Synergistic Inhibition of Microbial Sulfide Production by Combinations of the Metabolic Inhibitor Nitrite and Biocides

MICs of six broad-spectrum biocides and two specific metabolicinhibitors and fractional inhibitory concentration indexes (FICIs)for controlling a sulfide-producing consortium were determined.Nitrite was synergistic (FICI < 1) with all but one biocidedue to its specific inhibition of dissimilatory sulfite reductase.Hence, combining nitrite with biocides allows more efficientand cost-effective control of sulfate-reducing bacteria.

INTRODUCTION

Top

Introduction

Sulfate-reducing bacteria (SRB) contribute to souring, the productionof sulfide in oil and gas fields. Sulfide is corrosive and toxic(14) and causes reservoir plugging (7). Control of biogenicsulfide production decreases operating costs and can be achievedthrough the application of biocides (18, 24), nitrate (31),or nitrite (29). Environmental regulations and development ofoil reservoirs in environmentally sensitive areas have spurredthe development of easily degradable "green" biocides that areless toxic to higher, nontarget organisms, like fish (10). Nitrite,a specific metabolic inhibitor of SRB (13, 15), is also relativelynontoxic and inexpensive and has been successfully used to inhibitsulfide production in oil field settings (6, 31).

The structures of the biocides used are shown in Fig. 1. Theirchemical natures and supposed mechanisms of action are summarizedin Table 1. The metabolic inhibitors nitrite and molybdate specificallyinhibit SRB metabolism. Nitrite inhibits dissimilatory sulfitereductase, which catalyzes the reduction of sulfite to sulfidein all SRB (13), blocking sulfate respiration. The nitrite doserequired for inhibition depends on the concentration of theavailable electron donor, on the biomass concentration, andon the presence of nitrite reductase, which prevents inhibitionof SRB by nitrite (13, 15). ATP sulfurylase activates sulfatewith ATP to adenosine phosphosulfate. Use of molybdate in thisreaction gives adenosine phosphomolybdate, which is unstableand hydrolyzes spontaneously to AMP and molybdate, depletingATP reserves (30, 32).

View larger version (19K):
[in this window]
[in a new window]

FIG. 1. Structures of the six biocides used in this study.

View this table:
[in this window]
[in a new window]

TABLE 1. Chemical classes and mechanisms of action of biocides and metabolic inhibitors used to kill or inhibit SRB

It is clear from Table 1 that biocides have much broader activitiesthan nitrite or molybdate. Use of compounds with two differentmechanisms of action may lead to synergy and increased timefor microbial resistance to develop, resulting in decreasedcosts and environmental toxicity (16). Although synergisticcombinations of biocides, such as combinations of glutaraldehydewith quaternary ammonium compounds, have been used in the oiland gas industry (1, 11, 12, 20, 26), the potential of synergybetween metabolic inhibitors and biocides has not been exploredpreviously. We report here that combining biocides with nitriteresults in considerable synergy in SRB control, allowing significantlylower biocide concentrations to be used.

Bacterial strains, media, chemicals, and analyses.

Top

Bacterial strains, media,...

A Desulfovibrio species-containing SRB consortium was obtainedby inoculating saline Postgate's medium C (22, 25) with producedwater from the Coleville oil field in Saskatchewan, Canada,and subsequently transferring the culture monthly (13). Inhibitionexperiments were done in modified CSB medium (22). Anaerobicmodified CSB medium (100 ml in an 158-ml serum bottle) was inoculatedwith 2 ml of the SRB consortium and incubated at 30°C inthe dark. Biocides and/or metabolic inhibitors were added atmid-log phase (t_ML), when approximately 5 mM sulfide had beenproduced. Incubation was continued for 1 month following thisaddition. Cultures in which sulfide production did not recoverduring this time were considered to be permanently inhibited.Glutaraldehyde, bronopol, and tetrakishydroxymethylphosphoniumsulfate (THPS) were obtained from Sigma (St. Louis, MO). Benzalkoniumchloride was obtained from ICN (Aurora, OH). The cocodiaminebiocide used, designated T-397, was generously provided by BrenntagCanada (Etobicoke, Ontario, Canada). Formaldehyde (37% solutionin 15% methanol with the balance water) was obtained from Fisher(Fair Lawn, NJ). All other chemicals were laboratory reagentgrade and were obtained from standard chemical supply companies.Sulfide, sulfate, and nitrite concentrations were determinedas described elsewhere (8, 22). Molybdate caused the mediumto become yellow-brown, which interfered with sulfide analysis.Hence, possible resumption of sulfide production was demonstratedby sulfate removal only. Some biocides caused precipitates toform in the medium. These precipitates were briefly centrifugedout of samples before analyses to prevent interference withthe spectrophotometric methods.

Determination of the MIC and the FICI.

Top

Determination of the mic...

The MIC, defined as the concentration of a compound that inhibitedsulfide production by the SRB consortium for 1 month when itwas added at mid-log phase, was determined for the six biocidesand two metabolic inhibitors. The results are summarized inTable 1. Because solid benzalkonium chloride and liquid cocodiamineconsist of multiple, related compounds having various molecularweights, their concentrations were not expressed in mM. Formixtures, concentrations were varied in the ranges 0 < [A]< MIC_A and 0 < [B] < MIC_B using a checkerboard approach,where MIC_A is the MIC of compound A and MIC_B is the MIC of compoundB. The fractional inhibitory concentration index (FICI) forcombinations of a biocide(s) and/or a metabolic inhibitor(s)that caused inhibition was calculated as follows (16): FICI= [A]/MIC_A + [B]/MIC_B.

For a synergistic mixture (FICI, <1) the MIC (MIC_AB) combinationsmay be related by the solid line in Fig. 2. The lowest FICIobtained for all inhibitory combinations on the checkerboardwas considered the FICI for the pair. For example, when MIC_Awas 1 mM, MIC_B was 5 mM, and a mixture of 0.5 mM compound Aand 0.5 mM compound B was inhibitory, then FICI was 0.5/1 +0.5/5 or 0.6. If this was the lowest value for the pair, thenall MIC_AB combinations for [A] and [B] could be representedby the curved solid line in Fig. 2, which has the equation givenabove with FICI equal to 0.6 as the tangent. Hence, althoughthe curved solid line crosses a range of FICI values (Fig. 2),only the lowest value is relevant and is reported here.

View larger version (16K):
[in this window]
[in a new window]

FIG. 2. MIC (MIC_AB) of a mixture of two biocides or metabolic inhibitors A and B as a function of concentration (solid line). The line is theoretical. Mixture compositions above the line are not inhibitory, whereas compositions below the line are inhibitory. The dotted lines represent the equation given in the text, using an MIC_A of 1 mM, an MIC_B of 5 mM, and FICI values of 0.2, 0.4, 0.6, 0.8, and 1.0, as shown.

Examples of the checkerboard approach used to determine theFICI values for each combination of antimicrobial compoundsare shown in Fig. 3. Combinations of cocodiamine and THPS wereeffective only when the MIC of both compounds were added (Fig.3D). Hence, the checkerboard filled with open triangles (representingno inhibition of sulfide production) and a maximal FICI valueof 2 was estimated, indicating that these biocides were antagonistic.In the case of nitrite and bronopol most combinations were inhibitory(Fig. 3H). The lowest FICI was estimated to be 0.28, indicatingthat this combination was synergistic. Combinations of THPSand benzalkonium chloride appeared to be indifferent, with anFICI of 1 (Fig. 3B). The estimated FICI for each pair of antimicrobialagents tested is shown in Table 2. In the case of THPS and glutaraldehydeboth antagonistic and synergistic mixtures were observed (Fig.3A). These two compounds were considered indifferent based onthe complete data set. Glutaraldehyde and benzalkonium chloride(Fig. 3E and Table 2) (FICI, 0.35) are known to be synergisticand are considered a standard for industrially available biocidecombinations. However, the following five other combinationsproved to have FICI values that were comparable to or lowerthan the value for this standard pair (Table 2): nitrite andglutaraldehyde (Fig. 3F), nitrite and benzalkonium chloride(Fig. 3G), and nitrite and bronopol (Fig. 3H), as well as benzalkoniumchloride and bronopol, and glutaraldehyde and bronopol. Of the28 combinations tested, 16 were synergistic, 2 were synergisticor indifferent, 7 were indifferent, and 3 were antagonistic.Antagonistic reactions were observed with THPS and formaldehyde,cocodiamine, or bronopol (Table 2). The number of synergisticcombinations and their average FICI were 6 and 0.44, respectively,for nitrite; 6 and 0.48, respectively, for benzalkonium chloride;5 and 0.36, respectively, for glutaraldehyde; 4 and 0.32, respectively,for bronopol; 4 and 0.54, respectively, for molybdate; 4 and0.58, respectively, for formaldehyde; and 3 and 0.52, respectively,for cocodiamine. THPS was indifferent or antagonistic in allcombinations tested. Hence, adding the metabolic inhibitor nitritetogether with a biocide was beneficial in all but one case.

View larger version (18K):
[in this window]
[in a new window]

FIG. 3. Inhibition ( ${blacklozenge}$ ) or lack of inhibition ( ${triangleup}$ ) of sulfide production by an SRB consortium with mixtures of antimicrobial compounds. (A) Glutaraldehyde plus THPS. (B) Benzalkonium chloride plus THPS. (C) Cocodiamine plus formaldehyde. (D) Cocodiamine plus THPS. (E) Glutaraldehyde plus benzalkonium chloride. (F) Nitrite plus glutaraldehyde. (G) Nitrite plus benzalkonium chloride. (H) Nitrite plus bronopol. The FICI values for these mixtures derived from the data shown, as well as the FICI values for all other mixtures tested, are listed in Table 2. The criterion for inhibition was no sulfide production for 1 month for addition at mid-log phase.

View this table:
[in this window]
[in a new window]

TABLE 2. Estimated FICI values for combinations of antimicrobial agents

Biocidal activity was evaluated by determining the most probablenumbers (MPN) (cells ml^–1) of viable cells remaining 24h after the MIC of glutaraldehyde, benzalkonium chloride, ornitrite was added at t_ML. Duplicate samples from treated andcontrol cultures were taken at t_ML and at t_ML plus 24 h. MPNs,determined by standard three-tube MPN assays (2), decreasedmost when the biocides glutaraldehyde and benzalkonium chloridewere added (no viable cells remaining) and least when nitritewas added (5% viable cells remaining) (results not shown).

Synergy of the metabolic inhibitors nitrite and molybdate ininhibiting sulfide production by oil field SRB and in inhibitingabiotic corrosion (21, 23), as well as synergy among biocidesin inhibiting SRB (17), has been demonstrated previously. Glutaraldehydeand quaternary ammonium compound blends have also been foundto be more effective in killing SRB and general aerobic bacteriathan the individual compounds (1, 11, 12, 20, 26). Althoughinsufficient combinations of concentrations were tested to clearlydemonstrate synergy, the improved killing by the combinationsof compounds suggested that synergy did occur with these blends.However, the usefulness of combining biocides (Fig. 1) and themetabolic inhibitors nitrite and molybdate, as suggested here,has not been reported previously.

The combinations of biocides and/or metabolic inhibitors usedare ranked on the basis of their FICI values in Table 2. Thesynergistic mixtures included glutaraldehyde and formaldehyde(FICI, 0.48), both of which are classified as cross-linkingaldehydes, indicating that their modes of action may not bethe same despite this similar classification. Similarly, thecationic detergents cocodiamine and benzalkonium chloride appearedto be somewhat synergistic (FICI, 0.63). THPS did not exhibitsynergy in any of the combinations tested; three of the sevencombinations tested were in fact antagonistic (Table 2), suggestingthat the structure of THPS or its mode of action was not compatiblewith the other compounds.

Bronopol, glutaraldehyde, and, to a lesser extent, benzalkoniumchloride interacted synergistically with most other compounds.Bronopol has a different structure and mechanism of action thanthe other biocides tested, which may have resulted in its excellentcapacity for synergy. Glutaraldehyde has been demonstrated tobe more effective than formaldehyde (4, 12, 20), explainingthe generally lower FICI values for mixtures including glutaraldehyde(Table 2).

Strong synergy was observed between nitrite and the biocidesglutaraldehyde, benzalkonium chloride, and bronopol (0.28 <FICI < 0.30). The very specific mode of action of nitriteas an inhibitor of dissimilatory sulfite reductase (13, 15)can apparently be favorably combined with broad-spectrum biocidalactivities to achieve synergy. Nitrite is reduced to eithernitrogen or ammonia, products that cause little environmentalharm. Although both nitrite and molybdate are specific metabolicinhibitors of the SRB sulfate reduction pathway, we found thatmolybdate was less suitable than nitrite (0.37 < FICI <1.17) (Table 2). Because it is not transformed to similarlyharmless products and is more expensive than nitrite, molybdateis inferior in metabolic inhibitor-biocide combinations. Hence,from a practical viewpoint, use of synergistic combinationsof nitrite and other biocides to inhibit souring is most suitablefor decreasing bioicide use, which decreases toxicity and resultsin potential cost savings in oil and gas recovery operationswhere sulfide control is needed.

ACKNOWLEDGMENTS

This research was funded by Natural Sciences and EngineeringResearch Council of Canada (NSERC) strategic grant STPGP 234833-00entitled "Sulfide removal with nitrate-reducing, sulfide-oxidizingbacteria" and by ConocoPhillips. E.A.G. was supported by anNSERC postdoctoral fellowship.

FOOTNOTES

* Corresponding author. Mailing address: Department of Biological Sciences, University of Calgary, 2500 University Dr. NW, Calgary, Alberta T2N 1N4, Canada. Phone: (403) 220-6388. Fax: (403) 289-9311. E-mail: voordouw{at}ucalgary.ca.

Published ahead of print on 22 September 2006.

REFERENCES

Top