Temperature response of N cycling in Arctic Sediments
Accepted Article
Received date: 01/17/2014 Accepted date: 08/03/2014 Temperature response of denitrification and anammox rates and microbial community structure in Arctic fjord sediments 1 Andy Canion1, Will A. Overholt2, Joel E. Kostka2*, Markus Huettel1, Gaute Lavik3 and Marcel M.M. Kuypers3
1
Earth, Ocean, and Atmospheric Science Department, Florida State University, Tallahassee, FL
2
Schools of Biology and Earth and Atmospheric Science, Georgia Institute of Technology, Atlanta, GA
3
Max Planck Institute for Marine Microbiology, Bremen, Germany
Running Title: Temperature and N cycling in Arctic Sediments Key Words: Anammox / Arctic / Denitrification / Temperature Subject Category: Microbial ecology and functional diversity of natural habitats
*Corresponding author mailing address: Joel E. Kostka Georgia Institute of Technology Schools of Biology and Earth and Atmospheric Science Room 225, Cherry Emerson Bldg. 310 Ferst Drive Atlanta, Georgia, 30332-0230 Phone: 404-385-3325 Fax: 404-894-0519 Email address: [email protected]
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/1462-2920.12593
1 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments ABSTRACT
Accepted Article
1 2
The temperature dependency of denitrification and anaerobic ammonium oxidation
3
(anammox) rates from Arctic fjord sediments was investigated in a temperature gradient block
4
incubator for temperatures ranging from -1 to 40 °C. Community structure in intact sediments
5
and slurry incubations was determined using Illumina MiSeq SSU rRNA gene sequencing. The
6
optimal temperature (Topt) for denitrification was 25 – 27 °C, whereas anammox rates were
7
optimal at 12 – 17 °C. Both denitrification and anammox exhibited temperature responses
8
consistent with a psychrophilic community, but anammox bacteria may be more specialized for
9
psychrophilic activity. Long-term (1 – 2 months) warming experiments indicated that
10
temperature increases of 5 – 10 °C above in situ had little effect on the microbial community
11
structure or the temperature response of denitrification and anammox. Increases of 25 °C shifted
12
denitrification temperature responses to mesophilic with concurrent community shifts, and
13
anammox activity was eliminated above 25 °C. Additions of low molecular weight organic
14
substrates (acetate and lactate) caused increases in denitrification rates, corroborating the
15
hypothesis that the supply of organic substrates is a more dominant control of respiration rates
16
than low temperature. These results suggest that climate-related changes in sinking particulate
17
flux will likely alter rates of N removal more rapidly than warming.
18 19 20 21 22 23 24
2 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments INTRODUCTION
Accepted Article
25 26
Denitrification and anammox are the dominant processes for removal of fixed nitrogen
27
from the ocean (Canfield et al., 2010). Up to 50 % of global N removal by denitrification and
28
anammox occurs in coastal and continental shelf sediments (Gruber and Sarmiento, 1997;
29
Codispoti, 2007). Arctic shelf sediments contribute significantly to global N removal as more
30
than 50% of the seafloor of the Arctic Ocean is covered by continental shelf sediments
31
(Jakobsson et al., 2003), which account for 7 to 11% of the global burial rate of organic carbon
32
(Stein and Macdonald, 2004). Productive shelf regions in the Chukchi and Barents Seas have
33
been identified as active sites of denitrification (Devol et al., 1997; Glud et al., 1998). Based on
34
modeled denitrification rates, the contribution of Arctic sediments to global N removal may be as
35
high as 13% (Chang and Devol, 2009), even though the Arctic Ocean only comprises 2.5% of
36
the total ocean surface area. More recently, substantial contributions of anammox activity to N
37
removal were reported from Arctic shelf sediments (Rysgaard et al., 2004; Gihring et al., 2010a).
38
Given the importance of Arctic shelves as sites of N removal, it is crucial to understand
39
how global climate change may alter the biogeochemical cycling of N in Arctic shelf sediments.
40
In the Atlantic sector of the Arctic, warmer Atlantic water is likely to intrude further onto shelves
41
(Spielhagen et al., 2011). Also, loss of sea ice and stronger water column stratification from
42
freshwater inputs are expected to increase the importance of phytoplankton production relative to
43
ice algae, resulting in an altered sinking flux of particulate organic matter (Arrigo et al., 2008).
44
Major alterations in N removal rates on Arctic shelves have the potential to affect N fixation in
45
the North Atlantic due to the role of the Arctic basin as a through-flow for phosphate-rich waters
46
from the Pacific to the Atlantic Ocean (Yamamoto-Kawai et al., 2006).
47 48
In contrast to low latitude shelf environments, biogeochemical cycling in Arctic shelf
sediments occurs under permanently cold (< 5 °C) conditions. Cultivation-based surveys in the 3 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments Arctic have demonstrated that sediments are replete with psychrophilic and psychrotolerant
Accepted Article
49 50
microbes (Helmke and Weyland, 2004). Psychrophilic enzymes have been shown to maintain
51
relatively high catalytic activity at low temperatures relative to their optimal temperature (Feller
52
and Gerday, 2003), indicating that sediment microbial activity is not inhibited by permanently
53
cold conditions. Indeed, rates of polysaccharide hydrolysis, oxygen respiration, denitrification,
54
and sulfate reduction from Arctic sediments may overlap with those of temperate sediments due
55
to the activity of psychrophiles (e.g., Arnosti et al., 1998; Thamdrup and Fleischer, 1998; Kostka
56
et al., 1999; Gihring et al., 2010a). A general consensus from studies of Arctic sediments is that
57
the supply of organic substrates is likely a more dominant control of organic matter turnover than
58
in situ temperatures.
59
The interaction of low temperature and substrate availability must also be considered as a
60
control of microbial activity in permanently cold environments (Pomeroy and Wiebe, 2001).
61
Cold temperatures can have a direct effect on the specific affinity of membrane transport proteins
62
(Nedwell, 1999), requiring elevated substrate concentrations to maintain growth in pure cultures.
63
At the microbial community level, the availability of low molecular weight monomeric
64
substrates for terminal respiration may depend on extracellular hydrolysis rates, which in turn,
65
are influenced by the enzymatic capabilities and community composition of exoenzyme-
66
producing microbes (Arnosti, 2004). Thus, even though marine sediments have a relatively high
67
bulk carbon concentration as compared to the water column, respiration rates may still be limited
68
by organic matter quality.
69
The metabolic potential of nitrate-respiring communities will be an important
70
determinant of nitrogen removal rates in Arctic shelf sediments in response to climate-related
71
ecosystem changes. In the present study, we examined the role of temperature and organic
72
substrate availability in regulating both rates of fixed nitrogen removal (denitrification and 4 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments anammox) and microbial community structure in permanently cold sediments under both un-
Accepted Article
73 74
manipulated and experimentally warmed conditions. The following hypotheses were addressed:
75
1) Warming of sediments will increase the contribution psychrotolerant and mesophilic nitrate-
76
respiring microbes already present in sediments 2) Addition of labile carbon substrates will alter
77
the temperature response of nitrate respiration due to the interactive effects of temperature and
78
substrate availability.
79 80
METHODS
81
Sample sites
82
Sediment cores were collected in 13.6 cm diameter core liners using a Haps corer
83
(Kanneworff and Nicolaisen, 1972) in summer of 2010 and 2011 from three western fjords of the
84
Svalbard archipelago (Table 1, Figure S1). Sites were located in Smeerenburgfjorden (Station J,
85
79°42.01' N, 11°05.20' E), Kongsfjorden (Station Q, 78°59.43' N, 12°17.87' E), and Van
86
Keulenfjorden (Station AC, 77°33.25' N, 15°38.24' E). Sediments from Station J were black
87
clayey and rich with organic matter, while the sediments from Station AC and Station Q were
88
dark gray silty clay and reddish-brown loamy, respectively. Vertical profiles of pore-water
89
nitrate, nitrite, and ammonium concentration were measured at each station using Rhizon
90
samplers (Rhizosphere Research Products) that were inserted into ports drilled along
91
polycarbonate core liners (9.5cm diameter) (Seeberg-Elverfeldt et al., 2005). Rhizons were
92
rinsed with the first 0.5 mL of pore-water, and then 1.5 mL was collected for each 1cm depth
93
interval. Surface waters near the west coast of Svalbard are slightly warmer than the
94
surrounding Barents Sea due to the influence of the West Spitsbergen Current, and the western
95
fjords have been proposed as model sites for the study of warming effects on high latitude
96
ecosystems (Svendsen et al., 2002). 5 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments Denitrification and anammox rate measurements
Accepted Article
97 98
Intact sediment sub-cores (3.5 cm diameter) with 15 cm of sediment and 15 cm of
99
overlying water were collected from the Haps corer and were immediately transported to
100
temperature-controlled rooms maintained at 1 °C. Incubation experiments for rate
101
determinations were initiated within 24 hours and ran for 24 to 72 hours. Denitrification and
102
anammox rate determinations were made according to the 15N isotope pairing technique (IPT,
103
Nielsen, 1992) with a final overlying water label concentration of 50 µM Na15NO3- (99 atom%;
104
Cambridge Isotope Laboratories, Andover, MA). Details of the experimental protocol can be
105
found in Gihring et al. (2010a). Anoxic slurry incubations with 100 µM Na15NO3- were
106
performed in parallel with each intact core incubation to determine the relative contribution of
107
anammox (ra) to total N2 production (Thamdrup and Dalsgaard, 2002). Anoxic slurry
108
incubations were also used to determine the potential anammox and denitrification rates with
109
respect to depth in the upper 5 cm of sediment.
110
Linear regressions of excess 29N2 or 30N2 concentration against time were used to
111
calculate the rate of production of 29N2 and 30N2, and the standard errors of regression slopes
112
were used to calculate the standard errors of rates. All regression slopes were determined to be
113
significantly greater than 0 (t-test, p
Accepted Article
Received date: 01/17/2014 Accepted date: 08/03/2014 Temperature response of denitrification and anammox rates and microbial community structure in Arctic fjord sediments 1 Andy Canion1, Will A. Overholt2, Joel E. Kostka2*, Markus Huettel1, Gaute Lavik3 and Marcel M.M. Kuypers3
1
Earth, Ocean, and Atmospheric Science Department, Florida State University, Tallahassee, FL
2
Schools of Biology and Earth and Atmospheric Science, Georgia Institute of Technology, Atlanta, GA
3
Max Planck Institute for Marine Microbiology, Bremen, Germany
Running Title: Temperature and N cycling in Arctic Sediments Key Words: Anammox / Arctic / Denitrification / Temperature Subject Category: Microbial ecology and functional diversity of natural habitats
*Corresponding author mailing address: Joel E. Kostka Georgia Institute of Technology Schools of Biology and Earth and Atmospheric Science Room 225, Cherry Emerson Bldg. 310 Ferst Drive Atlanta, Georgia, 30332-0230 Phone: 404-385-3325 Fax: 404-894-0519 Email address: [email protected]
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/1462-2920.12593
1 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments ABSTRACT
Accepted Article
1 2
The temperature dependency of denitrification and anaerobic ammonium oxidation
3
(anammox) rates from Arctic fjord sediments was investigated in a temperature gradient block
4
incubator for temperatures ranging from -1 to 40 °C. Community structure in intact sediments
5
and slurry incubations was determined using Illumina MiSeq SSU rRNA gene sequencing. The
6
optimal temperature (Topt) for denitrification was 25 – 27 °C, whereas anammox rates were
7
optimal at 12 – 17 °C. Both denitrification and anammox exhibited temperature responses
8
consistent with a psychrophilic community, but anammox bacteria may be more specialized for
9
psychrophilic activity. Long-term (1 – 2 months) warming experiments indicated that
10
temperature increases of 5 – 10 °C above in situ had little effect on the microbial community
11
structure or the temperature response of denitrification and anammox. Increases of 25 °C shifted
12
denitrification temperature responses to mesophilic with concurrent community shifts, and
13
anammox activity was eliminated above 25 °C. Additions of low molecular weight organic
14
substrates (acetate and lactate) caused increases in denitrification rates, corroborating the
15
hypothesis that the supply of organic substrates is a more dominant control of respiration rates
16
than low temperature. These results suggest that climate-related changes in sinking particulate
17
flux will likely alter rates of N removal more rapidly than warming.
18 19 20 21 22 23 24
2 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments INTRODUCTION
Accepted Article
25 26
Denitrification and anammox are the dominant processes for removal of fixed nitrogen
27
from the ocean (Canfield et al., 2010). Up to 50 % of global N removal by denitrification and
28
anammox occurs in coastal and continental shelf sediments (Gruber and Sarmiento, 1997;
29
Codispoti, 2007). Arctic shelf sediments contribute significantly to global N removal as more
30
than 50% of the seafloor of the Arctic Ocean is covered by continental shelf sediments
31
(Jakobsson et al., 2003), which account for 7 to 11% of the global burial rate of organic carbon
32
(Stein and Macdonald, 2004). Productive shelf regions in the Chukchi and Barents Seas have
33
been identified as active sites of denitrification (Devol et al., 1997; Glud et al., 1998). Based on
34
modeled denitrification rates, the contribution of Arctic sediments to global N removal may be as
35
high as 13% (Chang and Devol, 2009), even though the Arctic Ocean only comprises 2.5% of
36
the total ocean surface area. More recently, substantial contributions of anammox activity to N
37
removal were reported from Arctic shelf sediments (Rysgaard et al., 2004; Gihring et al., 2010a).
38
Given the importance of Arctic shelves as sites of N removal, it is crucial to understand
39
how global climate change may alter the biogeochemical cycling of N in Arctic shelf sediments.
40
In the Atlantic sector of the Arctic, warmer Atlantic water is likely to intrude further onto shelves
41
(Spielhagen et al., 2011). Also, loss of sea ice and stronger water column stratification from
42
freshwater inputs are expected to increase the importance of phytoplankton production relative to
43
ice algae, resulting in an altered sinking flux of particulate organic matter (Arrigo et al., 2008).
44
Major alterations in N removal rates on Arctic shelves have the potential to affect N fixation in
45
the North Atlantic due to the role of the Arctic basin as a through-flow for phosphate-rich waters
46
from the Pacific to the Atlantic Ocean (Yamamoto-Kawai et al., 2006).
47 48
In contrast to low latitude shelf environments, biogeochemical cycling in Arctic shelf
sediments occurs under permanently cold (< 5 °C) conditions. Cultivation-based surveys in the 3 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments Arctic have demonstrated that sediments are replete with psychrophilic and psychrotolerant
Accepted Article
49 50
microbes (Helmke and Weyland, 2004). Psychrophilic enzymes have been shown to maintain
51
relatively high catalytic activity at low temperatures relative to their optimal temperature (Feller
52
and Gerday, 2003), indicating that sediment microbial activity is not inhibited by permanently
53
cold conditions. Indeed, rates of polysaccharide hydrolysis, oxygen respiration, denitrification,
54
and sulfate reduction from Arctic sediments may overlap with those of temperate sediments due
55
to the activity of psychrophiles (e.g., Arnosti et al., 1998; Thamdrup and Fleischer, 1998; Kostka
56
et al., 1999; Gihring et al., 2010a). A general consensus from studies of Arctic sediments is that
57
the supply of organic substrates is likely a more dominant control of organic matter turnover than
58
in situ temperatures.
59
The interaction of low temperature and substrate availability must also be considered as a
60
control of microbial activity in permanently cold environments (Pomeroy and Wiebe, 2001).
61
Cold temperatures can have a direct effect on the specific affinity of membrane transport proteins
62
(Nedwell, 1999), requiring elevated substrate concentrations to maintain growth in pure cultures.
63
At the microbial community level, the availability of low molecular weight monomeric
64
substrates for terminal respiration may depend on extracellular hydrolysis rates, which in turn,
65
are influenced by the enzymatic capabilities and community composition of exoenzyme-
66
producing microbes (Arnosti, 2004). Thus, even though marine sediments have a relatively high
67
bulk carbon concentration as compared to the water column, respiration rates may still be limited
68
by organic matter quality.
69
The metabolic potential of nitrate-respiring communities will be an important
70
determinant of nitrogen removal rates in Arctic shelf sediments in response to climate-related
71
ecosystem changes. In the present study, we examined the role of temperature and organic
72
substrate availability in regulating both rates of fixed nitrogen removal (denitrification and 4 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments anammox) and microbial community structure in permanently cold sediments under both un-
Accepted Article
73 74
manipulated and experimentally warmed conditions. The following hypotheses were addressed:
75
1) Warming of sediments will increase the contribution psychrotolerant and mesophilic nitrate-
76
respiring microbes already present in sediments 2) Addition of labile carbon substrates will alter
77
the temperature response of nitrate respiration due to the interactive effects of temperature and
78
substrate availability.
79 80
METHODS
81
Sample sites
82
Sediment cores were collected in 13.6 cm diameter core liners using a Haps corer
83
(Kanneworff and Nicolaisen, 1972) in summer of 2010 and 2011 from three western fjords of the
84
Svalbard archipelago (Table 1, Figure S1). Sites were located in Smeerenburgfjorden (Station J,
85
79°42.01' N, 11°05.20' E), Kongsfjorden (Station Q, 78°59.43' N, 12°17.87' E), and Van
86
Keulenfjorden (Station AC, 77°33.25' N, 15°38.24' E). Sediments from Station J were black
87
clayey and rich with organic matter, while the sediments from Station AC and Station Q were
88
dark gray silty clay and reddish-brown loamy, respectively. Vertical profiles of pore-water
89
nitrate, nitrite, and ammonium concentration were measured at each station using Rhizon
90
samplers (Rhizosphere Research Products) that were inserted into ports drilled along
91
polycarbonate core liners (9.5cm diameter) (Seeberg-Elverfeldt et al., 2005). Rhizons were
92
rinsed with the first 0.5 mL of pore-water, and then 1.5 mL was collected for each 1cm depth
93
interval. Surface waters near the west coast of Svalbard are slightly warmer than the
94
surrounding Barents Sea due to the influence of the West Spitsbergen Current, and the western
95
fjords have been proposed as model sites for the study of warming effects on high latitude
96
ecosystems (Svendsen et al., 2002). 5 This article is protected by copyright. All rights reserved.
Temperature response of N cycling in Arctic Sediments Denitrification and anammox rate measurements
Accepted Article
97 98
Intact sediment sub-cores (3.5 cm diameter) with 15 cm of sediment and 15 cm of
99
overlying water were collected from the Haps corer and were immediately transported to
100
temperature-controlled rooms maintained at 1 °C. Incubation experiments for rate
101
determinations were initiated within 24 hours and ran for 24 to 72 hours. Denitrification and
102
anammox rate determinations were made according to the 15N isotope pairing technique (IPT,
103
Nielsen, 1992) with a final overlying water label concentration of 50 µM Na15NO3- (99 atom%;
104
Cambridge Isotope Laboratories, Andover, MA). Details of the experimental protocol can be
105
found in Gihring et al. (2010a). Anoxic slurry incubations with 100 µM Na15NO3- were
106
performed in parallel with each intact core incubation to determine the relative contribution of
107
anammox (ra) to total N2 production (Thamdrup and Dalsgaard, 2002). Anoxic slurry
108
incubations were also used to determine the potential anammox and denitrification rates with
109
respect to depth in the upper 5 cm of sediment.
110
Linear regressions of excess 29N2 or 30N2 concentration against time were used to
111
calculate the rate of production of 29N2 and 30N2, and the standard errors of regression slopes
112
were used to calculate the standard errors of rates. All regression slopes were determined to be
113
significantly greater than 0 (t-test, p
