Vol. 45, No. 2 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 1983, p. 428-435 0099-2240/83/020428-08$2.00/0 Copyright C) 1983, American Society for Microbiology Adaptation of Natural Microbial Communities to Degradation of Xenobiotic Compounds: Effects of Concentration, Exposure Time, Inoculum, and Chemical Structuret JIM C. SPAIN* AND P. A. VAN VELD U.S. Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, Florida 32561 and Gulf Coast Research Laboratory, Ocean Springs, Mississippi 39564 Received 24 May 1982/Accepted 3 October 1982 Adaptation of microbial communities to faster degradation of xenobiotic compounds after exposure to the compound was studied in ecocores. Radiola- beled test compounds were added to cores that contained natural water and sediment. Adaptation was detected by comparing mineralization rates or disap- pearance of a parent compound in preexposed and unexposed cores. Microbial communities in preexposed cores from a number of freshwater sampling sites adapted to degrade p-nitrophenol faster; communities from estuarine or marine sites did not show any increase in rates of degradation as a result of preexposure. Adaptation was maximal after 2 weeks and was not detectable after 6 weeks. A threshold concentration of 10 ppb (10 ng/ml) was observed; below this concentra- tion no adaptation was detected. With concentrations of 20 to 100 ppb (20 to 100 ng/ml), the biodegradation rates in preexposed cores were much higher than the rates in control cores and were proportional to the concentration of the test compound. In addition, trifluralin, 2,4-dichlorophenoxyacetic acid, and p-cresol were tested to determine whether preexposure affected subsequent biodegrada- tion. Microbial communities did not adapt to trifluralin. Adaptation to 2,4- dichlorophenoxyacetic acid was similar to adaptation to nitrophenol. p-Cresol was mineralized rapidly in both preexposed and unexposed communities. Laboratory studies of pollutant biodegrada- tion are best conducted by using mixed culture systems taken from the field. The goal of such studies is to use degradation rates measured in the laboratory to predict degradation rates in the environment. A number of factors, such as tem- perature, salinity, pH, redox potential, microbi- al biomass, and prior exposure, can affect the degradation rate and, thus, the fate of a toxicant. If laboratory studies are to be used to predict biodegradation rates in the field, it is important that the rate-determining factors be understood. Previous work has shown that adaptation of microorganisms can play a major role in deter- mining biodegradation rates (3, 4, 6, 7, 9). Adap- tation is defined as a change in the microbial community that increases the rate of transforma- tion of a test compound as a result of a prior exposure to the test compound. This is an opera- tional definition, and no attempt is made to distinguish among mechanisms, such as gene transfer or mutation, enzyme induction, and population changes (7). It is likely that each of t Contribution 440 from the Gulf Breeze Laboratory. these mechanisms contributes to the changes in biodegradation rates observed after natural mi- crobial communities are exposed to new sub- strates. Changes in biodegradation rates due to perturbation of the system during sample collec- tion or subsequent laboratory manipulation are excluded from consideration by the use of prop- er controls. The effects of adaptation can be dramatic (7). Even when xenobiotic compounds are added at concentrations below 100 ppb (100 ng/ml), the degradation rates can be 1,000-fold higher in populations that are preexposed to the com- pound. The study of Spain et al. (7) raised several questions concerning the duration, con- centration dependence, site dependence, and compound dependence of microbial adaptation. In this report we describe attempts to answer these questions. MATERIALS AND METHODS Test system. Biodegradation of test compounds and adaptation of microbial communities were measured in ecocores as described previously (7). The ecocore test systems were filled with sediment and water from the test sites. Samples from all but the shallowest sites were collected by divers. Radiolabeled test com- 428 on September 11, 2020 by guest http://aem.asm.org/ Downloaded from
Transcript
Vol. 45, No. 2APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 1983, p. 428-4350099-2240/83/020428-08$2.00/0Copyright C) 1983, American Society for Microbiology
Adaptation of Natural Microbial Communities to Degradationof Xenobiotic Compounds: Effects of Concentration, Exposure
Time, Inoculum, and Chemical StructuretJIM C. SPAIN* AND P. A. VAN VELD
U.S. Environmental Protection Agency, Environmental Research Laboratory, Gulf Breeze, Florida 32561 andGulf Coast Research Laboratory, Ocean Springs, Mississippi 39564
Received 24 May 1982/Accepted 3 October 1982
Adaptation of microbial communities to faster degradation of xenobioticcompounds after exposure to the compound was studied in ecocores. Radiola-beled test compounds were added to cores that contained natural water andsediment. Adaptation was detected by comparing mineralization rates or disap-pearance of a parent compound in preexposed and unexposed cores. Microbialcommunities in preexposed cores from a number of freshwater sampling sitesadapted to degrade p-nitrophenol faster; communities from estuarine or marinesites did not show any increase in rates of degradation as a result of preexposure.Adaptation was maximal after 2 weeks and was not detectable after 6 weeks. Athreshold concentration of 10 ppb (10 ng/ml) was observed; below this concentra-tion no adaptation was detected. With concentrations of 20 to 100 ppb (20 to 100ng/ml), the biodegradation rates in preexposed cores were much higher than therates in control cores and were proportional to the concentration of the testcompound. In addition, trifluralin, 2,4-dichlorophenoxyacetic acid, and p-cresolwere tested to determine whether preexposure affected subsequent biodegrada-tion. Microbial communities did not adapt to trifluralin. Adaptation to 2,4-dichlorophenoxyacetic acid was similar to adaptation to nitrophenol. p-Cresolwas mineralized rapidly in both preexposed and unexposed communities.
Laboratory studies of pollutant biodegrada-tion are best conducted by using mixed culturesystems taken from the field. The goal of suchstudies is to use degradation rates measured inthe laboratory to predict degradation rates in theenvironment. A number of factors, such as tem-perature, salinity, pH, redox potential, microbi-al biomass, and prior exposure, can affect thedegradation rate and, thus, the fate of a toxicant.If laboratory studies are to be used to predictbiodegradation rates in the field, it is importantthat the rate-determining factors be understood.
Previous work has shown that adaptation ofmicroorganisms can play a major role in deter-mining biodegradation rates (3, 4, 6, 7, 9). Adap-tation is defined as a change in the microbialcommunity that increases the rate of transforma-tion of a test compound as a result of a priorexposure to the test compound. This is an opera-tional definition, and no attempt is made todistinguish among mechanisms, such as genetransfer or mutation, enzyme induction, andpopulation changes (7). It is likely that each of
t Contribution 440 from the Gulf Breeze Laboratory.
these mechanisms contributes to the changes inbiodegradation rates observed after natural mi-crobial communities are exposed to new sub-strates. Changes in biodegradation rates due toperturbation of the system during sample collec-tion or subsequent laboratory manipulation areexcluded from consideration by the use of prop-er controls.The effects of adaptation can be dramatic (7).
Even when xenobiotic compounds are added atconcentrations below 100 ppb (100 ng/ml), thedegradation rates can be 1,000-fold higher inpopulations that are preexposed to the com-pound. The study of Spain et al. (7) raisedseveral questions concerning the duration, con-centration dependence, site dependence, andcompound dependence of microbial adaptation.In this report we describe attempts to answerthese questions.
MATERIALS AND METHODSTest system. Biodegradation of test compounds and
adaptation of microbial communities were measured inecocores as described previously (7). The ecocore testsystems were filled with sediment and water from thetest sites. Samples from all but the shallowest siteswere collected by divers. Radiolabeled test com-
FIG. 1. Sampling sites on the Escambia River near Pensacola, Fla. The water depth at all sites was between 1and 2 m. Mi, Mile.
pounds were added without carrier, and parent com-pound disappearance and mineralization were moni-tored. Adaptation was demonstrated by measuring thedegradation of the test compounds in cores preex-posed to the compounds for appropriate intervals. Thedegradation rates in preexposed cores were then com-pared with the degradation rates in control cores, anddifferences were taken to indicate adaptation.
Extraction and analysis. Aqueous samples contain-ing p-nitrophenol (PNP), 2,4-dichlorophenoxyaceticacid (2,4-D), and p-cresol were acidified with concen-trated HCI (pH 1 to 2) and extracted with equalvolumes of ethyl acetate. Extracts were filteredthrough Fluoropore filters (pore size, 0.5 pm; MilliporeCorp, Bedford, Mass.) and analyzed by high-pressureliquid chromatography. High-pressure liquid chroma-tographic analyses were performed with a WatersAssociates liquid chromatograph equipped with a radi-al compression column and a UV absorbance detector.Samples that contained radiolabeled test compoundswere supplemented with unlabeled compounds at a
concentration of 50 mg/liter to allow detection by UVabsorbance. Peaks that contained radioactivity were
collected, and the radioactivity was measured byliquid scintillation counting. PNP was separated on a
silica gel column by using chloroform-methanol-aceticacid (160:40:1) as the mobile phase. UV absorbancewas monitored at 254 nm; p-Cresol was separated byreverse-phase chromatography, using an octadecysil-ane-bonded silica column and methanol-phosphatebuffer (80:20) as the mobile phase. The phosphatebuffer was 0.1 M phosphoric acid adjusted to pH 3.5
with sodium hydroxide. p-Cresol was detected byfollowing absorbance at 280 nm. 2,4-D was separatedby reverse-phase chromatography, using acetonitrile-1.0o acetic acid (1:1) as the mobile phase. UV absor-bance was measured at 280 nm.
Trifluralin was extracted from aqueous solutionswith equal volumes of hexane, and the extracts wereanalyzed by gas chromatography. Gas chromatograph-ic analyses were performed with a Hewlett Packardmodel 5830 gas chromatograph equipped with an elec-tron capture detector. The column (2 mm by 2 m) waspacked with 3% OV-101 on Chromosorb HP (80/100mesh) with support-bonded Carbowax-20M. The col-umn temperature was 155°C, and the flow rate of theargon-methane (90:10) carrier gas was 31 cm3/min.
Culture media and conditions. The culture media andthe methods used for enumeration and selection ofbacteria have been described previously (7).
Materials. PNP was obtained from Matheson, Cole-man and Bell Inc., Norwood, Ohio; p-cresol was fromAldrich Chemicals, Milwaukee, Wis.; aaa-trifluoro-2,6-dinitro-N,N-dipropyl-p-toluidine (trifluralin), 2,4-D, and 2,4-dichlorophenol were from the U.S. Envi-ronmental Protection Agency, Research TrianglePark, N.C.; [U-14C]PNP and p-[ring-U-14C]cresolwere from Pathfinders, St. Louis, Mo.; [ring-U-14C]2,4-D was from Calbiochem, La Jolla, Calif.; and[ring-U-14C]trifluralin was from New England NuclearCorp., Boston, Mass. The purity of chemicals wasverified by gas chromatography or high-pressure liquidchromatography before use.
Radioactivity measurements. Radioactivity was mea-
sured by liquid scintillation counting as described
previously (7).
RESULTSEffect of sampling site. Preliminary work (7)
demonstrated that microbial communities from atest site on the Escambia River adapted todegrade PNP faster after an initial exposureto the nitro compound. To determine whethercommunities taken from other freshwater sitescould adapt to mineralize PNP, ecocore sampleswere collected at five sites on the EscambiaRiver (Fig. 1). The mineralization data for corestreated initially with radiolabeled PNP (initialcores) are shown in Fig. 2A. After a lag periodduring which mineralization was negligible, therates of 14Co2 release increased substantially inall samples. When PNP was added to controlcores after 8 days of incubation, the lag periodand subsequent mineralization rates were virtu-ally identical to those of the initial cores (Fig.2B). In contrast, when cores preexposed tounlabeled PNP were exposed to [ 4C]PNP, com-
FIS CAMP
1 50F CHEMICAL PLANr
J 40
8 304
2020-ITJ
10
00 40 80 120 160 200 240
TIME , hours
FIG. 2. Mineralization of [14C]PNP in cores fromvarious sites on the Escambia River. (A) Cores weretreated with (14C]PNP (100 nmol/core) and incubatedat 220C. The 1'CO2 released from the cores wastrapped, and the radioactivity was measured by liquidscintillation counting. (B) Control cores were incubat-ed for 8 days with no supplement and then tested with100 nmol of ['4C]PNP. (C) Preexposed cores wereincubated for 8 days with 100 nmol of PNP per coreand then tested with 100 nmol of [14C]PNP.
munities from all sites were clearly adapted todegrade the nitro compound (Fig. 2C). Most ofthe communities adapted to a greater extent thanthe communities at the original test site (Fig. 1,site A).To test whether adaptation occurred in estua-
rine and marine communities, seven sites in theEscambia River estuary and surrounding area(Fig. 3) were sampled and tested for adaptationto PNP. The original river test site (Fig. 1, siteA) provided a freshwater control sample; theother test sites sampled were in Escambia Bay,the Gulf of Mexico, and Range Point saltmarshin Santa Rosa Sound. Only the initial cores wereused (Fig. 4) because there was no evidence ofmineralization in any of the saltwater samples.The river sample was the only sample thatcaused any appreciable mineralization of PNP.With the Escambia River core, there was adramatic increase in the rate of 14CO2 releaseafter 4 days of incubation. The bacteria in eco-cores were enumerated at the beginning of theexperiment (Table 1). There was no apparent
FIG. 3. Test sites in the Escambia River estuary and adjacent areas. Site 1 is on the Escambia River; sites 2,3, and 5 are in Pensacola Bay; site 6 is in Range Point saltmarsh; and sites 4 and 7 are in the Gulf of Mexico. Mi,Mile.
correlation between mineralization rate and totalbacterial biomass. Attempts to isolate PNP-degrading bacteria from the estuarine and salinesites failed. In contrast, PNP degraders were
readily isolated from the river test site (7).Subsequent experiments in which preexposedand control cores from the same sites were
tested with [14C]PNP gave identical results, and
F
8
0 1 2 3 4 5 6 7 8 9 10 11 12 13
TIME, doy
FIG. 4. Mineralization of PNP in ecocores fromdiverse sampling sites on the Escambia River estuaryand nearby Gulf of Mexico (see Fig. 3). Cores weretreated with 1.1 F.M [14C]PNP and incubated at 20°C.'4CO2 was trapped, and the radioactivity was mea-sured by liquid scintillation counting. Data are aver-
ages from duplicate cores. The upper curve representsdata from Escambia River cores. The mineralizationrates in cores from all other sites (lower curve) were
indistinguishable.
there was no evidence of adaptation in samplesfrom saline environments.
Duration of adaptation. To determine whetheradaptation is important in the environment, it isnecessary to know how long communities re-
main adapted after exposure to a xenobioticcompound. We designed an experiment to deter-mine how long adaption persists after exposureof river communities to PNP. A group of eco-
core samples from the Escambia River weretreated with unlabeled PNP. A similar group ofcontrol cores received no treatment. After incu-bation for different intervals, [14C]PNP was add-ed to preexposed and control cores, and miner-alization was studied by measuring 14CO2release (Fig. 5). The initial rate of 14CO2 releasewas used as a measure of adaptation. No miner-
TABLE 1. Bacterial biomass in ecocores fromvarious sites on the Escambia River estuary and the
nearby Gulf of Mexico
Site S Water Total bacteriano a SIte depth (m) (CFU/ml, x104)
1 Escambia River 2.0 4.92 Pensacola Bay, 6.0 18.4
Bayou Chico3 Pensacola Bay 12.0 4.54 Gulf, Nearshore 6.0 0.425 Pensacola Bay 4.6 4.4
FIG. 5. Duration of adaptation. Cores were takenfrom the Escambia River sampling site, preexposed to1.08 ,uM PNP, and tested with [14C]PNP after 0, 313,383, 473, 669, and 1,097 h. The 14Co2 released fromthe cores in the first 36 h of the test was trapped andmeasured by scintillation counting.
alization was detected in the control cores dur-ing the course of the experiment. Adaptationwas maximal 2 weeks after exposure and fellslowly until it was no longer detectable after 7weeks. Because bacterial activity in the coresincubated for the longer times was likely to havebeen less than optimal, adaptation in the envi-ronment would probably be different. Therefore,the times indicated by a static experiment suchas the one described here would underestimatethe duration of adaptation in natural systems.
Concentration effects. To determine whetherthe degree of adaptation depended on the con-centration of PNP used during preexposure oron the concentration used in the biodegradationassay, ecocores were prepared with sedimentand water from the Escambia River test site andtreated in the following ways: (i) different preex-posure concentrations and test concentrationsthe same as the preexposure concentrations(cores were preexposed to different concentra-tions of PNP, incubated for 5 days to allowcommunities to adapt, and then tested withradiolabeled PNP at the same concentrationsused for preexposure); (ii) different preexposureconcentrations and a test concentration of 0.72,uM (cores were preexposed to different concen-trations of PNP and incubated for 5 days, andthen 0.72 ,uM radiolabeled PNP was added toeach core); and (iii) 0.72 ,uM PNP preexposureand different test concentrations (cores werepreexposed to 0.72 ,uM PNP, incubated for 5days, and then tested with different concentra-tions of radiolabeled PNP).The rate of mineralization of PNP in adapted
cores (Fig. 6A) depended on the PNP concentra-tions used for preexposure and biodegradationtests. The mineralization rates in ecocorespreexposed to identical concentrations and test-ed at different concentrations depended on the
60
50W
e 40
0 30
w 20
8 0
60
o 50
(n
e 40c20 30w
w 20
CY0q1 0
PREEXPOSED TO 126 nMOLES / CORE
RESPIKED AT INDICATED CONCENTRATIONS
(C) PREEXPOSED AT INDICATED CONCENTRATION252
50.4
76.0o 126
12.6
0O
20 40 60 80 100
HOURS
FIG. 6. Effects of concentration on adaptation. (A)Cores were preexposed to unlabeled PNP at the indi-cated concentrations (nanomoles per core) for 6 daysand then tested with [14C]PNP at the same concentra-tions, and the "4CO2 released from the cores wastrapped and measured. (B) Cores were preexposed to126 nmol of PNP per core for 6 days and then testedwith [14C]PNP at the indicated concentrations. (C)Cores were preexposed to PNP at the indicated con-centrations for 6 days and then tested with 126 nmol of[14]PNP.
concentration ofPNP used in the biodegradationtest (Fig. 6B).Communities preexposed to different concen-
greater (Fig. 6C). Preexposure concentrationsinfluenced the rate to some extent; there was a
slight lag in mineralization rate in communitiespreexposed to 25.2, 50.4, or 76 nmol/core, butafter 70 h the total amounts of 14Co2 releasedwere similar. Communities in control cores thatreceived no preexposure gave no evidence ofadaptation to faster degradation of the nitrocompound, and communities preexposed to 12.6nmol/core showed only slight adaptation whenthey were tested with 126 nmol/core.Each core in the experiment described above
was matched with two identically treated cores
in which both parent compound disappearanceand 14CO2 release were followed. The resultsconfirmed the mineralization data shown in Fig.6.
Adaptation with other compounds. Trifluralin,PNP, p-cresol, and 2,4-D were tested in theecocore system to determine whether the micro-bial communities from the river test site couldadapt to degrade other compounds. For eachcompound, two cores were supplemented initial-ly with 14C-labeled compound; one of these wasused to follow mineralization, and the other wassampled at intervals to measure the concentra-tion of the parent compound in the water. Threecores were treated with unlabeled test com-pound, and three control cores received noaddition. Mineralization in the cores treatedinitially with radiolabeled p-cresol was rapid andimmediate. With 2,4-D and PNP, there was a lagperiod before rapid mineralization occurred. Nosignificant mineralization was detected in thecores that contained trifluralin. After the rates ofmineralization began to decrease in the initialcores that contained PNP, p-cresol, or 2,4-D,14C-labeled test compounds were added to con-trol and preexposed cores. Samples were re-
moved from two control and two preexposedcores for analysis of the parent compound. The14CO2 released from all three cores was trappedand used as a measure of mineralization.Cores preexposed to PNP rapidly released
14Co2, and there was a concomitant, rapid dis-appearance of the parent compound (Fig. 7A).Mineralization and disappearance of the parentcompound were much slower in control cores.
Similar results were obtained with 2,4-D (Fig.7B), and there was no accumulation of 2,4-dichlorophenol.Cores tested with p-cresol gave identical rates
of mineralization and parent compound disap-pearance regardless of whether they were preex-posed (Fig. 7C). In both cases, p-cresol was
attacked rapidly by the microbial community.Trifluralin was degraded only very slowly,
and there were no differences between rates in
preexposed communities and those in controlcommunities (Fig. 7D). No radioactivity wasreleased as 14CO2.
DISCUSSIONOur results indicate that adaptation of aquatic
microbial communities can last for severalweeks after exposure to a xenobiotic compound.Therefore, adaptation must be considered whenwe predict the fate of xenobiotic compounds inenvironments subjected to repeated or extendedexposures. Biodegradation tests should be donewith adapted microbial communities, and com-pounds that cause adaptation can be expectednot to persist. There is considerable evidence toindicate that such is the case in terrestrial sys-tems as well (6).The ability of communities to adapt varies
widely from site to site. Fresh- and saltwatercommunities clearly differ in the ability to miner-alize PNP. It is possible that, given sufficienttime, organisms from saline environments wouldalso adapt to degrade nitrophenol faster afterexposure, but we have been unable to demon-strate such a response.Baughman et al. (1) have shown that for many
compounds biodegradation rates are proportion-al to the total bacterial biomass. This relation-ship does not seem to be true for compoundsthat are attacked by only a few specific orga-nisms in the test system. Our results stronglysuggest that second-order rate constants basedon total bacterial biomass cannot be used topredict the biodegradation of certain pollutantsthat cause adaptation of the microbial communi-ty. Preliminary evidence suggests that the re-sponse of a community is governed by thepresence or absence of bacteria that are able touse the nitro compound as a carbon source.
Results obtained with different concentrationsof PNP indicate that the extent of adaptationachieved by a microbial community is not direct-ly proportional to the preexposure concentrationof the substrate. There seems to be a thresholdconcentration of PNP below which there is nodetectable adaptation of the community. Thethreshold concentration is similar to the concen-tration indicated in our previous work (7). Ac-tive microbial communities were produced bypreexposure to PNP concentrations above thethreshold, and higher preexposure concentra-tions caused only slight increases in biodegrada-tion rates. The mineralization rates shown byadapted communities were influenced more bythe test concentration than by the preexposureconcentration within the range tested. Futurestudies on the practical applications of this ob-servation should enable us to predict the fate oforganic pollutants more accurately. Predictionof microbial adaptation after exposure to a toxi-
HOURS HOUJRSFIG. 7. Mineralization of radiolabeled test compounds by preexposed and control populations from the
Escambia River. (A) PNP. (B) 2,4-D. (C) p-Cresol. (D) Trifluralin. Cores from the Escambia River test site wereexposed to the test compounds at concentrations of 1.08 pLM; control cores received no supplement. After 8 daysof incubation at 20°C, all cores were tested with 1.08 ,uM radiolabeled compound. During subsequent incubation14C02 release was measured, and water samples were analyzed for parent compounds by high-pressure liquidchromatography or gas chromatography. The values shown are averages of data from duplicate cores; sample-to-sample variation never exceeded 10%o of the mean.
cant will become easier if the extent of adapta-tion is the same over a relatively wide range ofexposure concentrations. Furthermore, primingof microbial communities near potential pollut-ant discharge sites with small amounts of thepollutant might be used to ensure a rapid micro-bial response after a spill.The results of the experiments with various
test compounds provide examples of severalways that microbial communities can respond inan adaptation test. The microbiota clearly de-graded PNP faster as a result of preexposure.There was a lag period of several days beforecommunities exposed to PNP began to attackthe test compound, but communities in corestested after the adaptation period attacked PNPimmediately. The data from the 2,4-D experi-ments lead to similar conclusions.When p-cresol was added to ecocores, the
results were markedly different. There was no
lag period before the population attacked p-
cresol, and preexposure in the laboratory did not
alter biodegradation rates. The following expla-nations for such a rapid attack are possible: (i)the adaptation may have been extremely fastand thus may not have been detected in ourexperiment; (ii) the enzymes for p-cresol degra-dation may have been constitutive in the com-munity; or (iii) the community in the river mayhave been exposed to p-cresol or a similarcompound before the samples were collected,which would have produced microbial popula-tions able to degrade p-cresol immediately. Suchexposure could be chronic at the test site, due tothe presence of cresols or similar compounds inrunoff from agricultural land and forests.
Trifluralin degradation was relatively slow,and there was no detectable adaptation. There-fore, the disappearance of this compound wasdue to abiotic factors or to biodegradation bynonspecific cooxidative processes (5).The results described above would be useful
in predicting the fate of the four test compoundsif the compounds were released in the river.
Trifluralin would biodegrade slowly and at arelatively constant rate, depending on the con-centration. p-Cresol would degrade rapidly andimmediately, whereas PNP and 2,4-D wouldpersist during adaptation of the microorganismsand then biodegrade rapidly.Measurements of mineralization do not al-
ways accurately reflect the disappearance of aparent compound in a biodegradation test (1).The results of our experiments clearly show thatfor PNP and p-cresol, mineralization rates close-ly parallel rates of parent compound disappear-ance. This is probably true in most cases wherethe initial step is rate limiting in a biodegradationsequence that involves complete mineralization.Degradation of the above compounds beginswith incorporation of molecular oxygen into thesubstrate (2, 8); subsequent steps in the pathwayare considerably less complex. In the case oftrifluralin, the initial transformation probablydoes not produce readily degradable products;therefore, mineralization is a poor indicator ofbiotransformation.
ACKNOWLEDGMENTSWe thank the dive team of the U.S. Environmental Protec-
tion Agency Environmental Research Laboratory, GulfBreeze, Fla., for collecting ecocore samples used in thisstudy.
This work was funded in part by U.S. Environmental
Protection Agency contract 68-D1-5043 with Battelle Colum-bus Laboratories and subcontract T-6423 (7197)4)37 with GulfCoast Research Laboratory, Ocean Springs, Miss.
LITERATURE CITED
1. Baughman, G. L., D. F. Paris, and W. C. Steen. 1980.Quantitative expression of biotransformation rate, p. 105-111. In A. W. Maki, K. L. Dickinson, and J. Cairns, Jr.(ed.), Biotransformation and fate of chemicals in the aquat-ic environment. American Society for Microbiology,Washington, D.C.
2. Dagley, S., and M. D. Patel. 1957. Oxidation ofp-cresol andrelated compounds by a Pseudomonas. Biochem. J.66:227-233.
3. Felset, A., J. V. Maddox, and W. Bruce. 1981. Enhancedmicrobial degradation of carbofuran in soils with historiesof Furadan use. Bull. Environ. Contam. Toxicol. 26:781-788.
4. Fournier, J. C., P. Codaccdoni, and G. Soulas. 1981. Soiladaptation to 2,4-D degradation in relation to the applica-tion rates and the metabolic behavior of the degradingmicroflora. Chemosphere 10:977-984.
5. Perry, J. J. 1979. Microbial cooxidations involving hydro-carbons. Microbiol. Rev. 43:59-72.
6. Simon-Sylvestre, G., and J. C. Fournier. 1979. Effects ofpesticides on the soil microflora. Adv. Agron. 31:1-81.
7. Spain, J. C., P. H. Pritchard, and A. W. Bourquin. 1980.Effects of adaptation on biodegradation rates in sediment/water cores from estuarine and freshwater environments.Appl. Environ. Microbiol. 40:726-734.
8. Spain, J. C., 0. Wyss, and D. T. Gibson. 1979. Enzymaticoxidation of p-nitrophenol. Biochem. Biophys. Res. Com-mun. 88:634-641.
9. Torstensson, N. T. L., J. Stark, and B. Goransson. 1975.The effect of repeated applications of 2,4-D and MCPA ontheir breakdown in soil. Weed Res. 15:159-164.
