APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 1992, p. 961-9680099-2240/92/030961-08$02.00/0Copyright ©) 1992, American Society for Microbiology

Role of Sulfate Concentration in Dechlorination of 3,4,5-Trichlorocatechol by Stable Enrichment Cultures Grown with

Coumarin and Flavanone Glycones and AglyconesANN-SOFIE ALLARD, PER-AKE HYNNING, MIKAEL REMBERGER, AND ALASDAIR H. NEILSON*

Swedish Environmental Research Institute, Box 21060, S-100 31 Stockholm, Sweden

Received 9 September 1991/Accepted 8 January 1992

Metabolically stable anaerobic enrichment cultures have been obtained from sediment samples contaminatedwith chlorophenolic compounds. Enrichment was carried out with esculin, esculetin, naringin, naringenin,fraxin, quercetin, and acetate in media with two sulfate concentrations. These cultures were used to examinethe 0-demethylation of 4,5,6-trichloroguaiacol and the dechlorination of 3,4,5-trichlorocatechol. Whereas0-demethylation was observed in all cultures, the occurrence of dechlorination was significantly morerestricted. The presence of the carbohydrate moiety in the cultures enriched with the glycones represseddevelopment of populations which were able to carry out dechlorination. Although sulfate at a concentrationof 2 g/liter in the primary enrichments blocked the development of populations able to bring aboutdechlorination, addition of sulfate at this concentration did not inhibit dechlorination in cultures possessing thiscapability. Different dichlorocatechol isomers were produced under the various conditions, so that in view ofthe established resistance of some of these to further dechlorination, the ultimate fate of 3,4,5-trichlorocatecholin the natural environment remains partly unresolved. No enrichment culture containing a low sulfateconcentration was able to dechlorinate either 2,4,5-trichlorophenol or 2,4,6-trichlorobenzoate.

The fate and persistence of organic chemicals dischargedinto the aquatic environment are problems of increasingconcern. The factors determining these are complex, so it ishighly desirable to make optimal use of the results oflaboratory experiments: while retaining the advantages ofcontrollable and reproducible systems, these may be de-signed to stimulate as closely as possible cardinal features ofnatural ecosystems. We have attempted to apply this prin-ciple consistently in our investigations (17).A number of chlorinated guaiacols and chlorocatechols

are formed during the production of pulp by the kraft processand are among the organochlorine compounds that havebeen identified in bleachery effluents (see references inreference 20). We showed that these compounds are parti-tioned from the aquatic phase into the sediment phase (23)and that they can be recovered from contaminated sedimentsin substantial quantities (24). Their persistence in the envi-ronment is therefore largely determined by their susceptibil-ity to anaerobic degradation. We showed that under anaer-obic conditions both 0-demethylation and dechlorinationreactions take place (1, 22), and one of the importantconclusions from a previous study (1) was that cultures ableto carry out these reactions need not have been obtained byselective enrichment with substrates containing aromaticmethoxy or chlorine substituents.Our attention has therefore been directed to a more

extensive investigation of the role of compounds structurallyrelated to those plausibly found as organic constituents innatural sediments (1, 22). At least in coastal areas, thesecompounds probably originate largely from terrestrialplants, and we have previously used as growth substratescompounds related to tannins, such as 3,4,5-trihydroxyben-zoate. We have now examined a series of plant-relatedbicyclic compounds: coumarins derived from 1,3-dihydroxy-

* Corresponding author.

benzenes and flavanones from 1,3,5-trihydroxybenzenes. Inan attempt to evaluate the competitive role of readily de-graded carbohydrates, we have examined growth substratesin which a carbohydrate and an aromatic structure arecombined in the same molecule. We have compared themetabolic activities of cultures grown with the glycones-inwhich the carbohydrate moieties are covalently linked to thearomatic portion by readily cleaved glycosidic bonds-withthe corresponding aglycones lacking carbohydrates. In ad-dition, since we are increasingly interested in both marineand brackish-water environments, the role of sulfate concen-tration has been systematically examined.The results showed that some coumarins and flavanone-

related compounds supported anaerobic growth and that theglycones were readily hydrolyzed to the aglycones, whichwere then further degraded. The enrichment cultures wereexamined for their ability to 0-demethylate 4,5,6-trichlo-roguaiacol and dechlorinate the resulting 3,4,5-trichlorocate-chol. 0-demethylation was observed in all of the culturesexamined, whereas the capacity for dechlorination was morerestricted and the isomers formed were dependent on thesubstrate used for primary enrichment. The glycone cultureswere significantly less active in dechlorinating 3,4,5-trichlo-rocatechol, and a high sulfate concentration was inhibitoryto development of populations capable of carrying outdechlorination. Enrichments with acetate in media with lowsulfate concentrations displayed only low dechlorinatingcapacity towards 3,4,5-trichlorocatechol, and none of theenrichments dechlorinated either 2,4,5-trichlorophenol or

2,4,6-trichlorobenzoate.

MATERUILS AND METHODS

Chemicals and analytical methods. To facilitate reference,the structural formulae of the bicyclic compounds used are

given in Fig. 1. Esculin, esculetin, fraxin, naringin, naringe-nin, and quercetin were purchased from Sigma Chemical Co.

961

Vol. 58, No. 3

APPL. ENVIRON. MICROBIOL.

RO MeO

HO OO HO 0OOR

A B

OH O OH O

ROA

HO A O

OH OH

C D OH

FIG. 1. Structural formulae of compounds used as substrates forenrichments. (A) Esculetin (R = H) and esculin (R = 1-glucosyl);(B) fraxetin (R = H) and fraxin (R = ,-glucosyl); (C) naringenin (R= H) and naringin (R = 3-rhamnoglycosyl); (D) quercetin.

(St. Louis, Mo.). Fraxetin was prepared by hydrolysis offraxin with 2 mol of HCI liter-1 for 12 h at 50°C and was usedwithout further purification. 4-Hydroxyphenylpropionic acidwas obtained from Janssen Chimica (Beerse, Belgium);3,4-dihydroxybenzaldehyde was from Riedel-de-Haen (Han-nover, Germany); and 3,4-dihydroxybenzoic acid, 2,4,6-trihydroxybenzoic acid, and phloroglucinol were from Ald-rich-Chemie (Steinheim, Germany). 4,5,6-Trichloroguaiacoland 3,4,5-trichlorocatechol together with the dichlorocate-chol isomers were synthesized as already described (21),2,4,5-trichlorophenol was purchased from Merck (Darm-stadt, Germany), and 2,4,6-trichlorobenzoic acid was fromAldrich-Chemie. Methanol and acetonitrile were high-per-formance liquid chromatography (HPLC) quality from Lab-Scan (Stillorgan, Ireland), and other solvents were distilled-in-glass quality from Burdick and Jackson (Muskegon,Mich.).HPLC for analysis of the growth substrates was carried

out with a Nucleosil C8 column (5-,um particle size) (JonesChromatography, Hengoed, United Kingdom). For the anal-ysis of fraxin, esculin, and their aglycones, the mobile phasewas 0.05 M phosphate buffer, pH 3, and methanol (75:25).The UV detector was set at 312 nm. For naringenin and itsaglycone, a mobile phase of phosphate buffer and acetoni-trile (70:30) was used with the UV detector set at 284 nm.The flow rate for all analyses was 1 ml min-'.Gas chromatographic (GC) analysis (11) of 2,4,5-trichlo-

rophenol and of chlorocatechols and GC-mass spectrometricidentification (11) of metabolites from the chlorocatecholswere carried out on samples derivatized as described previ-ously (1). Analysis of 2,4,6-trichlorobenzoic acid was carriedout with 1-ml samples by adding 4-bromobenzoic acid (20 RI,10 pLg ml-1) as a surrogate standard, extracting the acidifiedsamples with t-butyl methyl ether (two times, 1.5 ml), andesterifying the samples with diazomethane. A standard GCtemperature program (11) was used.

Metabolites produced from the growth substrates wereextracted, derivatized, and identified as follows. Samples (2ml) were acidified with a few drops of H2SO4 (1 mol liter-1),saturated with NaCl, extracted twice with ethyl acetate (1.5ml), and dried (Na2SO4). Phenols were acetylated withacetic anhydride-pyridine, excess reagents were removed(11), and carboxylic acids were methylated with diazometh-ane. Excess diazomethane and solvent were removed, andthe residue was dissolved in hexane-t-butyl ether (1:1)before GC-mass spectrometric analysis as described above.Authentic standards were treated identically.

Microbiological procedures. Sediment samples were ob-tained from two localities in the Gulf of Bothnia putativelycontaminated with organochlorine compounds from the pro-duction of bleached pulp. They were preserved in tightlycapped jars at 4°C immediately after collection, and enrich-ment experiments were begun within a month after collec-tion. All subsequent anaerobic operations were carried outin an Anaerobe Systems (San Jose, Calif.) growth chamber.Enrichments were carried out by using the basal media andthe procedures described previously (19, 22) except that thesulfate concentrations were adjusted to 0.2 or 2.0 g/liter;after initial incubation for 3 months, cultures were trans-ferred monthly at least 10 times to achieve metabolicallystable cultures. Except for esculetin, which was dissolved intetrahydrofuran-methanol (1:1), poorly water soluble sub-strates were dissolved in tetrahydrofuran, suitable volumeswere added to the bottles in which growth was carried out,and solvent was evaporated; the nominal concentrations ofthe growth substrates in the media were ca. 500 mg liter-'.

All metabolic experiments used test substrates at a con-centration of 100 ,ug liter-' and were carried out in sealedampoules incubated in darkness at 22°C as described previ-ously (1). Two sets of controls were used: (i) incubations ofthe test substrate in base medium without addition of cellsand (ii) incubation of autoclaved cells under anaerobicconditions in the presence of sulfide. The latter were pre-pared in the anaerobe chamber by transferring washed densecell suspensions to bottles fitted with rubber liners andaluminum screw caps. After removal from the chamber, thebottles were autoclaved, the contents were cooled in astream of nitrogen, the bottles were returned to the chamber,additional sulfide was added at the same concentration asused originally, the test substrate was added, and portionswere dispensed into ampoules. Experiments on sulfate ad-ditions were carried out by transferring cells grown inlow-sulfate medium to fresh medium containing a low sulfateor high sulfate concentration; the converse was carried outfor cells grown in high-sulfate medium. The test substratewas then added, and the media were dispensed into am-poules and incubated.

RESULTS

Metabolically stable cultures were obtained by enrichmentof the two sediment samples with esculin, esculetin, narin-gin, naringenin, fraxin, and acetate as carbon substrates.The basal media contained sulfate at concentrations of either0.2 or 2.0 g/liter, although parallel cultures at the twoconcentrations were not obtained from all of the growthsubstrates: cultures degrading fraxetin were obtained byinoculation with cells from the fraxin enrichment cultures.HPLC analysis of growth media showed that the glycones

were rapidly hydrolyzed to the aglycones within 24 h; theaglycones were then more slowly metabolized. In culturescontaining a low sulfate concentration, esculetin was com-pletely degraded in 40 days whereas ca. 60% of the initialconcentration of naringenin remained after 70 days; incultures containing a high sulfate concentration, degradationof esculetin occurred to the extent of ca. 50% in 40 days anddegradation of naringenin occurred to ca. 40% in 70 days.Complete degradation of both fraxin and fraxetin in mediacontaining either low or high concentrations of sulfate oc-curred in 15 days; degradation of quercetin at either sulfateconcentration occurred in 7 days.

Metabolites clearly derived from the growth substrateswere identified by comparison of the mass spectra and

962 ALLARD ET AL.

SULFATE AND DECHLORINATION OF 3,4,5-TRICHLOROCATECHOL 963

relative GC retention times of derivatized samples withthose of authentic compounds. 3,4-Dihydroxyphenylpropi-onic acid was formed from esculetin, and 4-hydroxyphenyl-propionic acid was formed from naringenin. A series of keymetabolites was identified from the degradation of quercetin:mass spectra of derivatives and those of authentic referencecompounds are shown in Fig. 2. From these structures, thepathway for the degradation of the heterocyclic ring could bededuced (Fig. 3). All of the metabolites were slowly de-graded further without evidence of identifiable intermedi-ates.

Concentrations of 4,5,6-trichloroguaiacol, 2,4,5-trichlo-rophenol, and 2,4,6-trichlorobenzoate in controls lackingcells were unaltered during the incubation periods. Theconcentration of 3,4,5-trichlorocatechol was constant duringincubation in base medium in the absence of cells or in basemedium containing autoclaved cells.

Rates of 0-demethylation were highly variable and werenot dependent on the sulfate concentration. Complete 0-de-methylation of 4,5,6-trichloroguaiacol occurred within 1 hfor cultures grown with esculin and fraxin; for fraxetin andquercetin, 0-demethylation occurred within 24 h; and foresculetin, naringin, and naringenin it occurred between 3 and7 days. Cultures enriched with acetate at a low sulfateconcentration 0-demethylated 4,5,6-trichloroguaiacol within2 days, whereas 0-demethylation by the high-sulfate enrich-ment culture was significantly slower (>30 days).

Cultures obtained by enrichment with esculin at either ofthe sulfate concentrations were unable to dechlorinate 3,4,5-trichlorocatechol after incubation for 80 days. Whereasthose obtained with esculetin at the high sulfate concentra-tion were also negative after 40 days, the correspondingculture in which sulfate was employed at the low concentra-tion dechlorinated 3,4,5-trichlorocatechol quantitativelywithin 30 days to 3,5-dichlorocatechol: the mass spectrumand GC retention time of the 0-acetate were identical tothose of an authentic specimen. When cells from the low-sulfate enrichment culture were incubated with the testsubstrate in medium containing the high sulfate concentra-tion, the rate of dechlorination was unaffected. Cells fromthe high-sulfate-concentration enrichment incubated in me-dium with the low sulfate concentration were unable to carryout dechlorination.The results from a similar series of experiments carried

out with naringin and naringenin were comparable, thoughthey differed in some important details. Cultures obtainedfrom naringin and low sulfate concentrations were unable todechlorinate 3,4,5-trichlorocatechol after incubation for 40days, whereas those using naringenin and a low sulfateconcentration dechlorinated the test substrate in a yield of40%. In these experiments 4,5-dichlorocatechol was exclu-sively produced and was identified by comparison of themass spectrum and GC retention time of the 0-acetate withthose of an authentic specimen. The corresponding culturegrown with naringenin in which a high sulfate concentrationwas employed showed no dechlorination capability afterincubation with 3,4,5-trichlorocatechol for 50 days. Whencultures from the low-sulfate enrichment were incubated inthe presence of the high sulfate concentration, dechlorina-tion of 3,4,5-trichlorocatechol occurred.Enrichment cultures with fraxin and sulfate at both low

and high concentrations exhibited dechlorination after alengthy lag, though the yields were different in the twocultures. Yields were higher in the high-sulfate cultures, and3,4,5-trichlorocatechol produced 4,5-dichlorocatechol withca. 20% 3,4-dichlorocatechol. No positive enrichment was

obtained with fraxetin, so subenrichments were made fromthe fraxin culture using fraxetin as the carbon source. Thesecultures employing sulfate at both low and high concentra-tions dechlorinated 3,4,5-trichlorocatechol to 3,5-dichloro-catechol in yields of ca. 30% without any appreciable lag:smaller amounts of 3,4-dichlorocatechol were also formed.Identification of the dichlorocatechol isomers-3,4-, 3,5- and4,5-dichlorocatechol-was carried out by comparison of therelative GC retention times of the 0-acetates, the 0-trime-thylsilyl ethers, and the 0-heptafluorobutyrates with thoseof authentic samples.The quercetin cultures enriched with both low and high

concentrations of sulfate dechlorinated 3,4,5-trichlorocate-chol to 4,5-dichlorocatechol, which was identified by com-parison of the mass spectrum and GC retention time of the0-acetate with those of an authentic sample.Whereas cultures enriched with acetate in medium with

the low sulfate concentration dechlorinated the 3,4,5-trichlo-rocatechol formed by 0-demethylation of 4,5,6-trichlorogua-iacol to 4,5-dichlorocatechol in low yield, the high-sulfateacetate enrichments were unable to carry out significantdechlorination of 3,4,5-trichlorocatechol within 60 days.To facilitate comparison, the results of all experiments on

dechlorination have been summarized in Table 1.The cultures enriched with naringenin and esculetin at low

sulfate concentrations were unable to dechlorinate either2,4,5-trichlorophenol or 2,4,6-trichlorobenzoate after 60days.

DISCUSSION

We attempted to incorporate a degree of environmentalrealism into our experiments, which can be summarized asfollows. (i) For primary enrichment of sediment samples weused compounds plausibly related to organic components ofnatural sediments; in addition, the growth substrates werestructurally unrelated to the compounds under investigationand, for example, lacked chlorine substituents. (ii) Succes-sive transfers of the initial enrichment were carried out sothat no organic matter from the original sediment remainedin the cultures used for metabolic studies. These experi-ments therefore differed from others which have been re-ported (9, 16, 27) and had the advantage that these metabol-ically stable cultures could be used in subsequentinvestigations. (iii) Xenobiotic concentrations were at thesubmicromolar level comparable to those encountered inreceiving waters. (iv) The low concentrations of the testsubstrates probably did not contribute significantly togrowth of the organisms, and the metabolism of thesexenobiotics was carried out concurrently with that of thegrowth substrates in the natural sediment phase.The control experiments clearly showed that the observed

transformations were microbiologically mediated, and ourexperiments using autoclaved cells showed that intact cellswere necessary. On the other hand, these experiments donot exclude the possible role of corrins (26), coenzyme F430(4), or porphyrin-containing cytochromes in mediating de-chlorination within the cells. These compounds-which areconstituents of many anaerobic bacteria-have been shownto bring about chemical dehalogenation of a variety ofhalogenated compounds (reference 5 and references in ref-erence 18) under reducing conditions. This is not an entirelytheoretical issue, since the question may be posed whether,even if the reactions were chemical-though not enzymat-ic-but could be accomplished only in the presence of cell

VOL. 58, 1992

APPL. ENVIRON. MICROBIOL.

A

5L559 6978 839B65J 70 1

i0

.5559 6; 1 7003 qi 96124

III II. 137l0i 150

126B

55 63 79s 05 Bt 1 IBR 121~ Iil

2

167 120 Ii

*X1'*SXI'*5

B4

226

195,1 211

200

+Xt,O02.0

268

1 2H9 253 279250

*X3*OX3 .0*

160

139 151

210

193 C.160 00 100 120 140 1608 1TI0 200 T T 220 240 T 260

100

Be-20.70.

60.

50240.38.20.10l0 ..55 63 I 7.9 9052 109

*Xt *0Xt2o'6

'X3 *OX3*0+*

*XtS*0XIS.*,

160

121 .11. 134139 151

210

193

252

1.80 10 12li0 1t0 0 22020- -

FIG. 2. Comparison of mass spectra of derivatized metabolites from the degradation of quercetin with those of authentic compounds. (A)2,4,6-Triacetoxybenzoic acid methyl ester; (B) phloroglucinol triacetate; (C) 3,4-diacetoxybenzaldehyde; (D) 3,4-diacetoxybenzoic acidmethyl ester. Mass spectra of metabolites are given in the upper panels; authentic standards are given in the lower panels.

1521090e0706050403020100

51

104

*xIs sxI*s*

*X3 *XP'S+

*XI5O0xIs.*o.

226

124-mi -.-- JL

10137 1627 170.

15I

195

200211

260

253 ,. 229T T 8250 T T TT300

*X3 *8X3*0

*x1S *oXISg.O *100.

Bo0070605040.30.20.10l050

100.9B00706B50403820100

310300

4XI5'OX25I*O

252

964 ALLARD ET AL.

1!

SULFATE AND DECHLORINATION OF 3,4,5-TRICHLOROCATECHOL 965

130C

_60, 6,9 7.3 71,1. 085 98160 00

109105.i. 119 151 1631 120 *a 160too 120 140 160

*Xi.IvXI,*s

l0ie '

X30 *X30*0 *

222

287280 '220

91 109l I. 119

*XI *5XI' *S

180

151 163

222

282 1,118380 100 120 140 160 100 200 220

,,,, 5,96,3 69 757,9

16B

109R 12319J39 I t 151 163 ii.

*X3 *0X3* +*

210

179193

60 B0 19 1280 140 160 l10 200

Ii

137

,,,, 5,96,3,6, 75..7,5,I109

9,195 123 5.1 163,IL

11

*X3 *0X3 0

0X50 *0XSO *0

252

221220 240 260

.X50.X50 *0

210

179

60 B0 10 120 140 160 180

FIG. 2-Continued.

193

252

ill1 221

200T T 220 2i0 T I260

100.90!8070.60.50.40-3020.10.A

100908070605040302010le

xX30 0X30 '0

11, b663 0174 221.1.60

D109080706050403020180

100901Be.706o5040.38.20.100

VOL. 58, 1992

I

1,32

I;9

APPL. ENVIRON. MICROBIOL.

//

CHO

OH

OH

H02C

OH

OH

OH

HO OH

FIG. 3. Proposed pathway for the anaerobic degradation ofquercetin, based on metabolites identified in this study.

components, this is biotransformation in the strict sense ofthe word.The results of this study clearly supported the conclusion

of earlier investigations (1, 22) that dechlorination of chlo-rocatechols may effectively be accomplished by organismsenriched with substrates which do not have chlorine substi-tuents, and they increased the structural spectrum of com-pounds plausibly related to organic components of natural

TABLE 1. Summary of dechlorination experiments usingdifferent enrichment cultures: yields of dichlorocatecholsproduced by dechlorination of 3,4,5-trichlorocatechol,

the principal isomers formed, and the timerequired to achieve the stated yields

Enrichment, Yield (%) Isomer Time (days)sulfate concn

Esculin, low 0 80Esculin, high 0 80Esculetin, low 100 3,5- 30Esculetin, high 0 40Naringin, low 0 40Naringenin, low 40 4,5- 40Naringenin, high 0 50Fraxin, low 5 4,5- 50Fraxin, high 25 4,5- 50Fraxetin, low 30 3,5- 10Fraxetin, high 30 3,5- 10Quercetin, low 85 4,5- 20Quercetin, high 85 4,5- 20Acetate, low 10 4,5- 30Acetate, high 0 60

sediments which were able to support growth under anaer-obic conditions and carry out 0-demethylation and dechlo-rination reactions.Although the degradation of flavanones by different

groups of anaerobic bacteria has been investigated previ-ously (2, 13, 14, 28), the wider metabolic potentialities ofthese organisms have not been explored previously. Thepresent results showed that a range of natural plant-derivedsubstrates, including coumarins and flavanones, were able toserve as carbon sources and were degraded under anaerobicconditions with the formation of a series of hydroxylatedbenzoic acids, some of which were further metabolized. Thestructures of these metabolites are consistent with the path-way of degradation proposed by other workers (2, 13, 14, 28)for structurally related compounds and add further details tothe pathway for degradation of quercetin (Fig. 3).The cultures obtained by enrichment with the coumarins

and the flavanone were able to carry out reactions unrelatedto those involved in degradation of the growth substrates:0-demethylation of 4,5,6-trichloroguaiacol and dechlorina-tion of 3,4,5-trichlorocatechol. Since the glycones containeda carbohydrate moiety covalently bound to the aromaticstructures, it was possible to compare the influence of areadily degraded carbohydrate group and a less readilydegraded aromatic group on the 0-demethylation and de-chlorination reactions.Comparison of the metabolic capability of the cultures

obtained from the glycones with those from the aglyconesshowed that dechlorination was more effectively carried outby the cultures grown with the aglycones, though low yieldswere obtained from the fraxin enrichment. The presence ofthe readily degraded carbohydrate moiety in the originalenrichment apparently led to the development of culturesunable to carry out dechlorination of 3,4,5-trichlorocatechol.These results suggest therefore that the simultaneous pres-ence of a readily degradable substrate such as a carbohy-drate and a less readily assimilable one may be inimical tothe development of organisms capable of dechlorinatingchlorocatechols. Carbohydrates alone were not used forenrichment, since these are readily degradable by the pre-dominating facultative anaerobes which dominate sedimentsfrom many areas of the Baltic. Instead, we used acetatewhich cannot be used by these organisms under anaerobicconditions. The results obtained with the acetate enrichmentfurther substantiated the foregoing conclusions on the coun-terselective role of readily degraded growth substrates onenrichment for organisms with dechlorination capability andwere consistent also with observations that acetate was notan effective electron donor for dechlorination of 3-chlo-robenzoate by Desulfomonile tiedjei (3).Whereas these observations may be of restricted rele-

vance in natural ecosystems where the concentrations ofreadily degraded substrates such as carbohydrates and low-molecular-weight carboxylic acids will generally be ex-tremely low, they may be of considerable importance insystems used for anaerobic biological treatment of industrialwaste. Whereas addition of, for example, glucose mayfacilitate biodegradation of pentachlorophenol in an anaero-bic reactor (10), the present results suggest that such addi-tions would be counterselective to enrichment of the appro-priate dechlorinating organisms from the initial inoculum.Attention should therefore be directed to this issue inassessing, for example, the relative merits of systems fortreating conventional bleachery effluents (4) which containsubstantial concentrations of readily degraded organic sub-stances as well as a range of chlorinated compounds, includ-

OH 0OH

HO N0 H\NOH

OH

OH

r NiCHOHO> OH

OH

C02H

HO OH

966 ALLARD ET AL.

SULFATE AND DECHLORINATION OF 3,4,5-TRICHLOROCATECHOL

ing 3,4,5-trichlorocatechol. It is clearly premature, however,to generalize our conclusions to structurally different orga-nochlorine compounds. Consistent with the results of aprevious study (1), different dichlorocatechol isomers wereproduced from 3,4,5-trichlorocatechol: indeed, all three iso-mers, 3,4-dichloro-, 4,5-dichloro-, and 3,5-dichlorocatechol,have been produced by cultures enriched with differentcarbon sources. The ultimate fate of 3,4,5-trichlorocatecholis therefore difficult to predict in view of the apparentresistance of some of the dichlorocatechols to further de-chlorination (1).There was a significant difference in the effect of high

concentrations of sulfate present during the initial enrich-ments and its effect on actively dechlorinating culturesobtained by enrichment with low sulfate concentrations.Most cultures from initial enrichments using high sulfateconcentrations were significantly less effective in carryingout the dechlorination of 3,4,5-trichlorocatechol than thosein which low sulfate concentrations were present. In thesecases, the presence of sulfate at the higher concentrationwas apparently counterselective to the development of pop-ulations of organisms carrying out dechlorination-presum-ably because of the competition in these populations ofsulfate for reducing equivalents. On the other hand, dechlo-rination of 3,4,5-trichlorocatechol was effectively carried outin experiments in which high concentrations of sulfate wereadded to cultures from low-sulfate enrichments: the pres-ence of high concentrations of sulfate did not thereforeinhibit dechlorination by these cultures. This is consistentwith observations of pure cultures of D. tiedjei (3). On theother hand, enrichments with quercetin with high or lowsulfate concentrations dechlorinated 3,4,5-trichlorocatecholto 4,5-dichlorocatechol equally rapidly.These results are in broad agreement with those from

other laboratories (6-8) which have illustrated the counter-selective effect of sulfate in development of cultures able todechlorinate chloroaromatic compounds. On the other hand,slow degradation of monochlorophenols and 2,4-dichlo-rophenol by sediment slurries in sulfate-containing mediumhas been demonstrated (9, 12), and the presence of sulfate inaquifer slurries did not affect the dechlorination of tetrachlo-roaniline, though further dechlorination of the initiallyformed 2,3,5-trichloroaniline was inhibited (15). Collec-tively, the data suggest that in natural environments, thesalinity-and concomitantly the sulfate concentration-ofthe system may play a cardinal role in the persistence ofchlorophenolic and other organochlorine compounds. Inaddition, the presence of significant concentrations of sulfatein wastewater streams containing organochlorine com-pounds may present a problem in anaerobic biological treat-ment systems which has not apparently been consideredhitherto.The isomers of dichlorocatechol that were formed by

dechlorination of 3,4,5-trichlorocatechol were influenced by(i) the sulfate concentration in the medium and (ii) whetherthe growth substrate was conjugated with a carbohydrate.Comparable differences in the isomers formed by dechlori-nation of a wider range of chlorocatechols were found in aprevious study using cultures which had been grown withphloroglucinol and catechin (1). All these results may, how-ever, simply reflect the existence of metabolically distinctpopulations of bacteria present in the various enrichmentcultures. In the environment, a range of products maytherefore be expected to result from the anaerobic dechlori-nation of chlorocatechols.On the basis of the collective evidence, anaerobic dechlo-

rination of chlorocatechols may be expected to occur insediments containing natural plant detritus consisting ofcompounds structurally related to coumarins and fla-vanones. Support for the role of compounds similar to thoseused as growth substrates in this study was obtained fromthe identification in sediment samples of a series of com-pounds structurally related to the metabolites identified inthis investigation: aromatic acids, including benzoate, phe-nylacetate, 4-hydroxyphenylacetate, and phenylpropionate.Whereas the results of this and previous studies (1, 22)

have clearly demonstrated the possibility of dechlorinationof catechols in laboratory experiments, extrapolation tonatural ecosystems should be made cautiously. Evaluationof persistence in the environment would require access toresults from monitoring over a prolonged period of time-comparable, for example, to that carried out for neutralorganochlorine compounds such as 1,1-bis(4-chlorophenyl)-2,2,2-trichloroethane (DDT) and polychlorinated biphenyls(25). The cardinal issue is the rate at which dechlorinationtakes place under natural conditions, and this cannot realis-tically be assessed from the kinds of experiments presentedhere. Evidence from laboratory experiments in which con-taminated sediments have been maintained under anaerobicconditions for extended periods (24) suggests, however, thatat least the more highly chlorinated catechols are apprecia-bly persistent under natural conditions even in highly con-taminated sediments. The degree of accessibility of the testsubstrates to the relevant organisms (17, 20) is, however, anunresolved issue.

ACKNOWLEDGMENT

We thank the Knut and Alice Wallenberg Foundation for provi-sion of funding towards the purchase of a mass spectrometer.

REFERENCES1. Allard, A.-S., P.-A. Hynning, C. Lindgren, M. Remberger, and

A. H. Neilson. 1991. Dechlorination of chlorocatechols by stableenrichment cultures of anaerobic bacteria. Appl. Environ. Mi-crobiol. 57:77-84.

2. Cheng, K.-G., H. G. Krishnamurty, G. A. Jones, and F. J.Simpson. 1971. Identification of products produced by theanaerobic degradation of naringin by Btiyrivibrio sp. C3. Can. J.Microbiol. 17:129-131.

3. DeWeerd, K. A., F. Concannon, and J. F. Suflita. 1991. Rela-tionship between hydrogen consumption, dehalogenation, andthe reduction of sulfur oxyanions by Desulfomonile tiedjei.Appl. Environ. Microbiol. 57:1929-1934.

4. Fitzsimons, R., M. Ek, and K.-E. L. Eriksson. 1990. Anaerobicdechlorination/degradation of chlorinated organic compounds ofdifferent molecular masses in bleach plant effluents. Environ.Sci. Technol. 24:1744-1748.

5. Gantzer, C. J., and L. P. Wackett. 1991. Reductive dechlorina-tion catalyzed by bacterial transition-metal coenzymes. Envi-ron. Sci. Technol. 25:715-722.

6. Genthner, B. R. S., W. A. Price, and P. H. Pritchard. 1989.Anaerobic degradation of chloroaromatic compounds in aquaticsediments under a variety of enrichment conditions. Appl.Environ. Microbiol. 55:1466-1471.

7. Gibson, S. A., and J. M. Suflita. 1986. Extrapolation of biodeg-radation results to groundwater aquifers: reductive dehalogena-tion of aromatic compounds. Appl. Environ. Microbiol. 52:681-688.

8. Gibson, S. A., and J. M. Suflita. 1990. Anaerobic biodegradationof 2,4,5-trichlorophenoxyacetic acid in samples from a meth-anogenic aquifer: stimulation by short-chain organic acids andalcohols. Appl. Environ. Microbiol. 56:1825-1832.

9. Haggblom, M. M., and L. Y. Young. 1990. Chlorophenol deg-radation coupled to sulfate reduction. Appl. Environ. Micro-biol. 56:3255-3260.

967VOL. 58, 1992

APPL. ENVIRON. MICROBIOL.

10. Hendriksen, H. V., S. Larsen, and B. K. Ahring. 1991. Anaero-bic degradation of PCP and phenol in fixed-film reactors: theinfluence of an additional substrate. Water Sci. Technol. 24:431-436.

11. Hynning, P.-A., M. Remberger, and A. H. Neilson. 1989. Syn-thesis, gas-liquid chromatographic analysis and gas chromato-graphic-mass spectrometric identification of nitrovanillins, chlo-ronitrovanillins, nitroguaiacols and chloronitroguaiacols. J.Chromatogr. 467:99-110.

12. Kohring, G.-W., X. Zhang, and J. Wiegel. 1989. Anaerobicdechlorination of 2,4-dichlorophenol in freshwater sediments inthe presence of sulfate. Appl. Environ. Microbiol. 55:2735-2737.

13. Krishnamurty, H. G., K.-J. Cheng, G. A. Jones, F. J. Simpson,and J. E. Watkin. 1970. Identification of products produced bythe anaerobic degradation of rutin and related flavonoids byButyrivibrio sp. C3. Can. J. Microbiol. 16:759-767.

14. Krumholz, L. R., and M. P. Bryant. 1986. Eubactenum oxi-doreducens sp. nov. requiring H2 or formate to degrade gallate,pyrogallol, phloroglucinol and quercetin. Arch. Microbiol. 144:8-14.

15. Kuhn, E. P., G. T. Townsend, and J. M. Suflita. 1990. Effect ofsulfate and organic carbon supplements on reductive dehaloge-nation of chloroanilines in anaerobic aquifer slurries. Appl.Environ. Microbiol. 56:2630-2637.

16. Larsen, S., H. V. Hendriksen, and B. K. Ahring. 1991. Potentialfor thermophilic (50°C) anaerobic dechlorination of pentachlo-rophenol in different ecosystems. Appl. Environ. Microbiol.57:2085-2090.

17. Neilson, A. H. 1989. Factors determining the fate of organicchemicals in the environment: the role of bacterial transforma-tions and binding to sediments, p. 74-112. In L. Landner (ed.),Chemicals in the aquatic environment. Springer-Verlag KG,Berlin.

18. Neilson, A. H. 1990. The biodegradation of halogenated organiccompounds. J. Appl. Bacteriol. 69:445-470.

19. Neilson, A. H., A.-S. Allard, P.-A. Hynning, and M. Remberger.

1988. Transformations of halogenated aromatic aldehydes bymetabolically stable anaerobic enrichment cultures. Appl. En-viron. Microbiol. 54:2226-2236.

20. Neilson, A. H., A.-S. Allard, P.-A. Hynning, and M. Remberger.1991. Distribution, fate and persistence of organochlorine com-pounds formed during production of bleached pulp. Toxicol.Environ. Chem. 30:3-41.

21. Neilson, A. H., A.-S. Allard, P.-A. Hynning, M. Remberger, andL. Landner. 1983. Bacterial methylation of chlorinated phenolsand guaiacols: formation of veratroles from guaiacols and high-molecular-weight chlorinated lignin. Appl. Environ. Microbiol.45:774-783.

22. Neilson, A. H., A.-S. Allard, C. Lindgren, and M. Remberger.1987. Transformations of chloroguaiacols, chloroveratroles, andchlorocatechols by stable consortia of anaerobic bacteria. Appl.Environ. Microbiol. 53:2511-2519.

23. Remberger, M., A.-S. Allard, and A. H. Neilson. 1986. Biotrans-formations of chloroguaiacols, chlorocatechols, and chlorover-atroles in sediments. Appl. Environ. Microbiol. 51:552-558.

24. Remberger, M., P.-A. Hynning, and A. H. Neilson. 1988. Com-parison of procedures for recovering chloroguaiacols and chlo-rocatechols from contaminated sediments. Environ. Toxicol.Chem. 7:795-805.

25. Schmitt, C. J., J. L. Zajicek, and P. H. Peterman. 1990. Nationalcontaminant biomonitoring program: residues of organochlorinechemicals in U.S. freshwater fish, 1976-1984. Arch. Environ.Contam. Toxicol. 19:748-781.

26. Stupperich, E., H.-J. Eisinger, and S. Schurr. 1990. Corrinoidsin anaerobic bacteria. FEMS Microbiol. Rev. 87:355-360.

27. Van Dort, H. M., and D. L. Bedard. 1991. Reductive ortho andmeta dechlorination of a polychlorinated biphenyl congener byanaerobic microorganisms. Appl. Environ. Microbiol. 57:1576-1578.

28. Winter, J., L. H. Moore, V. R. Dowell, and V. D. Bokkenheiser.1989. C-ring cleavage of flavonoids by human intestinal bacte-ria. Appl. Environ. Microbiol. 55:1203-1208.

968 ALLARD ET AL.