diversity of microbial communities on the different pipeline materials are discussed.
METHODS AND MATERIALS
Municipal tap water was used as influent water, and the heated influent water simulated hot water. Dissolved organic carbon (DOC) values were measured with a total organic carbon analyzer (Vario TOC, Elementar, Germany), and turbidity values were measured with a turbidimeter (2100N, HACH, USA). The concentration of chemical oxygen demand (COD) was determined according to the Chinese drinking water standard (Water quality – Determination of permanganate index (1989)). The concentration of residual chlorine was measured using a portable residual chlorine rapid analyzer (S-CL501, Qingshijie, China) and temperature values were measured by a temperature analyzer (HANA HI 9813-65, HANA, Italy), and UV254 values were measured using a UV-spectrophotometer (Shimadzu UV2600, Shimadzu, Japan). Although UV254 and DOC are not included in Chinese drinking water standards, they are the important parameters to accurately detect the organic matter of the influent, in order to avoid the storage of tap water resulting in concentrations of organic matter that might cause interference with the biofilm growth test. The main water quality parameters during the experiment are shown in Table 1.
|DOC (mg L−1)||1.707–2.262|
|COD (mg L−1)||0.81–1.33|
|Free residual chlorine (mg L−1)||0.00–0.19|
|DOC (mg L−1)||1.707–2.262|
|COD (mg L−1)||0.81–1.33|
|Free residual chlorine (mg L−1)||0.00–0.19|
The values in this table were the main values of these water quality parameters during the experiment.
A biofilm annular reactor (BAR) is often used to simulate the growth of microorganisms in drinking water and study the effects of materials on biofilm growth (Gomes et al. 2014). Biofilm in DHWS in residential buildings was cultivated in a BAR with a continuous flow of hot water. In order to control the hydraulic conditions and temperatures, the test was operated in one BAR reactor. The BAR is cylindrical, including an outer cylinder and rotating inner cylinder, and the rotational speed stationary is controlled by the motor. There is an inlet, an outlet and a biofilm sampling port on the outer cylinder. The inner cylinder can hold 18 removable slides (17.6 cm2 per slide) for biofilm growth, including six PPR slides, six PVC slides and six SS slides. And any one of the slides can be removed at any time when the reactor is in operation. The effective volume of the reactor is 800 ml, and the shear pressure was simulated by the motor control of the inner cylinder at the speed of 30 r/min. The total flow rate of the reactor was 6.5 mL/min, making the hydraulic retention time about 2.1 h, which can minimize the planktonic growth of suspended heterotrophic organisms (Butterfield et al. 2002). The temperature of the influent water in the heat transfer aluminum tube was increased by means of a water bath, and the hot water temperature in the reactor was maintained between 37.4 °C and 40.5 °C.
Prior to experimental use, the BAR reactor was disinfected with a certain concentration of NaOCl solution of 5 mg/L. Then it was connected to a barrel container with tap water to start operation. The microbial growth test was stopped when the number of bacteria in the biofilm was significantly reduced. The duration of the whole experiment was 80 days.
Samples were taken during the operation of the reactor, including the biofilm for biomass calculation and the influent quality index. The slides were removed from the reactor and the colonies attached to the surface of the slides were rinsed with sterile water. After using sterilized cotton swabs to wipe samples from the slides, they were then put into the 10 ml tubes with sterilized ultrapure water. Cotton swabs were then placed in the ultrasonic cleaning device (KQ-500B, Kunshan Ultrasonic Instrument Co., Ltd, Jiangsu, China) for 20 min, at a temperature of 20 °C and a frequency of 40 KHz. The biofilm on the cotton swabs was suspended in sterilized ultrapure water and then homogenized at 20,000 rpm for 1 min using a tissue homogenizer (Model M37610-33, Barnstead International, Iowa, USA) (Gagnon & Slawson 1999; Lu et al. 2005). In order to determine the total amount of bacteria, the plate count method (Standard Test Method for Hygienic Standard of Drinking Water) was used to dilute the water sample to a certain proportion. Using the filtration method as the Escherichia coli assay method, the HPC assay was carried out using heterotrophic plate counts (Standard Test Method for Hygienic Standard of Drinking Water).
On the operation of the 20th, 40th, 60th and 80th days, the slides were taken out of the pipeline system.
DNA was extracted using the EZNA Soil DNA Kit (OMEGA). The 16S rRNA genes of the bacteria were amplified from genomic DNA using a polymerase chain reaction (PCR) instrument (T100™ Thermal Cyeler, BIO-RAD). All PCR reactions performed were duplicated for each sample using a pipettor (Research plus 0.5–10 μl, Eppendorf). Quantitative quantification of genomic DNA used the Qubit2.0 DNA assay kit to determine the amount of DNA to be added to the PCR reaction. The V3–V4 interval of the 16S rRNA gene was amplified by using the 341F primer (CCCTACACGACGCTCTTCCGATCTGCCTACGGGNGGCWGCAG) and the 805R primer (GACTGGAGTTCCTTGGCACCCGAGAATTCCAGACTACHVGGGTATCTAATCC). The PCR system was configured to perform PCR amplification according to the reaction conditions, and the second round of amplification was performed by introducing Illumina Bridge PCR compatible primers (Douterelo et al. 2014a; Miller et al. 2017).
Sequencing of metagenomics and statistical analysis
Sequencing was implemented on a MiSeq Sequencing instrument (Illumina). Sequences were clustered into operational taxonomic units (OTUs) by setting a 0.03 distance limit (equivalent to 97% similarity) using the MOTHUR program. The species richness estimators, Shannon index and Simpson index were generated in MOTHUR for each sample. Sequences were phylogenetically assigned to taxonomic classifications using MOTHUR via the RDP database. After phylogenetic allocation of the sequences down to the phylum and genus level, the relative abundance of a given phylogenetic group was calculated (Lu et al. 2012).
Biofilm morphology detection
The surface structure of the treated biofilm was observed using a scanning electron microscope (SEM) (FEI nova nano450, The Netherlands), and the appropriate magnification was selected to photograph.
RESULTS AND DISCUSSION
Biofilm growth characteristics
The growth of the total number of bacteria, E. coli and HPC in the biofilm is shown in Figure 1. The microbial growth cycles of the total amount of bacteria and E. coli on the walls of the SS, PVC, and PPR pipelines were about 60 days, and the HPC cycle was approximately 40 days. During the first 10 days, the total amount of bacteria on the three kinds of pipelines increased, noticeably, the total amount of bacteria on the wall of the PVC pipeline reached the maximum, of about 7.3 × 104 cfu/cm2 on the 40th day and it began to decrease quickly to 9.6 × 103 cfu/cm2 on the 60th day. Meanwhile, the total amount of bacteria on the walls of the PPR and SS pipelines increased slowly, and reached the higher number of 1.2 × 104 cfu/cm2, and 9.3 × 103 cfu/cm2 respectively on the 40th day. The wall of the SS pipeline could be attached to less biomass in DHWS, which is more consistent with the conclusion of Jang et al. (2011), specifying that SS pipelines attached a lower amount of bacteria in the drinking water distribution system (DWDS).
As for E. coli, the amount of E. coli on the SS pipeline surface increased rapidly after 10 days, reaching 1.3 × 103 cfu/cm2, then decreased and stabilized at around 7.0 × 102 cfu/cm2. It continuously increased up until 40 days, at 1.7 × 103 cfu/cm2, and then began to decrease on the wall of the PVC pipeline. In contrast, the amount of E. coli on the PPR surface continuously grew at a lower level until about 40 days reaching 5.6 × 102 cfu/cm2 and then decreased.
As for HPC, the amounts of HPC on the walls of the three types of pipeline materials were slowly increasing during the first 40 days. They began rapidly increasing on the 40th day, and reached the maximum amount on the 60th day. The amounts of HPC on the walls of the PPR, PVC and SS pipelines were 5.4 × 105 cfu/cm2, 7.6 × 105 cfu/cm2 and 2.5 × 105 cfu/cm2 respectively, and the amount on the PVC pipeline was always higher. Pedersen (1990) found that 4 months was sufficient to allow the amount of bacteria in biofilm on SS and PVC to reach the so-called ‘steady state’ when the free chlorine content was 0.1 mg/L. Nevertheless, residual chlorine decays faster due to the high temperature of the water in the DHWS, and it is often lower than 0.1 mg/L. Therefore, it was discovered that the biomass on PPR, PVC and SS pipelines can be quickly stabilized.
Biofilm community species
At the phylum level, Proteobacteria and Firmicutes are two types of bacteria in larger amounts before exfoliation of biofilm (shown in Figure 2), and the proportion of Proteobacteria on the walls of the PPR, PVC and SS pipelines is 63.26%, 66.34% and 56.51% respectively. Studies showed that Proteobacteria was the main phylum in the surface biofilm both on the pipelines of DWDS (Hong et al. 2010; Liu et al. 2014) and on the pipelines of DHWS. A long period of delivering hot water through the pipelines causes the bacteria of Proteobacteria to accumulate on the walls of pipelines, resulting in a larger amount of bacterial species in the biofilm. After the exfoliation of biofilm, the amount of Verrucomicrobia increased significantly. Verrucomicrobia is a bacterium species which is widely found in aquatic and soil environments (Bergmann et al. 2011). Deinococcus–Thermus is also a bacterial species that appeared in large amounts on the wall of all three kinds of pipeline samples, and it includes some thermophilic bacteria, which are highly resistant to environmental hazards (Griffiths & Gupta 2007). The temperature of hot water in the pipeline system is high, resulting in the multiple reproduction of such bacteria, and the proportions of such bacteria on the PPR, PVC and SS were 5.88%, 5.21% and 5.00%, respectively. However, it is seldom found on the pipeline surface of DWDS. Zacheus et al. (2000) found that the biofilm composition of PVC, PE and SS was very similar under the same conditions. Also Percival et al. (1998) found that the biofilm on the SS surface in DWDS was mainly Pseudomonas, Acinetobacter and Micrococcus, but the DWDS is a constantly changing environment. This occurs especially in the processes of the phytoplankton community, the nutrients and the flow of water, which can affect the dominant bacteria in the bacterial community of the surface biofilm.
The proportion of Proteobacteria was 61.46% in the exfoliated biofilm, which is similar to the proportion of the biofilm on the pipelines, so Proteobacteria could be considered easy to attach and exfoliate.
At the genus level, the changes of the microbial community structure on the walls of the PPR, PVC and SS pipelines were analyzed. Among them, bacterial species such as Sphingopyxis, Spartobacteria genera incertae sedis, Gemmata, Sphingobium, Novosphingobium, and Terrimicrobium grew well on the wall of all pipelines (shown in Figure 3), most of which are common bacteria that live in fresh water and DWDS (Simoes et al. 2008). The amounts of such bacteria were drastically reduced when the biofilm was exfoliated. In contrast, Meiothermus, Rhodobacter and Hyphomicrobium had lower proportions on the walls of the pipelines, and the amounts of them were more stable in spite of decreasing with the exfoliation of biofilm.
The dominant bacteria on the walls of the PPR, PVC and SS pipelines, before and after the exfoliation of biofilm, were slightly different. Before the biofilm was exfoliated, Novosphingobium was a common dominant species, and the proportion was 29.85%, 23.95% and 28.30% respectively. For the PPR pipeline, Pseudomonas and Sphingobium were also the dominant species; for the PVC pipeline, Sphingobium and Terrimicrobium were the dominant species, and for the SS pipeline, Terrimicrobium and Sphingopyxis were the dominant species. When the biofilm was exfoliated, the dominant species on the wall of the pipelines changed; Terrimicrobium and Spartobacteria genera incertae sedis were the common dominant species for the PPR, PVC and SS, at 32.79% and 15.97% on the PPR pipeline, 28.30% and 18.31% on the PVC pipeline, and 37.26% and 13.40% on the SS pipeline respectively. But for the wall of the PPR pipeline, Sphingopyxis was also the dominant bacteria, and for the wall of the PVC pipeline, Gemmata was also the dominant bacteria, and for the wall of the SS pipeline, Sphingobium was also the dominant bacteria.
Novosphingobium, Sphingopyxis and Sphingobium all belong to the phylum of Proteobacteria, and they can produce abundant extracellular polysaccharides, and they are also less restricted in living conditions and have a stronger resistance to residual chlorine, which allows them to grow in oligotrophic environments and become the dominant bacteria of the biofilm (White et al. 1996; Furuhata et al. 2007). Meiothermus, a kind of thermophilus, belongs to the phylum of Deinococcus–Thermus. It is common in hot water and exists in biofilm in large quantities, and is able to reduce nitrate to nitrite. Hyphomicrobium belongs to the phylum of Proteobacteria, a Gram-negative bacterium, which can grow or use nitric acid for anaerobic growth, with the ability of nitrogen fixation and denitrification (Wang et al. 2014). In the shed bacteria, Sphingobium was of large amount, but it had less proportion in the bacteria of exfoliated biofilm, indicating that more species of biofilm were exfoliated.
Pseudomonas is a potential pathogen that produces extracellular polysaccharides to provide biofilm with a ‘stabilizing effect’ (Ghafoor et al. 2011). There was a higher proportion of Pseudomonas on the wall of the PPR pipeline, with the proportion of 51.29%, but it dropped to 0.19% after the exfoliation of the biofilm (shown in Figure 4). Meanwhile, the proportion of Pseudomonas on the walls of the PVC and SS pipelines was much lower, only with the proportions of 4.19% and 7.48% respectively, and it then dropped to 0.08% and 0.28% when the biofilm was exfoliated, which shows that Pseudomonas could attach to the pipeline in a large amount, and dropped sharply into the hot water with the exfoliation of the biofilm. It has a harmful impact on human health (Moritz et al. 2010).
Legionella, which is susceptible to propagation at any water temperature between 20 °C and 45 °C, was detected on the walls of the three pipeline materials, and it was more infectious at 37 °C (Amara et al. 2017). However, the amount of Legionella on the walls of the PPR, PVC and SS pipelines was slightly different (shown in Figure 4), for the SS pipeline has more Legionella than that on the PPR and PVC pipelines. Simultaneously, SS has more Legionella when exfoliated. Although the proportion of Legionella decreased when the biofilm was exfoliated, it still increased in the next microbial growth cycle. Therefore, a series of measures should be taken to reduce biofilm growth potential in hot water and on pipeline surfaces to prevent the rapid growth of bacteria (van der Kooij et al. 2005).
Richness and diversity of bacterial phylotypes
The growth process of biofilm on the walls of the PPR, PVC and SS pipelines on the 20th, 40th, 60th and 80th days was sequenced by high-throughout pyro-sequencing, and the microbial community diversity was analyzed.
With OTUs, at a 3% distance, the Shannon and Simpson indices are shown in Figure 5. The Shannon index is usually more sensitive to species richness, while the Simpson is more sensitive to species evenness (Nagendra 2002). The higher the Shannon value, the higher the diversity of the community. The larger the Simpson index, the lower the diversity of the community. Figure 5 shows that the diversity of the SS, PPR, SS and PVC were higher on the 20th, 40th, 60th and 80th days, respectively, and the Shannon values were 4.37, 4.70, 4.78 and 4.43. It also can be seen that the biofilm diversity was constantly changing – SS had more species richness in the biofilm growth stage, while the PPR and PVC had more species richness when the biofilm was exfoliated. So the material of the pipeline had a certain impact on the species in the exfoliation of the biofilm.
Surface structure of biofilm analysis
SEM can show a better observation of the apparent structure of the biofilm because there is no need to scrape the biofilm off. With the accumulation of microorganisms on the surface of the pipeline, the biofilm that occurs is usually thin and patchy (Wingender & Flemming 2011). SEM analysis showed that there was a dense block structure and clearly visible rod structure all on the wall of the PPR, PVC and SS pipelines, but the biofilm on the PVC pipeline was more solid (Figure 6), which might be related to the growth of more biomass on the pipeline. Percival et al. (1998) found that the morphology of the biofilm observed by SEM was related to the conditions during biofilm growth, with the changing of major microbial species, assimilable organic carbon, the value of pH and the temperature of the biofilm.
In this research, the growth of biofilm and the diversity of the microbial community on the walls of PPR, PVC and SS pipelines were studied. The results showed that the growth of bacteria on the surface of pipelines was stable after around 60 days. Compared with the walls of the PPR and SS pipelines, the PVC pipeline had a larger biomass of the total number of bacteria, E. coli and HPC, which indicates that PVC more easily causes adhesion of bacteria in hot water than other materials.
Different microbial species in the DHWS can selectively settle on the surface of the walls of PPR, PVC and SS pipelines. Although the types of bacteria on the pipelines were similar, the proportions of species were slightly different. Proteobacteria was the dominant bacteria at the phylum level on all the walls of the PPR, PVC and SS pipelines, which was consistent with the biofilm of pipelines in DWDS. However, its proportion in DHWS was lower than that in DWDS, which might be due to the fact that the thermophilic, such as Deinococcus–Thermus, increased. At the genus level, Novosphingobium was the common dominant species before the exfoliation of biofilm; but the dominant bacteria changed when the biofilm was exfoliated.
Some potential pathogens, such as Pseudomonas and Legionella, were detected in biofilm. It is suggested that effective biofilm disinfection should be considered to ensure microbiological safety in DHWS.
The material of the pipeline had a certain effect on the diversity of bacteria in the biofilm; SS has more species richness in the biofilm growth stage, while after the exfoliation of biofilm, PVC and PPR have more species richness.
Through the SEM photos, it was found that the apparent structure on the wall of the PVC pipelines was denser than on the PPR and SS, for much of the biomass attached on the PVC pipeline.
This experiment was financially supported by National Science and Technology Major Project of the Ministry of Science and Technology of China (2014ZX07406002).