Nitrite as a Stimulus for Ammonia-Starved Nitrosomonas europaea

Ammonia-starved cells of Nitrosomonas europaea are able to preservea high level of ammonia-oxidizing activity in the absence ofammonium. However, when the nitrite-oxidizing cells that formpart of the natural nitrifying community do not keep pace withthe ammonia-oxidizing cells, nitrite accumulates and may subsequentlyinhibit ammonia oxidation. The maintenance of a high ammonia-oxidizingcapacity during starvation is then nullified. In this studywe demonstrated that cells of N. europaea starved for ammoniawere not sensitive to nitrite, either when they were starvedin the presence of nitrite or when nitrite was supplied simultaneouslywith fresh ammonium. In the latter case, the initial ammonia-oxidizingactivity of starved cells was stimulated at least fivefold.

INTRODUCTION

Top

Introduction

Chemolithotrophic nitrifying bacteria play an important rolein the global nitrogen cycle by connecting the most reducedand oxidized parts of nitrogen. During the oxidation of ammoniumby ammonia-oxidizing bacteria, nitrite is formed as an end productthat is further oxidized to nitrate by nitrite-oxidizing bacteria.Under oxic conditions, chemolithotrophic ammonia-oxidizing bacteriaare entirely dependent on ammonium for the generation of energy.Under ammonium-limited conditions, the ammonia-oxidizing activityof cells of Nitrosomonas europaea is repressed by the presenceof ammonium-assimilating heterotrophic bacteria and/or plantroots (16). Ammonium oxidation is also repressed in the fieldduring the growing season of the wetland plant Glyceria maxima(2). Hence, ammonia-oxidizing bacteria have to withstand longerperiods of ammonium starvation. To be able to profit quicklyfrom the reappearance of fresh ammonium after a period of starvation,the population has to maintain a high ammonia-oxidizing potential.Pure cultures of ammonia-oxidizing bacteria are able to preservetheir ammonia-oxidizing activities for several weeks. For example,cultures of Nitrosomonas cryotolerans starved for ammonia wereable to maintain their cell densities and cellular compoundsand started ammonia oxidation immediately upon addition of ammonia(7-9). Also, ammonia-starved cells of N. europaea respondedimmediately to addition of fresh substrate (18). The maintenanceof a high ammonia-oxidizing capacity has also been demonstratedin mixed cultures of ammonia- and nitrite-oxidizing species(4, 14). In mixed cultures, however, repression of ammonia oxidationoccurred within the first few hours after addition of freshammonium to cells of N. europaea that had been incubated underammonium-limited conditions for several weeks (4). Since thenitrite-oxidizing cells of Nitrobacter winogradskyi were notable to keep pace with the ammonia-oxidizing cells, nitriteaccumulated after addition of fresh ammonium. It was hypothesizedthat the ammonia-oxidizing cells had been inhibited by the freshlyproduced nitrite. Stein and Arp (12) showed that the specificloss of ammonia-oxidizing activity in cells of N. europaea deprivedof ammonium was due to toxicity of accumulated nitrite in theincubation medium.

Effect of the presence of nitrite during starvation.

Top

Effect of the presence...

To study the potential inhibition of ammonia oxidation by nitritein cells that had been starved for ammonia for longer periods,cells of N. europaea strain ATCC 19781 were grown in nonshakenslurries containing 4 g of river sand in 20 ml of HEPES-bufferedmineral medium at 26°C in the dark. The mineral medium contained(per liter) 330 mg of (NH₄)₂SO₄, 585 mg of NaCl, 55 mg of KH₂PO₄,49 mg of MgSO₄ · 7H₂O, 147 mg of CaCl₂ · 2H₂O,75 mg of KCl, 10 g of HEPES, and 1 ml of trace elements solution,as described by Verhagen and Laanbroek (15). Bromothymol blue(3.2 mg/liter) was added from a 0.04% stock solution as a pHindicator. The pH was adjusted with 1 N NaOH to 7.7 before autoclaving.Before inoculation the sand slurries were autoclaved and cooledto room temperature. After 24 h, the slurries were decanted,fresh mineral medium was added, and the preparations were autoclavedagain. Altogether, three cycles of autoclaving, cooling, anddecanting were performed. Usually within 1 week after inoculationwith N. europaea the ammonium was converted to nitrite, thesand slurries were decanted, and fresh mineral medium was added.This subculturing was repeated three times. The use of sandcultures facilitated the axenic replacement of spent mediumby fresh medium without the loss of a large portion of the cells.On average, 8% of the medium remained in the culture after decanting.Samples used for determination of nitrite and nitrate contentswere centrifuged at 15,000 x g with an EBA 12 table centrifugefor 10 min. Griess-Ilosvay analyses of nitrite in the supernatantsof the cultures were performed immediately or within 24 h aftertransfer to -20°C.

Sand cultures of N. europaea that had been starved for 1, 2,and 3 months in their own spent medium (i.e., in the presenceof 5 mM nitrite) were decanted, and fresh mineral medium containing5 mM ammonium was added. The ammonium-starved cells startedto produce nitrite immediately (Table 1). During accumulationof nitrite, no repression of ammonia-oxidizing activity wasobserved during the first 24 h of incubation. A decline in ammonia-oxidizingactivity was occasionally observed during the second day ofincubation, which was likely due to ammonium limitation in theculture. In a parallel experiment with the same original batchof sand cultures, the spent medium was replaced by fresh mediumwithout inorganic nitrogen compounds at the start of the starvationperiod. After 1, 2, and 3 months of ammonium starvation in theabsence of nitrite, the cultures received fresh medium with5 mM ammonium. The potential ammonia-oxidizing activities werelow compared to those observed in the experiment described above(Table 1). Again, no inhibition of ammonia-oxidizing activitywas observed until ammonium had been exhausted after 6 daysof incubation. When we compared the potential ammonia-oxidizingactivities after starvation for ammonia in spent and fresh media,it appeared that 81 to 88% of the activity had been lost dueto starvation of the cells in fresh medium without inorganicnitrogen compounds. According to an analysis of variance (Statisticasoftware package, 99 edition; StatSoft Inc., Tulsa, Okla.),starvation time had no effect on the size of the ammonia-oxidizingpotential after starvation (P = 0.47161, n = 3), whereas thestarvation medium had an effect (P = 0.00003, n = 3). The presenceof nitrite in fresh starvation medium had no effect on the maintenanceof the ammonia-oxidizing potential (Table 2). Again, starvationin fresh medium had a significant (P < 0.05) negative effecton the level of the potential activity of starved cells. Underthese conditions, 76% of the potential activity was lost onaverage when the cells were starved in fresh medium. Lossesof cells during replacement of the medium could have been responsiblefor the differences in the levels of potential ammonia-oxidizingactivities in spent and fresh media. The numbers of free andattached cells in sand cultures were determined by flow cytometryin a separate experiment. The average total numbers of cellsin slurries prepared with cultures that had been starved inspent and fresh media and that were subsequently used to determinethe remaining ammonia-oxidizing potential were 1.62 x 10⁸ and1.34 x 10⁸, respectively. The difference in the number of cells(17%) is not large enough to explain the large differences inthe observed potential ammonia-oxidizing activities of cellsstarved in spent and fresh media. Hence, replacement of spentmedium by fresh medium at the onset of starvation seems to havediluted some unknown factor that had a positive effect on themaintenance of the ammonia-oxidizing potential. Such a factormight have been a homoserine lactone (1).

View this table:
[in this window]
[in a new window]
TABLE 1. Potential ammonia-oxidizing activities of starved cells of N. europaea measured for 4 h after addition of fresh mineral medium with 5 mM ammonium

View this table:
[in this window]
[in a new window]
TABLE 2. Potential ammonia-oxidizing activities of starved cells of N. europaea measured for 5 h after addition of fresh mineral medium with 5 mM ammonium

Effect of the presence of nitrite during potential activity measurements.

Top

After ammonium starvation for 4 months in spent medium containing5 mM nitrite, all sand cultures were supplied with fresh mediumcontaining 5 mM ammonium plus different combinations of nitrite,nitrate, and active nitrite-oxidizing cells of N. winogradskyistrain ATCC 25391, as indicated in Table 3. Nitrite and nitratecontents were determined by using high-pressure liquid chromatographyas described by Bollmann and Laanbroek (3).

View this table:
[in this window]
[in a new window]
TABLE 3. Potential ammonia-oxidizing activities of cells of N. europaea starved for ammonium for 4 months in spent medium

Sand cultures that had received 5 mM nitrite in addition to5 mM ammonium exhibited high potential ammonia-oxidizing activitiesfor 5 h irrespective of the presence of cells of N. winogradskyi(Table 3). Sand cultures that had received 5 mM nitrate alsoexhibited higher potential ammonia-oxidizing activities, althoughthese activities were significantly lower than those of thesand cultures incubated with 5 mM nitrite in the medium. Sandcultures that had received spent medium from N. winogradskyicultures exhibited the lowest potential activities during thefirst 5 h of incubation in the presence of ammonium. After theinitial 5 h the nitrite production rates in the different sandcultures were similar over the next 19 h (Fig. 1). The potentialammonia-oxidizing activities shown in Table 3 were determinedover the first 5 h of incubation after addition of fresh mediumwith ammonium. However, as Fig. 1 shows, the activities declinedagain after 3 h of incubation in the presence of 5 mM nitrite.Hence, the short-term effect of nitrite on the ammonia-oxidizingactivity of ammonia-starved cells was even more pronounced thanthe effect shown in Table 3. As calculated from the observednitrite and nitrate conversion rates, the nitrite-oxidizingactivity of the cells of N. winogradskyi was 179 nmol of nitriteh^-1 g (dry weight) of sand^-1, which is well above the potentialammonia-oxidizing capacity of starved cells of N. europaea,as shown in Table 3.

View larger version (15K):
[in this window]
[in a new window]
FIG. 1. Ammonia oxidation by cells of N. europaea starved for ammonium for 4 months in spent medium containing 5 mM nitrite. After starvation all sand cultures were decanted and supplied with fresh mineral medium containing 5 mM ammonium. Some cultures received in addition 5 mM nitrite ( ${blacksquare}$ ), 5 mM nitrite plus actively nitrite-oxidizing cells of N. winogradskyi (•), spent medium from an N. winogradskyi culture ( ${circ}$ ), or 5 mM nitrate ( ${blacktriangleup}$ ). Some cultures were supplied with only 5 mM ammonium and served as controls ( ${blacklozenge}$ ). The values are means based on the results obtained with five replicate cultures, and 95% confidence levels are indicated.

The assumed toxicity of nitrite for ammonia-starved cells ofN. europaea was not observed in the experiments described above,either when nitrite was present in the starvation medium orwhen it was present in the resuscitation medium. On the contrary,nitrite stimulated the ammonia-oxidizing activity of starvedcells during the first few hours after addition of fresh mediumwith 5 mM ammonium. Stimulation of ammonia oxidation by nitriteis difficult to explain based on the two-step mechanism of ammoniaoxidation in which hydroxylamine is an intermediate product,as has been described for these chemolithotrophic bacteria (6).It is hard to imagine how the end product of the second step(nitrite) could stimulate the oxidation of ammonia. However,nitrite can be reduced to NO by the enzyme nitrite reductase.Nitrite reduction to NO has been shown to occur in N. europaea(5, 10, 11). Recently, Whittaker et al. (17) showed that theNirK component of the nitrite reductase system was present inN. europaea under oxic conditions. Also, the gene for NO reductasewas present in the genome of this organism, but the genes forN₂O and nitrate reductases were missing (17). Zart et al. (19)demonstrated that NO indirectly stimulates aerobic ammoniumoxidation by Nitrosomonas eutropha. It was hypothesized thatNO provides the cells with NO₂ (or N₂O₄), which could act asthe cosubstrate for the enzyme ammonia monooxygenase in thefirst step of chemolithotrophic ammonia oxidation. The postulatedrole of NO in the stimulation of aerobic ammonia oxidation couldalso be valid for the observed short-term stimulation of ammoniaoxidation by nitrite in ammonia-starved cells in the experimentsdescribed here. The presence of nitrite during starvation apparentlyhad no effect on the preservation of the ammonia-oxidizing potential.

The observed stimulation of ammonia oxidation by nitrite inN. europaea contradicts the results obtained by Stein and Arp(13) with cells of N. europaea. The most obvious explanationfor the difference is the physiological condition of the cellsin the two studies. Our observations were made with ammonia-starvedcells, whereas Stein and Arp (13) used late-log-phase cellsfor their inhibition experiments.

Is stimulation of ammonia oxidation in ammonia-starved cellsby nitrite also feasible in the natural environment? Due totheir lack of competitive ability with respect to ammonium uptake,cells of N. europaea have to deal with periods of ammonium limitationor starvation (2, 16). Under oxic conditions, ammonium starvationalso leads to nitrite starvation for the nitrite-oxidizing bacteria.When ammonium reappears in an oxic environment, no nitrite isavailable to be used as a stimulus for ammonia oxidation bystarved cells. Hence, the results obtained with monoculturesshould be interpreted with care from an ecological point ofview, but they may be helpful in understanding the metabolismof ammonia-oxidizing bacteria, at least those belonging to theN. europaea-N. eutropha group.

ACKNOWLEDGMENTS

We thank Bertus Beaumont, Annette Bollmann, Paul Bodelier, andIngo Schmidt for valuable discussions. Paul Bodelier is alsoacknowledged for performing the statistical analyses.

FOOTNOTES

* Corresponding author. Mailing address: NIOO-KNAW Centre for Limnology, P.O. Box 1299, 3600 BG Maarssen, The Netherlands. Phone: 31 294 239336. Fax: 31 294 232224. E-mail: laanbroek{at}cl.nioo.knaw.nl.

Publication 2893 of the NIOO-KNAW Centre for Limnology, Nieuwersluis,The Netherlands.

REFERENCES

Top