Effects of Tubifex Tubifex (Oligochaeta: Tubificidae) on N-Mineralization in Freshwater Sediments, Measured with 15~Isotopes

AQUATIC MICROBIAL ECOLOGY Vol. 9: 289-294, 1995 Published December 21 Aquat microb Ecol

Effects of Tubifex tubifex (Oligochaeta: Tubificidae) on N-mineralization in freshwater sediments, measured with 15~isotopes

Silvia P. Pelegri*, T. Henry Blackburn

Department of Microbial Ecology, Institute of Biological Sciences, University of Aarhus, Ny Munkegade Bd. 540, DK-8000 Aarhus C, Denmark

ABSTRACT Sediment cores conta~ningdifferent dens~tiesof Tubifex tublfex ranging from 0 to 70000 ind m-' were incubated in the laboratory Rates of O2 and NO, uptake NH,' production n~tr~fication and denitnfication were determined from sediment-water fluxes Pore water NH,+was measured at the end of the expenment At natural densities -50000 ind m ', there were increased rates of O2 consumption (x2),denitnfication of water phase NO; (x3) and NH,' efflux (x26) Nitrification was stimulated at low worm densities, but inhibited at higher worm densities The transport of reduced compounds and organic matter, with the fecal pellets, to the sediment surface stimulated anoxic cond~tions in the inhabited microcosms These anoxlc cond~tionsled to increased rates of denitnfication and were responsible for the decrease in nitnfication at h~gherworm densities Approximately 25% of the NO? produced by nitnfication within the sedlment was subsequently den~trlfied Denitnfication was respon- slble for 25% of the NO, disappearance from the system The h~gherrates of denitrification were counterbalanced by higher rates of NH,' flux from the sediment It is likely however that the presence of T tubifex resulted in a net loss of nitrogen that could otherwise have been used by the primary producers

KEYWORDS Bioturbation . Nitrification Denitnfication Sed~ments

INTRODUCTION Tevesz 1982). However, their role in nutrient cycling, specifically nitrogen cycling, has received less atten- Macrobenthos, through their burrowing, feeding, tion (Kikuchi & Kurihara 1977, 1982, Chaterpaul et al. locomotive, respiratory and excretory activities, play 1979, 1980). an important role in mediating both physical and Tubificid worms feed primarily in the top 2 to 8 cm of chemical processes near the sediment-water interface sediment (McCall & Tevesz 1982), adjusting their feed- (Fisher 1982, McCall & Tevesz 1982). Tubificid worms ing depth to lower strata at higher worm densities are often, along with chironomid larvae, the most (Robbins et al. 1979, McCall & Fisher 1980). Tubificid abundant macrofauna species in eutrophied stream worms are 'conveyor belt' feeders (Rhoads 1974). They and lake sediments. Population densities can be as typically live and feed head down in the sediment. high as millions of individuals per m2 (Palmer 1968). Some portion of the posterior of the worms may project Many studies concerning the effect of these worms on above the sediment-water interface. The worms selec- the physical properties of the sediment have been car- tively ingest silt and clay particles at depth and digest ried out (McCall & Fisher 1980, Fisher 1982, McCall & the attached microflora, primarily bacteria (Davis 1974). Fecal pellets are deposited at the sediment- water interface, where they may form a pelletized layer. 'Present address: Department of Biology, University of South- western Louisiana, PO Box 42451, Lafayette, Louisiana Studies of the effect of bioturbation by marine 70504-2451, USA E-mail. sxp09270usl.edu infauna on nitrification and denitrification processes Aquat microb Ecol 9:289-294, 1995

(Kristensen et al. 1991, Pelegri et al. 1994, Pelegri & again following the same procedure as described Blackburn 1995) have shown that the particular feed- above. Two days later a final incubation took place ing and burrowing strategies of the various animals (reservoir water containing 62.1 pM NO3-, 81'5) can lead to variable stimulation of aerobic respiration, I5NO3-). nitrification and denitrification. The aim of this study is At the end of the experiment the cores were cut into to clarify the effect of tubiflcid worms on nitrogen 0-0.5, 0.5-1, 1-2, 2-3 and 3-5 cm sections. Pore water metabolism and on fluxes across the sediment-water was obtained by centrifuging the sediment for 10 min, interface. at 3000 rpm (2000 X g) in double centrifuge tubes (Blackburn 1980). The supernatant was immediately collected and frozen for later analysis. Unfortunately, MATERIAL AND METHODS due to the small volumes of pore water, the analyses were run on single samples. Undisturbed sediment cores (19.5 cm long and Analysis. Oxygen was measured directly in the 3.5 cm inner diameter), containing no visible macro- microcosms with an oxygen microsensor provided with fauna, were collected in Giber A stream (Denmark). a guard cathode (Revsbech et al. 1989). NO3' + NO2- This stream is dominated by runoff from agricultural was measured by standard methods (Grasshoff et al. areas and receives fluctuating amounts of NO3- and 1983) in a flow injection analyzer (Tecator, Hoganas, organic matter (Nielsen et al. 1990). Tubifex tubifex Sweden) and analyses of NH4+were done by the sali- Miiller (Oligochaeta: Tubificidae), collected from cylate hypochlorite method (Bower & Holm-Hansen sieved (500 pm) Universitetsparken Lake sediment, 1980). The formation of 15N labeled dinitrogen pairs were added to these cores (n = 11) at densities ranging (I4Nl5Nand 15Ni5N)by denitrification was measured from 0 to 70 ind. core-' (70000 ind. m-2). The worms on an isotope ratio mass spectrometer as described by dug immediately into the sediment. The cores were Nielsen (1992). The concentration of the dinitrogen placed in a reservoir containing well-aerated tap pairs in the samples was calculated by multiplying water, and kept in darkness at 10.8 * 0.8"C. An atmos- their ratio (14N15Nor 15N15N/totalN) by the N, concen- pheric NH4+ trap, consisting of 3 gas washing bottles tration In the water (Weiss 1970). The %I5NO3- enrich- containing 0.1 M H2S04,50 mM K2HP0, and distilled ment was analyzed on a mass spectrometer after using water connected in line, was installed in the aerating denitrifying bacteria cultures to transform NO3- to circuit between the air pump outlet and the reservoir. gaseous NZ (Risgaard-Petersen et al. 1993). Ammonium concentrations in the reservoir water Calculations. Rates of denitrification were measured ranged from 0.3 to 1.1 PM during the experiment. using the isotope palring technique (Nielsen 1992). Experimental procedure. Every second day 02, NH,+ The formation rates of single- (I4Nl5N) and double- and NO3' + NOz- fluxes were measured in order to labeled (15NlSN)dinitrogen pairs were used to calcu- determine when steady state was reached. This was late the denitrification of NO3- coming from the overly- done by closing the cores for 2 h, after which samples ing water (dw) and of NO3- generated within the from the overlying water were collected and immedi- sediment by nitrification (dn): ately frozen. Eleven days after setup, when steady state was reached, 15N03- (5 mM K15N03-, 99% 15N stock dI5 = (14NL5N)+ 2(15NI5N) (Koike & Hattori 1978); solution) was added to the reservoir water to a final concentration of 38.3 pM NO, (69.7 ''NO3-). After 2 d, the cores were stoppered and incubated for 4 to 5 h, which resulted in less than 20% O2 depletion. At the end of the incubation the stoppers 7.~1ereremoved and water sam- (Nielsen 1992) ples were taken and stored. Samples for analysis of dissolved NZ gas were stored In 6 m1 glass vials (Exetainer, where e15is the 0/0l5NO3-enrichment of the reservoir Labco Ltd, Buclunghamshire, UK), containing 2% ZnC1, water. in order to stop microbial activity. Nitrogen gas was Nitrification was calculated by a modification of the extracted from these water samples with 0.9 m1 helium, equation described by Blackburn (1993) for '5~~4+ and analyzed by mass spectrometry. Samples for ana- turnover in sediments: lyzing NH,', NO3- + NO,- and %I5NO,- enrichment were collected in plastic vials and immediately frozen. In order to test the applicability of the denitrification assay in our bioturbated sediment, the water in the reservoir was exchanged with water containing 73 VM where d is the nitrification rate, i is the NO3- consump- NO3- (83% 15N03-) and the cores were incubated tion rate, e is the %15N03- enrichment, c is the NO3- Pelegri & Blackburn: Effects of Tubjfextubifex on N-m~neralization

concentration in the water phase (0 = initial, t = final) The isotope pairing technique assumes that the and 0.003663 is the natural abundance of I5N in nature. increase in concentration in the water by the addition The rates of dw and NO3- uptake are dependent on of I5NO3- does not influence denitrification of the sed- the NO3- concentration in the water column, which iment-produced I4NO:; and that the NO; species was different in the various incubations. These rates (''No3- and 14N03-) are uniformly mixed in the deni- were normalized to 50 pM NO; concentration in the trification zone. Since the production of unlabeled N2 overlying water for all incubations. ('4N'4N)is not measured directly but estimated from the production of single- (l4Nl5N)and double-labeled (15N'5N) dinitrogen pairs, changes in the ratio of dw RESULTS (denitrification of NO3- coming from the overlying water) to dn (denitrification of NO3- generated within Oxygen consumption (Fig. lA), NH4+fluxes (Fig. 1B) the sediment oxic layer) due to bioturbation may and NO3- consumption (Fig 1C) were stimulated by underestimate the dn values. This can be tested by the presence of the worms. incubating with different concentrations of I5~O3-. Thus at higher ''NO3- levels more ''No3- will be paired with ''NO3- to form measurable 14N15N and less I4Nl4N is formed. Rates of dn, measured with NO3- concentrations in the overlying water ranging from 38.3 to 73 I-IM, were not significantly different (p < 0.001). Thus the addition of 38.3 pM NO3- to the overlying water was sufficient to avoid underestima- tion of dn rates. Rates of nitrification and dn (Fig. 2A) increased significantly (p < 0.05) by increasing the worm density from 0 to 20000 ind. m-'. However, nitrification rates decreased significantly (p < 0.05) when worm density increased from 20000 to 70000 ind. m-2. Around 25% of the total NH,+ nitrified was subsequently denitrified in the suboxic sediment layer. Denitrification of NO3- coming from the overlying water (dw) was stimulated by the presence of the worms (Fig. 2B). An avel-age of 25% of the NO3- uptake by the sediment was removed from the system by denitrification. Rates of dw represented from 20 to 90 0/0 of the total denitrification rates in microcosms containing worm densities ranging from 0 to 70000 ind. m-2. Ammonium concentrations increased with depth, but the increase, compared with the control, was less with < 20 000 ind. m-' and greater with >20 000 ind. m-2 (Fig. 3).

DISCUSSION

In our experiments, oxygen uptake was increased 3-fold at a tubificid density of 70000 ind. m-', whereas McCall & Fisher (1980) reported a 2-fold increase at

Animal density (10 ind. 100000 ind. m-2, 20 % of which was due to worm respiration. Our results indicated that the worms (20000 Fig. 1. Tubifex tubifex. (A) Oxygen uptake (mm01 m-2 h-'), ind. m-2) increased denitrification of NO, coming from (B)ammonium effluxes (pm01 n2r2 h l) and (C) nitrate uptake the overlying water by 175%. Chaterpaul et al. (1980) (pmol m-2 h-') corrected to 50 pM NO3- in the water phase, reported that Limnodrilus hoffmeisteri and Tubifex versus tubificid worm density. Each data point is the average of 6 replicates, except for microcosms containing 10000 ind. tubifex (12000 to 16000 ind. m-2) increased denitrifica- m-Lwith only 3 replicates. Error bars represent standard errors tion rates by only 80 %, in spite of NO3- concentrations Aquat microb Ecol9: 289-294, 1995

m1A tot. nitrlf.

50 100 Pore water NH (nmol sed.)

Fig. 3. Tubifex tubifex. Sediment concentration profiles of pore water NH4+In microcosms containing tubilicid worm densities ranging from 0 to 70000 ind m-'

tubifex (100000 m-2). Wood (1975), however, attributed an increased diffusion of rhodamine B into sediment to irrigation by tubificid worms. Increased porosity and/or diffusivity results in a deeper penetration by 0 ; I 1 I I oxygen (Davis 1974, Fukuhara 1987) and NO,-, but the 0 2 0 4 0 60 8 0 transport of reduced material to the sediment surface Animal density (103 ind. m-*) and the accumulation of organically enriched fecal pellets can result in increased rates of oxygen and Fig. 2. Tubifex tubifex. (A) Total nitrification and denitrifica- NO3- utilization. McCall & Fisher (1980) reported a 2- tion of NO3- generated within the sediment (dn) and (B)denitrification of NO3-coming from the overlying water (dw), cor- fold increase in organic content at the 0-1.5 cm layer, rected to 50 pM NO3- in the water phase, versus tubificid probably attributable to the higher organic content in worm density. Each data point is the average of 6 replicates. fecal pellets compared to bulk sediment (35%; except for microcosms containing 10000 ind. m-2 with only 3 Brinkhurst et al. 1972). replicates. Error bars represent standard errors. Units: pm01 m-2 h-l The pore water NH4+content increased in sediments at worm densities >20000 ind, m-'. Slmilar increases occurred in the 0-5 cm layer of rice-field soil contain- in the overlying water 4 times higher in their expen- ing Limnodrilus hoffmeisteri and Branchyura sowerbyi ment. (Kikuchi & Kurihara 1977) and in sediment with mixed The increase in oxygen consumption rates and in populations of L. hoffmeisteri and L. cervix (Fisher nitrate uptake, in bioturbated microcosms, could be 1982). This increased NH,' might have been due either associated with a change in the characteristics of the to animal excretion or to enhanced microbial activity surface sediment, due to the particular feeding strat- (Matisoff et al. 1985). Certainly, tubiflcids excrete NH4+ egy of Tubifex tubifex or to active irrigation of the bur- at qxite high rates, 0.6 nmol NI-I,' ind.-' h-' (Gardner el row by the worms (Pelegn et al. 1994). Tubificid worms al. 1983), and this occurs well below the sediment sur- are capable of affecting many sediment properties by face (Wisniewski & Planter 1985). It is probable that the formation of a stable layer of fecal pellets at the most of the increased NH,' in our sediments could be sediment surface. Egested material is packaged into attributed to direct worm excretion, rather than to mucus-bound pellets which may maintain their in- increased microbial activity. Increases in porosity tegrity for some time after deposition on the sediment and/or diffusivity at the sediment surface due to surface (Flsher 1979). McCall & Fisher (1980) reported Tubifex tubifex may lead to enhanced NH4+fluxes out increases in sediment porosity of 80 % at the 0-1.5 cm of the sediment, compared to non-inhabited micro- layer due to fecal pellets (Limnodrilus hoffmeisten and cosms. Our measurements showed indeed increased T. tubifex, 3000 ind. m-2). They believed that changes NH,' fluxes in the presence of the worms. in sediment porosity were responsible for an increased Nitrification rates increased by 23 % from 0 to 20000 diffusion of chloride tracer in sediment containing T. ind. m-', and decreased by 30% from 20000 to 70000 Pelegri & Blackburn: Effects of Tubifex tubifex on N-mineralization 293

ind. m-'. This accords well with increased rates of nitri- often has a high C:N ratio, because of its origins in fication as organic degradation increases up to the plant structural material. Micro-organisms utilizing highest organic loading rates, at which an inhibition of this material can be nitrogen-starved (Fenchel & nitrification took place (Blackburn 1990, Blackburn & Blackburn 1979). The higher rates of denitrification Blackburn 1993, Caffrey et al. 1993). As the rate of were counterbalanced by higher rates of NH4+ flux organic degradation increased, the quantity of NH4+ from the sediment, due mainly to worm excretion. It is increased, but the availability of O2 decreased result- likely, however, that the presence of bioturbating ing in a compression of the zone of nitrification into a worms, stimulating rates of denitrification, resulted in thinner band, closer to the sediment surface. Eventu- a net loss of nitrogen that could otherwise have been ally, even though NH,' concentrations might be high, used by the primary producers. the rate of loss of NH,' by diffusion exceeds the rate at which it could be oxidized. This scenario fits the Acknowledgements. This work was supported by sectorial observed data: NH,' concentrations increased with grant no. B/STEP-900024 from the Commission of the Euro- worm density and this was coupled to higher rates of pean Communltles and by the Danish Minlstry of Environ- O2 uptake. The NH,' was probably mostly produced mental Program. Mre thank Preben Ssrensen and Bloconsult by the worms and at least some of the increased 0' (Denmark) for their valuable help. uptake was also attributed to worm uptake. It is likely, however, that 0' penetration into the sediment surface LITERATURE CITED decreased with increased worm population densities, partially due to the microbial activity in the fecal pellet Blackburn ND, Blackburn TH (1993) A reaction diffusion layer at the sediment surface. The relatively low model of C-N-S-0 species in a stratified sediment. FEMS Microbiol Ecol 102:207-215 degree of coupling between nitrification and denitrifi- Blackburn TH (1980) Seasonal variations in the rate of cation (-25 %) also suggests that nitrification was close organic-N mineralization in anoxic marine sediments. In: to the sediment-water interface and that nitrate could Blogeochimie de la matiere organique a l'lnterface eau- escape by diffusion into the overlying water. Chater- sediment mann. Colloq int CNRS 293:173-183 Blackburn TH (1990) Denitrification model for manne sedi- paul et al. (1980) found that the presence of ment. In: Revsbech NP, Ssrensen J (eds) Denitriflcation in Limnodrilus hoffmeisteriand Tubifex tubifex (12 000 to soil and sediment. John Wiley and Sons Ltd, Chichester, 16000 ind, m-2) increased sediment nitrification rates p 323-337 by 40% during a substantially longer incubation Blackburn TH (1993) Turnover of lSNH4+tracer in sediments. period. In: Kemp P, Sherr BF, Sherr EB, Cole JJ (eds) Handbook of methods in aquatic microbial ecology. Lewis Publishers. Boca Raton, FL, p 643-648 Bower CE, Holm-Hansen T (1980) A salicylate-hypochlorite CONCLUSION method for determining ammonia in seawater. Can J Fish Aquat Sci 37 :794-798 Brinkhurst RO, Chua KE, Kaushik NK (1972) Interspecific The presence of Tubifex tubifex enhanced O2 and interactions and selective feeding by tubificid oligo- NO3- uptake, associated with changes in surface sedi- chaetes. Limnol Oceanogr 17:122-133 ment diffusivity, due to their feeding behavior and/or Caffrey JM, Sloth NP, Kaspar HF, Blackburn TH (1993) Effect to irrigation activity of the worms. Nitrification was of organic loading on nitrification and denitrification In a stimulated at low worm densities (20000 ind. m-') but marine sediment microcosm. FEMS Microbiol Ecol 12. 159-167 was inhibited at higher densities (70000 ind. m-2).It is Chaterpaul L, Robinson J'B, Kaushik NK (1979) Role of tubifi- likely that the O2 penetration depth decreased with cid worms on nitrogen transformations In stream sedi- increasing worm density, due to the transport of ment. J Fish Res Bd Can 36:673-678 reduced compounds and organic matter, with the fecal Chaterpaul L, Robinson JB, Kaushik NK (1980) Effects of tubificid worms on denitrification and nitrification in stream pellets, to the sediment surface. These anoxic condi- sediments. Can J Fish Aquat Sci 37:650-663 tions led to increased rates of denitrification and were Davis RB (1974) Stratigraphic effects of tubificids in profundal responsible for the decrease in nitrification at higher lake sediments. Limnol Oceanogr 19:466-488 worm densities. An average of 25 % of the NO3- uptake Fenchel TM, Blackburn TH (1979). The nitrogen cycle. In: by the sediment was removed from the system by de- Fenchel T, Blackburn TH (eds) Bacteria and mineral cycling. Academic Press. London, p 101-126 nitrification. There was some evidence for the reduc- Fisher JB (1979) Effects of tubificid oligochaetes on sediment tion of NO3- to NH4+at a population density of 70000 movement and the movement of materials across the sed- ind. m-' (data not shown). This process was not mea- iment-water interface. PhD thesis, Case Western Reserve sured at other worm densities. It is most likely that the University, Cleveland Fisher JB (1982) Effects of macrobenthos on the chemical dia- discrepancy between the rate of NO,- uptake and the genesis of freshwater sediments. In: McCall PL, Tevesz rate of denitrification was due to NO,- assimilation by MJS (eds) Animal-sediment relations. Plenum Press, New micro-organisms. The detritus in freshwater systems York, p 177-218 294 Aquat microb Ecol 9: 289-294, 1995

Fukuhara H (1987)The effect of tub~ficidsand chironom~dson N~elsenLP, Chnstensen PB,Revsbech NP, Ssrensen J (1990) particle redistribution of lake sediment. Ecol Res 2: Denitrificat~onand oxygen respiration in biofilms studied 255-264 with a microsensor for nitrous oxide and oxygen. Microb Gardner WS, Nalepa TF, Slavens DR, Laird G (1983)Patterns ECOI19:63-72 and rates of nitrogen release b\. benthic chironomids and Palmer MF (1968) Aspects of the respiratory physiology of oligochaeta. Can J Fish Aquat Sci 40:259-266 Tubifex tubifex in relation to its ecology. J 2001, Lond Grasshoff K, Erhardt M, Kremling K (1983) Methods of sea- 154:463-473 water analysis, 2nd edn. Verlag Chemie, Weinheim Pelegri SP, Blackburn TH (1995) Effect of bioturbation by Kikuchi E, Kurihara Y (1977) In vitro studies of the effects of Nereis sp., Mya arenana and Cerastodermasp., on nitrifi- tubificids on the biological, chemical and physical charac- cation and denitrification in estuarine sedirnents Ophelia teristics of submerged ricefield soil and overlying water. 42:289-299 Oikos 29:348-356 Pelegn SP, Nielsen LP, Blackburn TH (1994)Denitrification in Kikuchi E, Kurihara Y (1982) The effects of the oligochaete estuarine sediment stimulated by the irrigation activity of Branchiura sorverbyiBeddard (Tubiflcidae) on the biolog- the amphipod Corophium volutator (Pallas). Mar Ecol ical and chemical characteristics of overlying water and Prog Ser 105:285-290 soil in a submerged ricefield soil system. Hydrobiologia Revsbech NP, Christensen PB, Nielsen LP, Ssrensen J (1989) 97:203-208 Denitrification in a trickling filter biofilm studied by a Koike l, Hattori A (1978) Simultaneous determination of nitri- microsensor for oxygen and nitrous oxide. Water Res 23: fication and, nitrate reduction in coastal sediments by a I5N 867-871 dilution technique. Appl environ Microbiol 35:853-857 Rhoads DC (1974) Organism-sediment relations on the Kristensen E, Jensen MH, Aller RC (1991) Direct measure- muddy sea floor. Oceanogr Mar Biol A Rev 12:263-300 ment of dissolved inorganic nitrogen exchange and deni- Risgaard-Petersen N, Rysgaard S, Revsbech NP (1993) A sen- trification in individual polychaete (Nereis virens) bur- sitive assay for determination of "N/''N isotope distribu- rows. J mar Res 49:355-377 tion in NO3 J microb Meth 17:155-164 Matisoff G, Fisher JB, Matis S (1985) Effects of benthic Robbins JA, McCall PL, Fisher JB, Krezoski JR (1979)Effect of macroinvertebrates on the exhange of solutes between deposit feeders on migration of 137-CSin lake sediments. sediments and freshwater. Hydrobiologia 122:19-33 Earth planet Sci Lett 42:277-287 McCall PL, Fisher JB (1980) Effects of tubificid oligochaetes Weiss RF (1970) The solubility of nitrogen, oxygen and argon on physical and chemical properties of Lake Erie sedi- in water and seawater. Deep Sea Res 17:721-735 ments. In: Brinkhurt RO, Cook DG (eds) Aquatic ollgo- Wisniewski RJ, Planter M (1985) Exchange of phosphorus chaete biology. Plenum Press, New York, p 253-317 across sediment-water interface (with special attention McCall PL, Tevesz MJS (1982) The effects of benthos on phys- to the influence of biotic factors) in several lakes of ical properties of freshwater sediments. In: McCall PL, d~fferent trophic status. Verh int Verein Limnol 22: Tevesz MJS (eds) Animal-sediment relations. Plenum 3345-3349 Press, New York. p 105-176 Wood LW (1975) Role of oligochaetes In the circulation of Nielsen LP (1992) Denitrification in sediment determined from water and solutes across the mud-water interface. Verh int nitrogen isotope pairing. FEMS Microbiol Ecol86:357-362 Verein Limnol 19:1530-1533

Responsible Subject Editor: T Fenchel, Helslngsr, Denmark Manuscript f~rstreceived: ALI~LIS~ 2, 1995 Revised version accepted:November 10, 1995