An assessment of the use of native and denatured forms of okra seed proteins as coagulants in drinking water treatment

Water treatment improves the quality of water supplied to the general public by eliminating pathogens, turbidity and other contaminants in the raw water that may be harmful to human health. The availability of good drinking water at all times enhances human development and reduces the risk of contracting diseases that can emanate from polluted or contaminated water. However, many communities in developing countries lack adequate access to potable source of drinking water, with over 748 million people living in poverty still lacking access to clean water for domestic use (WHO & UNICEF 2014). In sub-Saharan Africa alone, the number of communities that are not supplied with safe water today remains higher than in the 1990s. Thus, water remains one of the greatest threats to mankind in the developing world. Water-related disease has claimed over 1.8 million lives annually, mostly those of children under five years of age (WHO 2007). Efforts by government at all levels to tackle this problem have failed in many developing countries due to lack of funds. Providing good water infrastructure and skilled personnel to improve domestic water quality will save the lives of many people in rural areas. To achieve this, coagulation and flocculation as a means of domestic water purification should be enhanced. These processes are employed essentially to improve the aggregation and settlement of particles that are later removed from sedimentation and filtration units. The process is, however, dependent upon the ability of the coagulant and flocculant aids to produce flocs with suitable characteristics. Coagulant and flocculant mixing processes play a vital role in the transformation of particles into flocs (Duan & Gregory 2003) and the subsequent bridging of the flocs into bigger macro flocs. Aluminium sulphate and ferric chloride are globally the most widely used coagulants in drinking water treatment (Duan & Gregory 2003). However, the presence of residual aluminium (Al) in the final water has posed some concern to water and wastewater operators (Driscoll & Letterman 1988), as prolonged ingestion of aluminium in water has been linked to the development of cognate decline in the human brain and Alzheimer's disease (Gauthier et al. 2000). Martyn et al. (1989) reported that the rate of Alzheimer's disease in England and Wales was higher in areas where the mean aluminium concentration exceeds 0.11 mg/l than in locations with less than 0.01 mg/l. Furthermore, some chemical polymers, e.g. acrylamide, are thought to be carcinogenic (Mallevialle et al. 1984). In addition, the application of aluminium sulphate in water treatment can consume the alkalinity and pH of the treated water that may result in reduced coagulation efficiency.

However, in many developing countries, indigenous materials of natural plant and animal origin have been used for decades in household water treatment. These materials, if properly applied, have significant potential to increase the efficiency, and reduce the overall cost, of water treatment. Natural plant seeds are biodegradable, widely available, environmentally friendly, and are non-toxic when consumed as food. Examples of natural plant species previously studied include Moringa oleifera (MO) (Jahn 1988; Ghebremichael et al. 2005), cactus (Zhang et al. 2006) and mustard seeds (Bodlund et al. 2014). Jahn (1988) used MO seed extract as a coagulant to alleviate domestic water supply problems in the developing world. Madsen et al. (1987) also used MO extract as coagulant and observed 99.5% reduction in turbidity accompanied by 80–99% reduction in coliform and faecal coliform count. Additionally, it has been reported that the coagulation potential of cactus plants is comparable to that of aluminium sulphate in removing turbidity in water with low dose (Zhang et al. 2006). Similarly, Bodlund et al. (2014) investigated the performance of different mustard seeds in pond and synthetic water and reported a coagulation activity of greater than 70% in mustard (large) compared to 60% in mustard (small), whereas mustard (yellow and cake) achieved 45 and 25%, respectively. To date, most research has focussed on the crude seed extract and the purified seeds’ proteins. Although MO has received more attention by researchers than the other plant seeds, there are potentially many vegetable plants that could perform as effectively and efficiently as MO in raw water treatment.

Okra is a plant widely grown in Nigeria and many other tropical regions of the world because of its nutrients and ease of cultivation. It can grow under different environmental conditions and reach maturity within three months of planting. Okra seed is a major source of protein, vitamin, calcium and oil, and is capable of curing ulcers and providing relief from haemorrhoids (Abidi et al. 2014). The okra pod contains carbohydrate and mucilaginous substances capable of removing turbidity in water, treating tannery and industrial wastewater (Agarwal et al. 2003). In their separate studies, De Jesus et al. (2013) and (Patale et al. 2012) applied okra powder obtained from a mature pod and achieved up to 99% reduction in turbidity within 10 minutes sedimentation time, due to the presence of mucilage substances in the pod. Conversely, the application of mallow and mucilage obtained from okra plants in water treatment revealed a major drawback due to the addition of organic substances from the plant in the final water (Anastasakis et al. 2009). Furthermore, the flocculating performance of various parts of okra plant, including the seed, was studied by Fahmi et al. (2014) on 55 NTU kaolin water and observed that only a 64.5% reduction in turbidity was achieved with okra seed extract. Okra is one of the most consumed traditional vegetables, eaten fried or boiled, steamed and may be added to salads, soups and stews. The extracted mucilage of the okra pod is used as a suspending agent and as a pharmaceutical adjuvant in paracetamol and other drug delivery (Sharma et al. 2013; Zaharuddin et al. 2014). It is also widely used in the cosmetic, pharmaceutical and food industries as a preservative.

In natural plant seed extracts, the coagulating compounds reported in the literature concern cationic protein (Ghebremichael et al. 2005). Studies have found the protein content in defatted okra seed to be as high as 40–50% (Oyelade et al. 2003), whilst in crude form, okra seed protein content was found to be in the range of 23.8%–25.5%. Protein, in a cationic state, can easily precipitate from solution with negatively charged substances. Okra seed protein contains over 100 amino acids and more than eleven major amino acids including three positively charged (arginine, lysine and histidine), and the two anions of aspartic and glutamic acids (Sami et al. 2013). The ɛ-amino group, the guanidine group and the imidazole group of the corresponding lysine, arginine and histidine residues in okra can give proton alkaline characteristics; they can bind the hydrogen ion and provide the protein molecule with a positive charge after they are fully ionised.

A protein is said to become denatured when its folding structure is altered as a result of exposure to certain elements of physical factors (e.g. heat), causing the protein to become biologically inactive. Proteins also degrade and denature upon storage, with such denaturation leading to visible aggregation and turbidity formation (Sharma & Luthra-Guptasarma 2009). In some instances proteins can be renatured but in most cases the denaturation is irreversible.

The research reported in this paper evaluates the performance of denatured Okra seed protein compared to its native state as an alternative water treatment coagulant and disinfectant in domestic water purification.

A good quality seed of okra was obtained at a local market in Hawul local government area of Borno State, Nigeria. In this market, fresh and old seeds of high quality species of okra are readily available. The seeds were sorted, packaged and labelled appropriately for ease of identification and transported to the UK for laboratory processing, preparation and analysis. The seed was cleaned by washing with tap water in order to remove contaminants such as dust, damaged seeds and plant debris which might affect the integrity of seeds during water treatment. The seeds were then dried in an oven at 60 °C for 6 hours before grinding.

Analytical grade chemicals and reagents (sodium chloride, sodium hydroxide, aluminium sulphate and hydrogen chloride) were obtained from Fisher Scientific, UK and kaolinFluka-60609, from Sigma Aldrich, Germany. Deionised (DI) water was used to prepare all the suspensions and concentration solutions in this study.

The seeds of okra were ground to fine powder using a laboratory miller (Tema mill, Germany) for 2 minutes to obtain the desired powder. The resulting seed powders were sieved through a set of stainless steel sieves (600 to 212 μm). The powders retained on the 212 and 300 μm were combined and used in the study.

The extract was prepared from the ground seed powders by adding 1.0 M NaCl solution to the seed powder to make 2% (w/v) suspension, i.e. 2 g of the seed powder in 100 ml NaCl. The suspension was vigorously stirred using a magnetic stirrer for 15 minutes at room temperature (19 ± 2 °C). In many tropical countries, room temperature ranges between 22 and 25 °C. The suspension was then centrifuged at 4,000 rpm for 10 minutes using a Heraeus Megafuge16 (Thermo Scientific, Germany). The suspension was decanted and the residual solids were dried in an oven at 50 °C overnight. The weight of the dried solid material was measured to ascertain the amount of seed powder used in making the suspension. The decanted suspension was then filtered through a Whatman No. 42 filter paper and the filtrate termed okra salt extract (OSEX). Similarly, the extract was prepared by dissolving 2 g of the seed powder in 100 ml of DI water to make 2% suspension, to extract the coagulating compound in the seed. The suspension was stirred using a magnetic stirrer for 15 minutes and then centrifuged at 4,000 rpm for 10 minutes. The suspension was decanted as in OSEX and then filtered through a Whatman No. 42 filter paper and the filtrate termed okra water extract (OWE). In addition, 2 g of MO was dissolved in 100 ml NaCl to make 2% suspension of MO extract and used in the study. Protein concentration in the extracts was estimated following Bradford (1976) with standard bovine serum albumin (BSA), absorbance was measured at 595 nm using Ultrospec 2100, UV/visible spectrophotometer.

The OSEX solutions were heated using a hot plate at temperatures of 60, 97 and 140 °C for 6, 4 and 2 hours, respectively. The wide temperature range was selected because most people in developing countries use firewood as a source of cooking energy which is difficult to control. The heated samples were then centrifuged at 4,500 rpm for 10 minutes and were filtered through a Whatman no. 42 filter paper and used in the study. Similarly, the extract was stored for 1, 3, 7, 10 and 14 days to denature the extract. Finally, the molecular weight (MW) of the extracts and the denatured protein samples were determined on 12% sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE).

The protein sample was resolved by SDS-PAGE (normal PAGE) and transferred to a nitrocellulose membrane (Protran BA-85, Pierce Protein Biology). The membrane was blocked in 5% milk/1X TBST (Tris-buffered Saline-TWEEN 20) and incubated on a rocker at room temperature for 30 minutes. After the blocking, the membranes were incubated with primary antibody polyclonal goat anti-GFP (AbD Serotec) 1:2000 of antibody in TBST overnight at 4 °C. At the end of the incubation, the membranes were washed three times with TBST at 5 minutes interval each and then incubated with the secondary antibody (polyclonal anti-Goat HRP) 1:1000 of antibody in TBST for 1 hour at room temperature on a rocker. The membranes were washed again as performed at the end of primary antibody incubation. The blots were then incubated with West Pico Chemiluminescent Substrate (Pierce) and then visualised with Gene Snap Software (SynGene).

Natural water samples were collected from the Bourn Brook River adjacent to the University of Birmingham in order to determine the bacterial inactivation capability of the heated OSEX on coliform count and E. coli present in the water. The river water was spiked with kaolin to bring the turbidity level to 45 NTU. After the jar test, the experiment was conducted using Colilert-18/Quanti-Tray (IDEX Inc., UK) for coliform/E. coli detection because of its ease of operation, flexibility, accuracy and speed. Colilert 18 was added to the water sample in a 100 ml sterilised vessel and the mixed reagent was then decanted into a tray, sealed and incubated at 35 °C for 18 hours. The presence of E. coli was determined with the help of a UV probe. The number of positive wells was counted and then read off on the most probable number (MPN) chart provided by IDEXX. The water sample was treated with two ranges of OSEX doses (50 and 80 mg/l) and the results compared with aluminium sulphate as a coagulant. The performance of OSEX in a natural water environment in terms of turbidity removal was also evaluated.

Jar tests were conducted using standard apparatus comprising six 1 litre beakers (Phipps & Bird, 7790-900B USA) to evaluate the optimum coagulant dose for the coagulation tests. For effective dispersion of the coagulant, the water was rapidly mixed at 200 rpm (G = 240 s⁻¹) for 1 minute during which various doses of the coagulant were added to the beakers. The mixing speed was then reduced to 30 rpm (G = 23 s⁻¹) for a further 30 minutes to simulate the flocculation stage. The suspension was then allowed to stand undisturbed to facilitate settlement for 1 hour. A final treated water sample (10 ml) was drawn 2 cm from the top surface of the water in the beakers using a syringe. The turbidity of the water was then measured using a turbidity meter (HI 93703, Hanna) and the water pH was measured with a pH meter (Mettler Toledo SevenGO, Switzerland). All experiments were conducted at room temperature (19 ± 2 °C).

The results of the bacterial reduction of coliforms and E. coli count using OSEX are presented in Table 1. In this study, colillert-18 Quanti-Tray method was adopted to assess the bacterial quality of the raw water before and after treatment with OSEX, and the result compared with that of aluminium sulphate. It is noteworthy that the number of large and small positive wells for coliform were the same in the raw water, before and after treatment because of the multiple presence of microbes other than E. coli. In this case, the number of coliform counts was found to be 2419.6 MPN/100 ml in both raw and treated water. However, there was a substantial reduction in total E. coli count as observed in Table 1, using 50 and 80 mg/l coagulant dosages. E. coli is a subset of total coliform, hence, the result show that E. coli count/100 ml in the raw water before treatment was 727.0/100 ml while after treatment, the total E. coli count was reduced to 79.9/100 ml and 54.5/100 ml, giving a percentage reduction in E. coli count of 89% and 93%, respectively. In addition, at optimum coagulant doses of 50 and 80 mg/l, turbidity removals were 88% and 75%, respectively, in the final water. In both cases, there was no observed decline in total coliform count after treatment, suggesting that some organic compounds in the seed were utilised as substrate by other microbes in the water to aid growth. Further investigation using aluminium sulphate as a coagulant achieved 86% reduction in E. coli count (from 727.0/100 ml to 102.2/100 ml) with a corresponding reduction of 28% in total coliform (from 2419.6 MPN/100 ml to 1553.1 MPN/100 ml). Similarly, there was substantial reduction in turbidity (approximately 98%) in final water treated with aluminium sulphate. The high removal of turbidity in alum treated water may be responsible for the reduction in total coliform count in this regard.

In the work reported here, OSEX in its native form was used as a coagulant in treating water with turbidity of 100 and 200 NTU. It was observed that an increase in OSEX dose resulted in a reduction of residual turbidity to a minimum value beyond which further addition of coagulant caused an increased turbidity as a result of re-stabilisation of the colloids (Figure 1). At this point, there were more positively charged species than the available surface charges on the colloids which encouraged stabilisation. At coagulant dose of 40–60 mg/l, maximum turbidity reductions of 91–98% were observed, as shown in Figure 1. Maximum removal efficiency was recorded in the high turbidity water (200 NTU) producing a residual turbidity of 4.9 NTU, a value which is compliant with the WHO and Nigerian drinking water standards of 5 NTU (SON 2007). This result is in agreement with finding reported by Katayon et al. (2004) who used MO extract to evaluate the performance of MO in low and high turbidity water. Turbidity removal was effective in high turbidity water because destabilisation is influenced by a high rate of colloidal interaction which encourages particle bridging (Gregory & Duan 2001).

In the work reported here, it was observed that OWE indicated poor performance in terms of turbidity removal. This means that DI water is not aggressive enough to extract the coagulating compounds while solvents such as NaCl solution were seen to be more aggressive in extracting the coagulating protein. Figure 2 shows that turbidity removal was 33% and 20% when OWE was used to treat 100 and 200 NTU water, respectively, compared with 91% and 98%, respectively, achieved with OSEX. Furthermore, the protein concentration in OSEX was 3.8 times higher than that in OWE. The effectiveness of OSEX may be due to the salting-in effect in the salt extracts causing an increase in protein solubility and dissociation as reported by Okuda et al. (1999), because the NaCl solution has a substantial effect on the solubility of a protein.

To investigate further the character of okra protein, the relationship between coagulant doses at different pH as it affects protein stability was investigated, as shown in Figure 3, using the optimal coagulant doses obtained in the previous tests. The results showed the effect of turbidity removal at lower pH to be significant (p<0.05), with a maximum of 99% efficiency removal at pH 4. This shows that low water pH had an important effect on the coagulation of turbid water with OSEX. Whilst it is not practical to treat water at such a low pH, OSEX has shown to perform effectively at pH of 6.5–7.5 as well. These were the ranges of water pH tested in the previous experiments. Each protein has an optimal pH to attain its biological function and its activity is normally affected by only a slight change in pH. Generally, an acidic environment is considered to be conducive to the binding of a proton from the dissociated carboxyl group and the transformation of protein into a cationic state, which is responsible for the charge neutralisation on particles in the water. It is known that inorganic particles are negatively charged in aquatic environments and the net surface charge of colloidal particles is reduced at low pH (Gregory 2005) which encourages the double layer compression. While study has reported that natural coagulants are most effective in coagulating water at pH in the region of pH 8 and above (Okuda et al. 2001), the study reported here observed the most effective pH to be lower, around pH 4. This demonstrates that the amino acid composition of proteins in different plants may have different coagulation activities. The difference in coagulating property may be attributed to the type of protein in okra seed although this requires further investigation.

The effect of coagulant addition on treated water pH and turbidity removal was investigated as shown in Figure 4. MO, the most studied natural plant; aluminium sulphate, a widely used synthetic coagulant in water treatment; and OSEX were each tested in water with a turbidity of 200 NTU. The results show that all the coagulants achieved approximately 98% turbidity removal efficiency with somewhat similar coagulation action. Turbidity reduction increases as coagulant dose increases, indicating the level of charge neutralisation to be similar, though the performance of OSEX was not as effective as that of MO and alum due to its high lipid content. Furthermore, the results show that incremental dosing of both OSEX and MO extracts yielded a plateau curve nature, meaning that the pH of the final water remained unaffected from its initial pH of 7.36 whereas alum was found to depress the pH of the water to pH 4.3. In aqueous solutions, amino acids contain weak α-amino groups (basic) and week α-carboxylic groups (acidic). Furthermore each of the basic and acidic amino groups contain in its side chain an ionisable group and so the combined actions of free amino acids and other amino acids in peptide linkages act as effective buffers during coagulation which resist a change in the pH of the water. Thus, natural coagulants offer an advantage over synthetic coagulants since no chemical addition is needed to control the pH of the treated water.

The efficacy and integrity of OSEX after denaturation was assessed based on different storage duration as shown in Figure 5, and temperature as indicated in Figure 6. Interestingly, it was observed that the coagulation efficiency of OSEX improved by approximately 8% with storage time from the third to tenth days, even though the increase was not appreciable and then degraded in quality thereafter. This suggests that, since there are many different sizes or bands arising from heterogeneity of one or more active proteins in seed (Ghebremichael et al. 2005), some of the proteins which were eliminated during the storage due to denaturation are protein compounds that hinder coagulation activity. However, the report presented here is not in agreement with the results reported by Katayon et al. (2004), who noted a decrease in turbidity removal efficiency of MO extracts stored longer than a day. This may be attributed to the difference in protein compounds in okra and MO seeds, because many proteins can be denatured within a few hours of storage. Here, the degradation in performance after the tenth day was caused by the aggregation, precipitation and repugnant odour emission from the protein sample. It was observed during the course of the study that, there was also the issue of physical protein agglomeration and adhesion on the container which could have added to the degraded performance after the tenth day. A wide range of characteristics can be exhibited by denatured proteins, from reduced solubility to communal aggregation.

OSEX was also heated (and so denatured) at different temperatures and its coagulation efficiency evaluated after the heat treatment as shown in Figure 6. It was observed that at the lower coagulant dose of 10 mg/l, the efficiencies of the heat-treated samples deteriorated compared to the native sample. This could be due to the disruption of both the secondary and tertiary structure of the proteins with only the primary structure available for activity which might have low coagulation potential at a lower dose. However, there were improvements in coagulation efficiencies at all doses above 10 mg/l. The highest performances were recorded at 80 mg/l for all the coagulants but the heated samples showed more than a 97% reduction in turbidity compared to 93% for untreated sample (native). It was further observed that the degree of improvement was rather varied across the denatured samples at doses of (20–60 mg/l) but still recorded approximately the same efficiency at 80 mg/l. Thus, heating can improve the coagulation potential of okra crude extract for people in developing countries as home water treatment coagulant, where access to clean water is a big challenge. At a coagulant dose of 100 mg/l, all the samples deteriorated in performance compared to 80 mg/l. It is noteworthy that the deteriorated performance of the denatured samples still outperformed the highest recorded efficiency of the non-denatured sample. This shows that the extract is stable after heat treatment. Again, it was seen that heating could improve the effectiveness of the filtration process. The time taken to filter 100 ml of the heated sample was between 30 and 40 minutes compared to more than 6 hours for non-treated sample. This demonstrates that the denaturation of proteins that are partially sensitive to storage time and temperature are beneficial, since their removal during the process further improves the quality of the coagulant protein. The two processes can therefore be considered as a simple protein purification technology which can easily be adopted in developing countries.

An assessment of the performance of the denatured sample on a very high turbidity water of 550 NTU was undertaken. Figure 7 shows the removal of turbidity was found to be more than 99% with a residual turbidity of 2.7 NTU at optimum coagulant dose of 200 mg/l. This is similar to river water turbidity found in most tropical countries of the world, especially after heavy rainfall. Therefore, a higher coagulant dose may be required to achieve the WHO water quality standard as shown in this experiment. Further work is required to assess the potential of treating natural water which may contain natural organic matter as contaminants with okra extract.

The antibacterial activity of the denatured OSEX and alum was tested on contaminated river water with low turbidity level (spiked with kaolin), using collilert-18 Quanti-Tray method. The Quanti-Tray test was conducted to assess the bacterial removal efficiency of the extract. The result revealed that at coagulant doses of 50 and 80 mg/l, the total E. coli count in the water was reduced by 89% and 93%, respectively, but the coliform counts remained unaffected after the test, as shown in Table 1. On the other hand, when alum was used as coagulant, total E. coli count in the water was reduced by 86% while there was also a 28% reduction in total coliform count. It is clear here that the reduction in E. coli in OSEX treated water was due to inactivation capability of the extract. Again the results revealed that some of the organic compounds in the seed can serve as substrates to many pathogens in the water which feed on it and hence remain unaffected during the treatment. This result shows that E. coli are more sensitive to the chemical compounds in the extract than the other microbes in the water. Madsen et al. (1987) had shown a direct relationship between E. coli reduction and removal of turbidity in treated water with MO extract at optimal coagulant dose. The okra seed contains phenol, alkaloids, flavonoids, saponins and ribosome-inactivation proteins (Kondo & Yoshikawa 2007) which show clearly that the extract has pathogen inactivation capability. Another possible reason for the reduction in E. coli count could be due to bacterial attachment to the floc during sedimentation as reported by Madsen et al. (1987). Furthermore, turbidity removal was 88% at optimum coagulant dose of 50 mg/l while at a higher dose of 80 mg/l, turbidity removal was only 75% with OSEX whereas alum achieved up to 98% reduction in turbidity in the water after treatment. It can be deduced from the study that the coagulation mechanism was the main cause of E. coli count reduction in alum treated water, while the reduction achieved with the extract was due to inactivation potential of OSEX as a coagulant since the coliform count remain unaffected after treatment.

The different protein bands in crude extract (Figure 8, lane 1) and denatured samples of okra (by heating – lanes 2 and 4, at 140 and 60 °C, respectively) and storage (lane 3) were compared in order to assess the effect of heat treatment and storage time on protein denaturation. The results show that the band in the CE with MW of 15 kDa was removed following heat treatment and was faint and less dense in the stored sample, indicating its susceptibility to heating and storage. The concentration of the band between 17 and 21 kDa was more discernible in the heat-treated samples than in the crude extract and stored samples. This indicates many overlapping protein compounds that were removed during heating and prompted the increase in coagulation efficiency. It is clear that the wider and denser band of proteins around 43 kDa in the heated samples was as a result of the effectiveness of the heat treatment. This facilitated the removal of the overlapping proteins in the band that possessed non-coagulating compounds. It is thought that these are the reasons why there was increased coagulation activity in the heat-treated sample compared with the crude extract. It is also believed that the presence of a protein–protein complex could hinder coagulation activity. Bodlund et al. (2014) also showed in a study where some mustard seed species were heated at 95 °C for 5 hours and observed the mustard seeds to be thermo-stable, which resulted in increased coagulation performances of the extracts. It was observed in this study that heating and proper storage of the extract samples before employing in water treatment can effectively remove the coagulation-hindering protein in the seed for effective performance.

Denaturing the protein of okra seed either by heating or storage destroyed both the secondary and tertiary structure of the protein, yielded an increase in MW of 21 and 43 kDa and gave rise to improved coagulation performance. Denaturing is, therefore, considered advantageous to people in developing countries where access to clean drinking water is still a major cause of health problems, though the increase was not much.

The crude extract sample of okra seed showed high coagulation activity in high turbidity water than in low turbidity water. Although not as effective or efficient as the denatured samples, it can still be considered as a good coagulant in terms of turbidity removal in home water treatment.

The bacterial inactivation capability of the extract is notable, eliminating E. coli by approximately 89% at optimum dose and 93% with a higher dosage, thought to be due to the presence of saponins and ribosome-inactivating protein in the extract. Further tests on its minimum inhibitory concentration on E. coli and other pathogenic bacteria found in water are required to reveal its further potential as a disinfectant.

Okra extract quality can be improved locally by simple heat treatment at household level without any requirement for more sophisticated heating facilities to achieve the desired improvement in treated water quality.

The first author appreciates and remains grateful to the Tertiary Education Trust Fund (TETFund AST & D/2013/14) and Nigerian government for funding this research work.

The authors wish to acknowledge the assistance in the laboratory work given by Mr Mark Carter of the School of Civil Engineering, University of Birmingham, UK.

The authors wish to declare that there are no conflicts of interest regarding the publication of this paper. The present research work was not financially supported by any person or organisations since the beginning to the end of this submitted work that may have any influence whatsoever on this paper.

The first author conducted the laboratory experiments and participated in the analysis of the results and writing up of the paper. The second author participated in reviewing the experimental procedures, results analysis and writing the paper.