Water intensive industries, such as the leather industry, are under increasing pressure to reduce water consumption and to improve the quality of the effluent they discharge. Tanneries may use up to 50m3 of water per tonne of raw material in the production of leather.

The production of leather also involves the addition of a large number of varied compounds including biocides, surfactants, syntans and dyes. As such, tanneries may produce a large quantities of effluent carrying a significant pollution loading in terms of biochemical oxygen demand (BOD) and chemical oxygen demand (COD).

Generally, tannery effluents are subject to primary treatment, such as settling, oxidation, coagulation and flocculation, and secondary treatment, in aerobic biological wastewater treatment plants, prior to discharge to the environment.

Secondary wastewater treatment plants include conventional activated sludge plants, trickling filters and rotating biological contactors. These systems will remove 60-95% and 60-85% of the effluent BOD and COD, respectively1.

Research at BLC Leather Technology Centre has demonstrated that greater reductions in both the BOD (100%) and COD (90%) loadings of tannery effluents can be achieved by application of membrane biological reactors (MBR)2. However, tannery effluents may contain a variety of compounds that are resistant to conventional biological effluent treatment.

These compounds contribute to a fraction of the effluent COD, described as the hard COD, which may, ultimately, be discharged into environment. Several specific groups of compounds have been identified as potential contributors to hard COD in tannery effluents. Polymers of naphthalene sulfonate, used as synthetic tanning agents, have been shown to be highly refractory to combined aerobic and anaerobic biological treatment3.

The toxic breakdown products of TCMTB, a commonly used fungicide in tanning, were found to be incompletely removed by MBR treatment4. Organophosphate and synthetic pyrethroid insecticides, applied to livestock to protect against ectoparasites, may persist through tannery processes and be discharged in the final effluent5.

The chemical structure of a compound will affect its resistance to biodegradation. Typically, aromatic compounds, such a naphthalene, are more resistant to biodegradation than alkyl compounds.

Increasing numbers of substitution groups, such as halogens, within the molecule will increase the resistance of the compound to degradation.

The structure of synthetic compounds may prevent degradative enzymes from reaching the active sites for degradation within the molecule6. High molecular weight compounds, such as polymers, may simply be too large to be taken into the cell of the microorganism and, therefore, are not exposed to the degradative mechanisms within the cell.

However, microbial degradation of recalcitrant compounds, that may be found in tannery effluents, has been reported. Bacteria, either singularly or in consortia, have been shown to degrade organophosphates, synthetic pyrethroids, nonylphenol ethoxylates and sulfonated detergents and dyestuffs.

The biological treatment of tannery effluents may, therefore, be enhanced by bioaugmentation of biological treatment plants.

Bioaugmentation is the inoculation of an environment with specifically enriched microorganism(s), to facilitate or increase the degradation of a target compound and provides a mechanism to reduce the acclimation period of competent indigenous microorganisms or, where competent indigenous organisms are absent, to introduce the desired catabolic activity into the contaminated environment7.

Several reports have been published describing successful studies of bioaugmentation of biotreatment plants. Bioaugmentation with the dehydroabietic acid (DhA)-degrading isolate, Zooglea resiniphilia DhA-35, restored DhA removal in high and low pH-stressed biomass of aerated lagoons treating paper mill effluent8.

Inoculation of reactors, fed peptone and phenol, with two phenol-degrading organisms shortened the start-up period for phenol removal9. Inoculation of membrane separation reactors (MBR) with 3-chlorobenzoate (3CBA)-degrading Pseudomonas putida BN210 increased the resistance of the MBR to shock-loading with 3CBA10.

Bacterial metabolic pathways are regulated by substrate induction mechanisms and global control systems (the activity of one cellular process being regulated by another).

Catabolite repression exerts a global control on the utilisation of various carbon sources and, in the presence of a preferred carbon source, the expression of genes involved in the metabolism of alternative sources, eg a recalcitrant compound, are repressed11.

As such, bioaugmentation of contaminated environments may fail if the selected isolates, found to utilise recalcitrant compounds as a sole source of carbon in the laboratory, preferentially utilise other carbon sources when inoculated into the contaminated environment, leaving the target compound undegraded. Similarly, in order for bioaugmentation to succeed, the inoculated organism(s) must continue to degrade the target compound at the concentrations at which it is expected to occur in the inoculated environment.

The bioaugmentation of biotreatment systems for the treatment of recalcitrant compounds in tannery effluent has been investigated at BLC Leather Technology Centre, using naphthalene-2-sulfonate (NSA) as a model compound (Figure 1). NSA is utilised in the synthesis of a number of tanning compounds, including syntans12 and dyestuffs13.

Concentrations of NSA of up to 61mg L-1 were detected in biologically-treated tannery effluent, indicating that it is resistant to biological treatment and may, therefore, contribute to the residual COD of the discharged effluent. An NSA-degrading isolate, Sphingomonas sp. strain RBN, was isolated from a municipal activated sludge plant treating effluent from a chemical manufacturing plant and used to evaluate bioaugmentation of biotreatment systems for the remediation of NSA-contaminated effluents.

The degradation by RBN, of 50mg NSA L-1 in a medium containing a number of readily-degradable carbon sources (CM), was assessed. The growth of RBN was assessed by measuring the turbidity of medium (OD600nm) and NSA removal was measured by HPLC.

The results are shown in Figure 2. The complete removal of NSA after 48 hours showed that degradation by RBN of NSA, at environmentally relevant concentrations, was not affected by the presence of alternative carbon sources.

In order for an isolate used for bioaugmentation to succeed, it must also be capable of survival and degradative activity in the presence of the diversity of other microorganisms in the bioaugmented environment. The use of RBN for the bioaugmentation of suspended floc reactor was investigated (Figure 3).

Supplementation of the reactor feed with 50mg L-1 each of NSA and compound #21, a phenol sulfonate-based syntan, led to an accumulation of NSA in the reactor (49.37 mg NSA L-1) and a fall in the gross COD removal rate. Six days after the addition of NSA, the mean COD removal had increased to 77.5% and the mean NSA concentration in the effluent had fallen to 32.66mg L-1, indicating that the biomass had adapted to facilitate limited degradation of NSA.

Twenty four hours after inoculation of the reactor with RBN, the effluent NSA concentration and was reduced to 15.26mg L-1, indicating that addition of RBN to the reactor had enhanced the removal of NSA.

The maximum organic removal rates, k, for the reactor in different states, were determined and are shown in Table 1.

The addition of the recalcitrant organic compounds to the reactor feed led to a fall in k, ie a fall in the maximum organic removal rate.

Augmentation of the biomass with RBN increased k, indicating that the addition of a metabolically competent organism had increased the biological treatment efficiency of the reactor biomass. This was concurrent with a reduction in the NSA concentration measured in the reactor effluent, suggesting that the increase in k was, in part, due to the degradation of NSA by RBN.

The data show that inoculation with RBN improved the treatment efficiency (k) of the reactor and that this improvement was due to degradation of NSA. Therefore, targeting a specific recalcitrant organic pollutant, NSA, by inoculation of the reactor with a NSA degrading isolate, RBN, was shown to be a successful strategy for improving the treatability of the feed.

However, inoculation of the reactor with RBN did not return k to the value calculated prior to the addition of the recalcitrant compounds to the reactor feed. The relatively reduced value of k after inoculation may have been due to the presence of compound #21 in the effluent from the reactor.

The reduced levels of COD removal (%) and reduced value of k, concurrent with the increased removal of NSA, before and after inoculation of the reactor, suggest that this compound was not degraded to any significant extent.

While it was possible to conclude empirically, from observed improvements (k), that the inoculation of the biomass with RBN had improved the biological treatment of the NSA-containing effluent, it was impossible to draw any conclusions from these data on the fate or persistence of RBN.

In order to examine its persistence when inoculated into a biomass, RBN was chromosomally marked with green fluorescent protein (gfp) by plate-mating with the donor strain Escherichia coli S17_pir. Bacteria expressing gfp will fluoresce under UV or blue light illumination (Figure 4). Gfp also provides a unique molecular marker which can be used to detect gfp-marked bacteria (Figure 5).

Gfp-marked RBN was inoculated into an activated sludge treating a feed of CM plus 50mg NSA L-1. Its survival was assessed by measurement of gfp. Plate-mating with Escherichia coli S17_pir also conferred resistance to the antibiotic kanamycin on to RBN and estimates of RBN numbers in the reactor were made by plate counts on agar containing kanamycin. Data for NSA removal, kanamycin-resistant cell counts and gfp are shown in figures 6, 7 and 8, respectively.

The reactor was operated for 28 days before inoculation with RBN. The biomass in the reactor was able to degrade 100% of the added NSA prior to inoculation with RBN. Eight days (day 36) after inoculation with RBN, the reactor was given a shock load of NSA.

As can be seen, the numbers of kanamycin-resistant cells and the amount of gfp measured in the reactor rose sharply, indicating a proliferation of RBN numbers in response to the shock load.

The NSA concentration in the reactor effluent was reduced to undetectable levels 24 hours after the shock load. Conversely, no increases in kanamycin-resistant cell numbers and gfp were measured in a reactor inoculated with a strain of RBN which could not degrade NSA, and NSA remained detectable in the reactor effluent for 48 hours after the shock load (data not shown).

The results show that RBN was able to persist after inoculation into an NSA-degrading reactor and increased the resistance of the reactor to shock loadings.

Inoculation with RBN was shown to improve both the kinetics of NSA degradation and NSA shock load resistance of activated sludges. RBN was also shown to persist in activated sludge regardless of its metabolic capacity for the degradation of the target compound.

Therefore, these results show that the bioaugmentation of biological wastewater treatment system for the degradation of NSA was successful.

It is possible then, that the degradation of other recalcitrant compounds in tannery effluents could be facilitated by bioaugmentation of biological wastewater treatment systems with specific microorganisms.

However, introduction of the desired catabolic capacity into the contaminated environment, may not always result in enhanced degradation of the target compound. Inoculated strains may fail to survive, or lose their desired catabolic activity in mixed highly competitive microbial ecosystems14,15. Application of competent isolates for bioaugmentation may fail for a number of reasons including:

* inability to adapt to the prevailing environmental conditions, including physicochemical factors, such as the presence or absence of oxygen, sub optimal pH or temperature

* predation by protozoa

* competition with the indigenous microbial population

* low substrate availability

* repression or irreversible loss of the desired catabolic function

Additionally, bioaugmentation with isolates competent for the degradation of a specific compound is unlikely to result in remediation of a wide range of recalcitrant compounds, as might be found in tannery effluent. RBN was unable to degrade compound #21, a phenol sulfonate-based syntan.

Similarly, RBN was unable to degrade a commercially available syntan based on NSA-formaldehyde condensation polymers. On this basis, it is conceivable that, in order to facilitate the degradation of a range of recalcitrant compounds in an effluent by bioaugmentation, all the compounds would have to be identified and a competent microorganism isolated for each compound.

As such, while bioaugmentation may be appropriate for the bioremediation of a given problem compound, the treatment of the hard COD in tannery effluents may require a different approach.

The microbial diversity in conventional biological treatment plants is low, with the community typically containing approximately 70 species16.

Similarly, although the concentration of biomass within MBR is significantly greater than that of conventional activated sludge plant, the population diversity (species richness) of both systems are reported as comparable and low17.

As functionality (ie capacity for biodegradation) is linked to population diversity18,19, an increase in diversity may lead to a greater capacity for these systems to treat recalcitrant xenobiotic compounds. Evidence is emerging to support this conception.

One of the simplest items of evidence is the effect of bacterial inoculum density on para-nitrophenol degradation. It was found that the probability of observing degradation of this compound was a function of the number of organisms inoculated and origin of the inocula.

Samples of river water were found to be more likely to contain degraders than samples of activated sludge20. This is consistent with recently published theoretical work that suggests that a millilitre of surface water (with 106 individuals) could contain up to 200 different populations, whilst an equivalent volume of activated sludge (with 106 individuals) could contain just 70 different species16. BLC Leather Technology Centre will shortly be launching a DTI-funded LINK project, in association with the University of Newcastle upon Tyne and a number of industrial partners including several tanneries, to investigate the influence of manipulating microbial diversity in biological treatment systems on their efficacy for the treatment of contaminated waters, including tannery effluents.

A successful conclusion to this work would provide strategies for increasing the microbial diversity in biological treatment systems with a resultant improvement in the quality of the effluent.

It is conceivable that such strategies could be used in combination with ‘conventional’ bioaugmentation, targeting specific compounds of concern, to provide leather manufacturers and those who treat their effluent, with novel effective means of reducing the impact of tannery effluents on the receiving environment by enhanced bioremediation of their organic content.