Dechlorination of lindane by the cyanobacterium Anabaena sp. strain PCC7120 depends on the function of the nir operon

Nitrate is essential for lindane dechlorination by the cyanobacteria Anabaena sp. strain PCC7120 and Nostoc ellipsosporum, as it is for dechlorination of other organic compounds by heterotrophic microorganisms. Based on analyses of mutants and effects of environmental factors, we conclude that lindane dechlorination by Anabaena sp. requires a functional nir operon that encodes the enzymes for nitrate utilization.

Cyanobacteria are photoautotrophic microorganisms common to a variety of environments including polluted ones. Earlier, we reported that two filamentous nitrogen-fixing cyanobacteria, Anabaena sp. strain PCC7120 and Nostoc ellipsosporum transformed lindane (14) first to ␥-pentachlorocyclohexene and then to a mixture of chlorobenzenes ( Fig. 1). This process was cometabolic and depended on the presence of nitrate (14).
Nitrate-dependent dehalogenation of organic compounds by different heterotrophic bacteria has been described in the past (1,9,21), but no mechanism for this process or link to genetic systems has been proposed. For some microorganisms dehalogenation was coupled with denitrification (9). In both cyanobacteria and anaerobic denitrifying microorganisms, nitrate uptake and reduction are initial processes of nitrate utilization, and the genes for these processes are organized in similar operons (4,17,19,24,26). At the level of nitrite reduction, metabolic pathways diverge and lead to the assimilatory chain for cyanobacteria (8), algae (3), fungi (12), and plants (27) and the dissimilatory chain for heterotrophic anaerobic microorganisms (28). In Anabaena sp. strain PCC7120, genes organized in an operon as 5Ј-nirA-nrtABCD-narB-3Ј encode nitrite reductase, nitrate transport proteins, and nitrate reductase (2,7), similar to other cyanobacteria (17,23).
We report that Anabaena sp. nirA, nrtC, nrtD, or narB mutants cannot dechlorinate lindane in the presence of nitrate. Dechlorination is also inhibited in the dark and in the presence of ammonium, both of which are environmental inhibitors of the function(s) encoded by the nir operon. Fifteen strains of wild-type cyanobacteria screened by us degraded lindane.
Effect of mutations in the nir operon on lindane dechlorination by an Anabaena sp. We analyzed lindane degradation by Anabaena sp. transpositional mutants TLN10 (insertion in the nirA gene), TLN12 (insertion in the nrtC gene), TLN21 (insertion at the 3Ј end of the nrtD gene), and DR796 (site-directed interposition in the narB gene), which have been described in detail by Cai and Wolk (2). Growth of the cultures at 28°C was monitored by measuring chlorophyll content. Experimental procedures were as described by Kuritz and Wolk (14). Addition of lindane to the cultures to a final concentration of 0.5 mg/liter allowed us to monitor the kinetics of the disappearance of lindane associated with the cells. To measure concentrations of cell-associated lindane, 2 ml of each culture was sampled, washed twice with 2 ml of sterile water, resuspended in 1 ml of sterile water, subjected to sonication in an ice water bath for 15 min, and extracted with 1 ml of hexane. Gas chromatography was performed as described by Kuritz and Wolk (14), except for the temperature gradient which was set up as linear from 100 to 190°C over 10 min and from 190 to 240°C over 4 min. Lindane retention time was 9.82 to 9.84 min.
Anabaena sp. mutants TLN10, TLN12, TLN21, and DR796 did not degrade lindane in the presence of nitrate in contrast to the wild-type culture ( Fig. 2; TLN21 is not shown), which result suggested the involvement of the nir operon in this process. Since lindane was associated with the cells in all mutants, it is unlikely that the products of the operon are involved in lindane uptake, though we know of nothing in the literature that suggests lindane uptake mechanisms. The presence of lindane did not affect the growth rate of the wild type or mutant cultures.
Effect of darkness or ammonium on lindane degradation. Wild-type cultures supplemented with 15 ng of lindane/liter dechlorinated the lindane in the presence of 5 mM nitrate only when illuminated; addition of 1 mM (NH 4 ) 2 SO 4 to nitratesupplemented wild-type cultures completely inhibited dechlorination (Fig. 3). Since ammonium and darkness are environmental inhibitors of the functions of the nir operon (6,8,15,16,19), these results also suggest a role for the products of the operon in lindane dechlorination.
Only the production of pentachlorocyclohexene is nitrate dependent. When supplied with the linA gene, the product of which dechlorinates lindane to ␥-pentachlorocyclohexene (11). Anabaena dechlorinated lindane to trichlorobenzenes in the absence of nitrate (14). We found no sequence homology to linA upon Southern hybridization of Anabaena DNA with an excess of this gene. Cyanobacteria and Pseudomonas paucimobidis may employ heterologous systems for this transformation.
Nitrate uptake and nitrite reductase activities do not correlate with lindane degradation. Nitrate reductase activity was measured as described by Herrero et al. (10) with modifications that included use of cell suspensions with chlorophyll content of ca. 10 to 15 g ml Ϫ1 and incubation at 28°C. Nitrate reductase activity was found in the wild-type cultures (21.3 Ϯ 9.9 pmol of NO 2 Ϫ ⅐ g of chlorophyll Ϫ1 ⅐ min Ϫ1 ) and in nrtC (13.0 Ϯ 3.8 pmol of NO 2 Ϫ ⅐ g of chlorophyll Ϫ1 ⅐ min Ϫ1 ) and nrtD (11.4 Ϯ 2.3 pmol of NO 2 Ϫ ⅐ g of chlorophyll Ϫ1 ⅐ min Ϫ1 ) mutant cultures. Nitrate reductase activity was present in the nirA mutant cultures, but its values were inconsistent through an independent series of experiments, possibly due to differences in the physiological state of the cultures that were sick on nitrate. Only the narB mutant had no nitrate reductase activity. An absence of the polar effect of the nirA, nrtC, and nrtD mutations (all of which are transpositional) on nitrate reductase activity may be due to either the presence of the weak promoter at the IS50R of the transposon (Cai and Wolk [2]) or the existence of a nitrate-regulated promoter upstream to the narB gene, or both.
Nitrate uptake experiments were carried out as described by Flores et al. (6). The specific rate of nitrate uptake was 35.7 Ϯ 12.0 nmol of NO 3 Ϫ ⅐ g of chlorophyll Ϫ1 ⅐ min Ϫ1 for the wild type and 30.6 Ϯ 13.5 nmol of NO 3 Ϫ ⅐ g of chlorophyll Ϫ1 ⅐ min Ϫ1 for the nirA mutant. The nrtC, nrtD, and narB mutant cultures did not evince nitrate uptake. Nitrate uptake inhibition by the mutation in the narB gene may be explained if nitrite is a transcriptional regulator of the nir operon, as was suggested by Kikuchi et al. (13). Cyanobacterial cultures supplemented with ammonium did not take up nitrate and had no nitrate reductase activity, which results agree with the results presented earlier by Flores et al. (5) and Omata et al. (18).
The amount of nitrite accumulated in nitrate-supplemented medium under CO 2 -limited conditions as reported by Suzuki et al. (25) served as a measure of the difference between the activities of nitrate and nitrite reductases. Concentrations of nitrite were determined as described in Snell and Snell (22) after the transfer of the cultures into fresh medium containing 100 M nitrate. The values for specific accumulation of nitrite (Fig. 4) were much higher for the TLN10 mutant (202.0 Ϯ 34.9 pmol of NO 2 Ϫ ⅐ mol of chlorophyll Ϫ1 ⅐ min Ϫ1 ), which took up nitrate and reduced nitrate but was deficient in nitrate reductase, than for the wild-type cultures (12.7 Ϯ 0.5 pmol of NO 2 Ϫ ⅐ g of chlorophyll Ϫ1 ⅐ min Ϫ1 ). TLN12 did not accumu- late nitrite probably due to its inability to take up nitrate. TLN21 accumulated nitrite at a lower rate than the wild-type culture (1.3 Ϯ 0.6 pmol of NO 2 Ϫ ⅐ g of chlorophyll Ϫ1 ⅐ min Ϫ1 ), which suggests that either the product of this gene is less essential for the function or the transposition did not completely inactivate the product. DR796 did not accumulate nitrate. We saw no direct correlation of lindane degradation with any individual enzyme activity encoded by the nir operon.
Present knowledge of the regulation of the nir operon-encoded pathways does not allow us to reach conclusions as to the mode of involvement of the products of this operon in lindane transformation. Our results illustrate the lack of coordinated regulation of the genes in the nir operon suggested by two other research groups (2,13). Nitrate reductase is an essential, although not sufficient, enzyme in what may be a fortuitous system for lindane degradation. One can speculate that lindane dechlorination may be a function of a membrane nitrate-nitrite reduction complex.
Lindane degradation by other cyanobacteria. The 15 strains of cyanobacteria attributed to three taxonomic groups by Rippka et al. (20) were able to degrade lindane although with different efficiencies (Fig. 5). In the past, the search for biological agents that degrade lindane and other organic xenobiotics involved only soil heterotrophic bacteria. Our observations show that cyanobacteria represent a largely unstudied resource in the search for biodegradative systems.