Diversity trumps acidification: Lack of evidence for carbon dioxide enhancement of Trichodesmium community nitrogen or carbon fixation at Station ALOHA Gradoville, M. R., White, A. E., Böttjer, D., Church, M. J., & Letelier, R. M. (2014). Diversity trumps acidification: Lack of evidence for carbon dioxide enhancement of Trichodesmium community nitrogen or carbon fixation at Station ALOHA. Limnology and Oceanography, 59(3), 645-659. doi:10.4319/lo.2014.59.3.0645 10.4319/lo.2014.59.3.0645 American Society of Limnology and Oceanography, Inc. Version of Record http://cdss.library.oregonstate.edu/sa-termsofuse Limnol. Oceanogr., 59(3), 2014, 645–659 E 2014, by the Association for the Sciences of Limnology and Oceanography, Inc. doi:10.4319/lo.2014.59.3.0645 Diversity trumps acidification: Lack of evidence for carbon dioxide enhancement of Trichodesmium community nitrogen or carbon fixation at Station ALOHA Mary R. Gradoville,1,* Angelicque E. White,1 Daniela Bo¨ttjer,2 Matthew J. Church,2 and Ricardo M. Letelier 1 1 Oregon State University, College of Earth, Ocean, and Atmospheric Sciences, Corvallis, Oregon 2 University of Hawaii at Manoa, Department of Oceanography, Honolulu, Hawaii Abstract We conducted 11 independent short-term carbon dioxide (CO2) manipulation experiments using colonies of the filamentous cyanobacteria Trichodesmium isolated on three cruises in the North Pacific Subtropical Gyre (NPSG). Dinitrogen (N2) and carbon (C) fixation rates of these colonies were compared over CO2 conditions ranging from , 18 Pa (equivalent to last glacial maximum atmospheric PCO2 )to, 160 Pa (predicted for , year 2200). Our results indicate that elevated PCO2 has no consistent significant effect on rates of N2 or C fixation by Trichodesmium colonies in the NPSG under present environmental conditions. Differences between PCO2 treatments were not modulated by phosphorus amendments, iron amendments, or light level. Sequencing the hetR, nifH, 16S, and internal transcribed spacer genes of Trichodesmium colonies revealed a highly diverse community of Trichodesmium and other N2-fixing colony-associated organisms. The species composition of Trichodesmium demonstrated spatiotemporal variability, but over half of total sequences were phylogenetically closely related (. 99% hetR sequence similarity) to isolate H9-4 of T. erythraeum, which showed no response to elevated PCO2 in previous laboratory experiments. Our handpicked Trichodesmium colonies included a substantial number of organisms other than Trichodesmium with the metabolic capacity for N2 and C fixation. We suggest that the diverse assemblage of Trichodesmium species and coexisting microorganisms within the colonies can explain the lack of an observed CO2 enhancement of N2 or C fixation rates, because different species are known to have different specific affinities for CO2. Burning fossil fuels and other human actions are adding by Trichodesmium, a group of filamentous, non-heterocys- carbon (C) to the atmosphere at unprecedented rates (IPCC tous cyanobacteria (Karl et al. 1997; Capone et al. 2005). 2013). This increase in atmospheric carbon dioxide (CO2) Recently, several laboratory studies on two isolates of propagates into the ocean, increasing the partial pressure of Trichodesmium erythraeum have shown enhanced N2 and, CO2 (PCO2 ) of seawater and causing a suite of changes to on some occasions, C fixation, in response to increases in ocean chemistry, collectively termed ocean acidification PCO2 (Barcelos e Ramos et al. 2007; Hutchins et al. 2007; (OA; Doney et al. 2009). Among the predicted effects of Levitan et al. 2007). In addition, an isolate of Crocosphaera OA is a possible upregulation of cellular processes by C- watsonii, a naturally abundant unicellular cyanobacterium limited phytoplankton. Because almost all phytoplankton (Moisander et al. 2010), also increases N2 and C fixation invest energy in carbon-concentrating mechanisms (CCMs) rates when exposed to elevated PCO2 (Fu et al. 2008; M. to help saturate their low CO2-affinity carboxylating Gradoville unpubl.). In these studies, a doubling of present- enzyme, ribulose-1,5-bisphosphate carboxylase-oxygenase day PCO2 increased Trichodesmium N2 fixation rates (RuBisCO; Badger et al. 1998), it has been proposed that between 35% (Kranz et al. 2009) and 138% (Levitan et al. future OA may allow certain phytoplankton species to 2007). An enhancement of global marine N2 fixation in this increase their C fixation rates and reallocate energy from range could have substantial consequences for the coupling CCMs to other metabolic processes (Hutchins et al. 2007). of elemental cycles, ocean primary productivity, carbon sequestration, and the biological pump. Cyanobacteria, which evolved under the high CO2 conditions of the Archaean Eon, can display an especially Still, it is unclear whether the early evidence for a PCO2 enhancement of N fixation by Trichodesmium and Croco- low CO2-affinity RuBisCO compared with other phyto- 2 sphaera isolates manipulated in the laboratory can be used plankton, and may thus benefit from future elevated PCO2 (Rost et al. 2008). There is a particular interest in the effect of to predict the response of diazotrophs under natural OA on a class of cyanobacteria with the capacity to fix environmental conditions. In most laboratory studies, Trichodesmium and Crocosphaera cultures are grown in dinitrogen (N2), termed diazotrophs. The bioavailable nitrogen (N) provided by marine diazotrophs helps fuel an artificial medium with excess nutrients uncharacteristic new production, especially in oligotrophic open-ocean of open-ocean habitats (Chen et al. 1996). These growth conditions may influence the effect of OA on N2 fixation. environments, where N2 fixation can account for up to half of the new N supporting export from the euphotic zone For example, under iron- (Fe-) limiting conditions, OA has (Karl et al. 1997). A large fraction of this fixed N is provided a negative or neutral effect on N2 fixation by Trichodes- mium and Crocosphaera isolates (Fu et al. 2008; Shi et al. 2012). To date, the only study of Trichodesmium strain * Corresponding author: [email protected] IMS101 grown in nutrient-amended seawater rather than 645 board incubators undercolonies varying were CO incubatedOahu in (Table surface-seawater 1). In cooled all deck- experiments, picked Methods and C fixation-rate measurements from P recent evidence suggestswhen that grown the in N C may seawater not (Shi befixation et rates the even primary al. in limitingYBCII Fe-replete 2012). media treatments, nutrient (Chen Furthermore, suggesting for et that this al. 1996) strain found that OA reduced N 646 water communities inresponses previous exhibited CO by this directionnaturally occurring is microbialof determining assemblages. how OA A will whether keyassociated affect atmospheric biological step feedbacks the rate require in processes an in understanding contrasting complex, diazotrophic communities to OA. results paint anet unclear picture al. of the 2012; response D. of natural Bo samples of exhibit the same positiveTrichodesmium response to CO naturally occurring strains [Hutchins et al.initial 2007, experiments 2013]). that used IMS101 and GBRTRLI101 regimes. Carbon and N (HOT) program the long-term fieldALOHA site (A Long-term of Oligotrophicthree the Habitat cruises Assessment), Hawaii aboard Ocean the R/V Time-series relationship betweenand P Southunamended Pacific whole-community gyresecologically experiments relevant in do interpretations theconditions not difficult. in North In these displayand experiments contrast, (Lomas a the et applicationLomas al. significant et 2012) of al. make 2012), phosphorus-Atlantic the specific and (P-) methodological approaches the andby Gulf Fe-replete of Mexico (Hutchins et al. 2009; ganisms utilized in ourTrichodesmium incubations. experiments, together with genetic characterization of the CO to characterize themethodology. To effect this ofphysiology, end, environmental the elevated conditions, aim and/or P ofments experimental the are present due study to is differences in community composition, nutrient and light conditions. Here,North we Pacific present results Subtropical ofoccurring N Gyre (NPSG) under a range of 202 quent deoxyribonucleic acid (DNA) extraction (Table 1). results. Although elevated P diazotrophic communities to OA have yielded conflicting Experimental design— Accurate forecasts of future marine nutrient cycling and In addition, experiments investigating the responses of Trichodesmium 2 Trichodesmium m manipulations can vary widely even within the genus m plankton net that was hand-towed at Trichodesmium Trichodesmium assemblage and colony-associated microor- (e.g., certain laboratory strains do not , colonies were collected at night using a ¨ttjer pers. comm.). These conflicting 100 km north of the Hawaiian island of colonies isolated from the Subtropical Trichodesmium Trichodesmium 2 CO Experiments were carried out on fixation rates were measured, and assemblages in the oligotrophic 2 colonies were saved for subse- CO and N Kilo Moana 2 has stimulated N 2 2 2 2 manipulation experi- colonies and whole , nutrient, and light fixation rates (Law fixation response to colonies and whole- CO CO 2 at and near Sta. observed in the Trichodesmium 2 2 manipulation on naturally , 2kmh 2 fixation Gradoville et al. 2 1 2 2 Table 1. Locations and ambient environmental conditions of seawater from which Trichodesmium colonies were collected. Sea surface temperature (SST) and