Removal of Cadmium by Microorganisms in a Two-Stage Chemostat

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, May 1984, p. 1158-1160 0099-2240/84/051158-03$02.00/0

Vol. 47, No. 5

Removal of Cadmium by Microorganisms in

a

Two-Stage Chemostat

C. HOUBA* AND J. REMACLE University of Liege, Institute of Botany, Microbial Ecology, Sart Tilman, B4000 Liege, Belgium Received 18 October 1983/Accepted 14 February 1984

A two-stage chemostat was used to study removal of cadmium by microorganisms in continuous culture. The medium was contaminated with 0.8 mg of Cd per liter. At 20°C, most of the microbial biomass formed aggregates which settled in the second stage of the chemostat. Effluent was free of bacteria. Up to 80% of the metal contained in the inlet flux was removed by the biomass, with 20% remaining in solution. At 10°C and with a shorter retention time, flocculation was poorer and metal removal by settling biomass did not exceed 35%.

Bacteria growing in the first stage of the chemostat presented no biotlocculation ability. Therefore, the turbidity of the effluent from the first stage remained high throughout the experiment. The turbidity of the effluent of the second stage decreased as the activated sludge developed (Fig. 2). An equilibrium seemed to be reached after day 5 and the values of the parameters reported in this paper were measured every 2 days after that time. At 20°C, good flocculation occurred in the second stage, and the biomass was almost entirely recovered by decantation of the effluent (Table 2). Lowering the temperature to 10°C and reducing the retention time to 3 days instead of 7 considerably hindered the flocculation ability of the culture. The biomass remained chiefly in suspension. COD values reported in Table 2 were measured after decantation (at 20°C) or after centrifugation (at 10°C) of the effluent of the second stage according to the French legal procedure (AFNOR, T90-101). When good culture conditions were achieved, i.e., at 20°C and with sufficient retention time, COD reduction reached 90%, since the initial COD content of the culture medium was 700 mg of C per liter. Concentrations of cadmium were measured every 2 days for 2 weeks in free bacteria growing in Cl and in aggregates formed in C2. Direct flame spectrophotometry was used after digestion of the material in a hot mixture of nitricperchloric (vol/vol) concentrated acid (Table 3). Differences due to growth temperature were not significant (P < 0.05; t test). Therefore, means were calculated irrespective of this parameter: 1,012 mg of Cd per kg dry weight in free bacteria and 2,160 mg of Cd per kg dry weight in aggregates. Aggregates thus accumulated twice as much metal as free bacteria. This difference was significant (P < 0.05; t test). The concentrations observed in bacteria forming a film on the internal walls of the chemostat were approximately the same as those for aggregates. Formation of aggregates and adhesion of bacteria to the walls of culture vessels depend on

The effect of cadmium on freshwater bacterial communities grown in continuous culture has been previously reported (12, 15). The presence of 0.5 mg of Cd per liter in the culture medium does not hinder bacterial development or biodegradation processes. Studies conducted with a treatment plant on a pilot scale lead to similar conclusion (2, 17). Microorganisms are able to concentrate heavy metals to a large extent (1, 3, 5, 9, 13). Several authors consider microorganisms as natural "biosorbants" (7, 10, 16). The present study was conducted in continuous culture. To allow natural interactions between microorganisms, plurispecific microbial communities were used as inoculum (2). The chemostat used for this purpose consisted of two parts (Fig. 1). The first stage was inoculated with water of the unpolluted Ourthe river collected near Liege (Belgium) and filtrated on an 8-p.m membrane filter (Millipore Corp.) (12). The microbial community in this stage consisted chiefly of heterotrophic bacteria (Pseudomonas spp. and Flavobacterium spp.). Oxyferm tubes (Roche) were used to perform this identification. Bacterial cells either remained freely in suspension (they are therefore referred to as free bacteria) or formed a biofilm adhering to the internal wall of culture vessel. The second stage was inoculated with activated sludge obtained from a municipal treatment plant at Tienen (Province Brabant, Belgium). The microbial community was mainly bacteria and ciliate protozoa (Opercularia spp. and Vorticella spp.). Bacteria formed aggregates and flocculation occurred. The composition of the culture medium (in grams per liter) was as follows: Bacto-tryptone (Difco Laboratories), 1; glucose, 0.5; salt solution, MgSO4, 0.125; NaCl, 0.125; Fe2(SO4)3, 0.0025; MnSO4, 0.0025. It was contaminated by addition of cadmium sulfate (Suprapure; Merck) to a final concentration of 0.8 mg/liter. The two vessels, C1 (free bacteria) and C2 (floc-forming bacteria), had different volumes (V): 500 ml and 5 liters, respectively. The effluent of C1 (E1) flowed directly and quantitatively into C2, so that fluxes (F) at the entrance of both stages were the same. Dilution rates (D = FIV [hours-]) or retention times (E = 1D [hours or days]) were different. Values of D and 0 for each run are presented in Table 1. Assays were conducted in duplicate for 2 weeks at 10 and 20°C. *

TABLE 1. D and 0 for the two stages of the chemostat in each assay

Corresponding author.

1158

Culture

Assay

temp

no.

10°C

1

200C

2 1 2

Second stage

First stage D 0 (h) (h- 1)

(h-1)

(days)

0.09 0.09 0.32 0.26

0.014 0.009 0.006 0.007

3 5 7 6

11.1 11.1 3.1 3.8

NOTES

VOL. 47, 1984

1159

TABLE 2. Description of the effluents from the second stage of the chemostats Culture Assay no. temp

10°C

El

200C

A

FIG. 1. Description of the two-stage chemostat. Abbreviations: M, sterile culture medium contaminated with cadmium; P, peristaltic pump; C1 and C2, culture vessels, first and second stages, respectively; El and E2, effluent of the first and of the second stage; A, forced sterile air.

the production of extracellular polymers (4, 8). At neutral pH, they are negatively charged and provide additional fixation sites for metal cations. This property might explain the higher concentration of cadmium observed in fixed or aggregated cells. More detailed experiments are needed, however, to understand the process of accumulation thoroughly. Cadmium budgets for each assay were calculated (Table 4). The amount of metal in each compartment of the system was expressed as a percentage of the total quantity introduced in the course of the experiment. In the effluent of the first stage, 55 to 89% of the total input was still in soluble

COD (mg

Free

iter)

(mg/liter)

Aggregates

(mg/liter)

218 + 150 156 ± 23 248 ± 106 138 ± 20 71 ± 30 1,202 ± 577 59 ± 18 2,226 ± 1,136

1 2 1 2

Tubdt Turbidity

0.58 ± 0.12 0.75 ± 0.05 0.32 + 0.15 0.14 ± 0.04

form, and 17 to 29% was being fixed by free bacteria. The biomass remaining in suspension flowed into the second stage of the chemostat, together with soluble metal. Additionally cadmium was removed by bacteria adhering to the walls of the first-stage vessel. The accumulation ability of that material is probably to a large extent controlled by the production of extracellular polymers, and therefore the thickness of the biofilm is an important factor. The biofilm is continuously formed and sloughed off. This might explain the large variations observed regarding that material (2 to 23% of the cadmium budget). In the effluent of the second stage, the proportion of soluble cadmium was reduced to 10 to 33%. The distribution of particulate metal depended on the bioflocculation ability of the culture. At 10°C and 0 = 3 to 5 days, 45 to 60% of the cadmium was associated with the microorganisms remaining in suspension and only 18 to 26% was removed with aggregates. At 20°C and 0 = 6 to 7 days, all the particulate metal consisted of aggregates (42 to 62%). To check the accuracy of the budgets the sums S, for the first stage and S2 for the second (see Table 4) should be equal to 100%; deviations can be attributed to analytical errors. The distribution of particulate metal is important regarding its fate since only aggregates can settle rapidly. Bioaccumulation is apt to occur in actual environment too (6, 11). Aggregates settling to the bottom of the rivers probably participate in the local transfer of metal from water to sediments, whereas free bacteria are transported further downstream. The results reported here also suggest the use of biological process in the case of actual industrial effluent (14). Financial support was generously provided by CEC and IRSIA (grant 76287). We thank N. Wilbert (Institute of Zoology, Bonn University, Federal Republic of Germany) for the identification of protozoa and to J. Emmerton (Ruhr University, Bochum, Federal Republic of Germany) for her help in the preparation of this manuscript.

1004

TABLE 3. Cadmium concentration in free and fixed bacteria and in aggregates

\

50

,

,I 5 ,

Assay

Free bacteria

no.

(C1)

10°C

1 2 1 2

948 ± 56 964 ± 201 1,205 + 303 1,000 ± 167

x

X L

Culture temp

,

,

,

*

10

15

DAYS

FIG. 2. Evolution of the turbidity of the effluent in the second stage of the chemostat (E2) as a function of time. Turbidity (optical density at 420 nm) at the beginning of the assay was considered to be 100%. Symbols: *, culture 2 at 10°C; X, culture 2 at 20°C.

200C Mean

1,012 ± 247

Cadmium concn (mg of Cd per kg [dry wt]) in: Bacteria attached to Aggregates the walls of (C2) vessel (CI)

1,595 ± 272

2,193 2,321 (8,288) 2,113

2,209

+

2,456 ± 653 2,227 ± 752 2,486 ± 720 105

2,160 ± 833

1160

APPL. ENVIRON. MICROBIOL.

NOTES TABLE 4. Cadmium budget

Cadmium distribution (% of total amt of metal introduced into the chemostat during each assay) in:

Assay no.

Bacteria fixed to the wall of C1 (a) 9 10 23 2

Soo in Microorganisms still in suspension in E2 E2 (f) after decantation (e) 10°C 1 18 89 26 45 19 116 2 17 77.5 18 60 10 104.5 200C +0 1 29 35 42 18 107 2 +0 23 74 62 33 99 a To check the accuracy of the budgets, the sums S5 = (a) + (b) + (c) for the first stage and S2 = (a) + (d) + (e) + (f) for the second should be equal to 100%. Differences are due to analytical errors. Culture

Free bacteria

Solution in

Aggregates

of C1 (b)

El (c)

(d)

LITERATURE CITED 1. Babich, H., and G. Stotzky. 1978. Effects of cadmium on the biota: influence of environmental factors. Adv. Appl. Microbiol. 23:55-117. 2. Bagby, M. M., and J. H. Sherrard. 1981. Combined effects of cadmium and nickel on the activated sludge process. J. Water Pollut. Control Fed. 53:1609-1619. 3. Beveridge, T. J., and S. F. Koval. 1981. Binding of metals to cell envelopes of Escherichia coli K-12. Appl. Environ. Microbiol. 42:325-335. 4. Brown, M. J., and J. N. Lester. 979. Metal removal in the activated sludge: the role of bacterial extracellular polymers. Water Res. 13:817-837. 5. Chopra, I. 1971. Decreased uptake of cadmium by a resistant strain of Staphylococcus aureus. J. Gen. Microbiol. 63:265-267. 6. Houba, C., and J. Remacle. 1981. The role of microorganisms in the cadmium fate in freshwater ecosystems, p. 246-249. In W. H. 0. Ernst (ed.), Proceedings of the 5th International Conference on Heavy Metals in the Environment. Amsterdam. 7. Kurek, E., J. Czaban, and J. M. Bollag. 1982. Sorption of cadmium by microorganisms in competition with other soil constituents. Appl. Environ. Microbiol. 43:1011-1015. 8. Marshall, K. C. 1976. Interfaces in microbial ecology. Harvard University Press, Cambridge, Mass.

S2a

99 98 83 97 stage

9. Matthews, T. H., R. J. Doyle, and U. N. Streips. 1979. Contribution of peptidoglycan to the binding of metal ions by the cell wall of Bacillus subtilis. Curr. Microbiol. 3:51-53. 10. Nakajima, A., T. Horikoski, and T. Sakaguchi. 1981. Studies on the accumulation of heavy metal elements in biological systems. XVII. Selective accumulation of heavy metal ions by Chlorella regularis. Eur. J. Appl. Microbiol. 12:76-83. 11. Patrick, F. M., and M. W. Loutit. 1977. The uptake of heavy metals by epiphytic bacteria on Alisma plantago-aquatica. Water Res. 11:699-703. 12. Remacle, J. 1980. Cadmium uptake by freshwater bacterial communities. Water Res. 15:67-71. 13. Remacle, J., and C. Houba. 1980. The influence of cadmium upon freshwater saprophytic bacteria. Environ. Technol. Lett. 1:193-200. 14. Remacle, J., and C. Houba. 1983. The removal of heavy metals from industrial effluents in a biological fluidised bed. Environ. Technol. Lett. 4:53-58. 15. Remacle, J., C. Houba, and J. Ninane. 1982. Cadmium fate in bacterial microcosms. Water Air Soil Pollut. 18:455-465. 16. Tsezos, M., and B. Volesky. 1981. Biosorption of uranium and thorium. Biotechnol. Bioeng. 23:583-604. 17. Weber, A. S., and J. H. Sherrard. 1980. Effects of cadmium on the completely mixed activated sludge process. J. Water Pollut. Control. Fed. 52:2378-2388.