Environmental Science and Pollution Research

https://doi.org/10.1007/s11356-020-10429-4

RESEARCH ARTICLE

Effects of elevated nitrogen on the growth and geosmin productivity

of Dolichospermum smithii
Qingyue Shen 1 & Kazuya Shimizu 2 & Hanchen Miao 1 & Shinya Tsukino 3 & Motoo Utsumi 2,4 & Zhongfang Lei 2 &
Zhenya Zhang 2 & Osamu Nishimura 5 & Yasuhiro Asada 6 & Naoshi Fujimoto 7 & Hirokazu Takanashi 8 & Michihiro Akiba 6

Received: 16 April 2020 / Accepted: 6 August 2020

# Springer-Verlag GmbH Germany, part of Springer Nature 2020

Abstract
Geosmin is one of the most common earthy-musty odor compounds, which is mainly produced by cyanobacteria in surface
water. Nitrogen (N) is an important factor affecting the growth of cyanobacteria and its secondary metabolites production due to
the eutrophication. In this study, we compared the effects of elevated N on the growth and geosmin productivity of
Dolichospermum smithii NIES-824 (synonym Anabaena smithii NIES-824), aiming to better understand the mechanisms in-
volved and give an important and fundamental knowledge to solve off-flavor problem. Results show that elevated N concen-
tration promoted more chlorophyll a (Chl-a) production, whereas the geosmin synthesis decreased, revealing a possible com-
petitive correlation between the Chl-a concentration and geosmin production of D. smithii NIES-824. The majority of geosmin (>
90%) was retained intracellularly during the 28 days of cultivation. The qRT-PCR analysis demonstrates that the expression level
of the geosmin synthase gene (geoA) was constitutive and decreased at the higher N concentration during the exponential growth
phase of cyanobacterial cells. Furthermore, the decrease of geoA expression during the decline phase suggested that geoA
transcription was closely related to cell activity and isoprenoid productivity.

Keywords Geosmin . geoA Expression . Cyanobacteria . Nitrogen . Earthy-musty odor

Introduction and Schindler 2009). Cyanobacterial blooms occurred period-

ically accompanied by earthy-musty odor in drinking water
Over the past few decades, excessive human activities and and aquaculture ponds have been widely reported worldwide
urbanization processes have led to severe eutrophication and (Giglio et al. 2008; Wang et al. 2011; Tsao et al. 2014; Watson
cyanobacterial blooms in aquatic ecosystems, bringing about et al. 2016). Although these odor compounds have no risk to
many water quality problems including off-flavor as well as human health, they could decline the water organoleptic qual-
toxic compounds in drinking water and aquaculture (Smith ity and affect the aquaculture and fisheries, triggering

Qingyue Shen and Kazuya Shimizu contributed equally to this work.

Responsible Editor: Philippe Garrigues

* Kazuya Shimizu 5
Department of Civil and Environmental Engineering, Tohoku
shimizu.kazuya.fn@u.tsukuba.ac.jp University, 6-6-06 Aramaki-Aza Aoba, Sendai, Miyagi, Japan

6
1
Graduate School of Life and Environmental Sciences, University of National Institute of Public Health, 2-3-6 Minami Wako,
Tsukuba, 1-1-1 Tennoudai, Tsukuba, Ibaraki, Japan Saitama, Japan
2
Faculty of Life and Environmental Sciences, University of Tsukuba, 7
Faculty of Applied Biosciences, Tokyo University of Agriculture,
1-1-1 Tennoudai, Tsukuba, Ibaraki, Japan
1-1-1 Sakuragaoka, Setagaya, Tokyo, Japan
3
Faculty of Life Sciences, Toyo University, 1-1-1 Izumino,
Ora-gun, Gunma, Japan 8
Department of Chemistry, Biotechnology and Chemical
4
Microbiology Research Center for Sustainability, University of Engineering, Kagoshima University, 1-21-40 Korimoto,
Tsukuba, 1-1-1 Tennoudai, Tsukuba, Ibaraki, Japan Kagoshima-city, Kagoshima, Japan
 Environ Sci Pollut Res

profound socio-economic impacts (Robin et al. 2006; Frank amount of activated carbon, and high-quality products in
et al. 2009). aquaculture. Up to now, however, little information is avail-
Geosmin and 2-methylisoborneol (2-MIB) are the most able on geosmin production from geosmin-producing
common earthy-musty odor compounds, which are produced cyanobacteria, especially with geoA-mRNA expression.
by cyanobacteria in water bodies, and actinomycetes in soils Recently, cyanobacteria kept predominant in eutrophicated
(Jüttner and Watson 2007). Humans are very sensitive to the water bodies, while nitrogen (N) and phosphorus (P) play an
earthy smell of these compounds (< 10 ng/L) (Cook et al. important role for cyanobacterial bloom. P was considered the
2001). In addition, it is difficult to remove these earthy- primary parameter for cyanobacterial biomass accumulation,
musty odor compounds by conventional water treatment pro- whereas it is widely observed that N limitation is a common
cesses. Therefore, the advanced oxidation and activated car- phenomenon in many lakes (Elser et al. 2007; Dolman et al.
bon adsorption processes are needed to efficiently remove 2012). Less attention has been concentrated on N than P, and
geosmin and 2-MIB with high operation costs (Westerhoff few researches were conducted on the effect of increased N
et al. 2005; Srinivasan and Sorial 2011). loading caused by human activities for earthy-musty odor
Geosmin can be produced by various cyanobacteria species production in aquatic environment. Moreover, it is still not
such as Dolichospermum (synonym Anabaena), Oscillatoria, clear how N affects geosmin synthesis in cyanobacteria from
Lyngbya, and Symploca (Smith et al. 2008). Dolichospermum the viewpoint of geoA-mRNA expression. In this study, we
is known as a common geosmin producer in eutrophic lakes investigated the effects of elevated N on the growth, geosmin
and reservoirs (Kutovaya and Watson 2014). Previous studies production, and geoA gene expression of a geosmin-
have suggested that environmental factors affected producing cyanobacterium, Dolichospermum smithii NIES-
cyanobacterial growth, and geosmin/2-MIB production in- 824 (synonym A. smithii NIES-824). The relationships among
clude nutrient concentration and their ratio (Smith 2003; the growth, geosmin production, and gene expression were
Olsen et al. 2016; Perkins et al. 2019), temperature, and light also discussed.
(Zhang et al. 2009; Kakimoto et al. 2014; Wang and Li 2015;
Cai et al. 2017; Alghanmi et al. 2018). The results from
geosmin/2-MIB production and the intra/extracellular propor- Materials and methods
tion can help to elucidate the relationship between earthy-
musty odor production and other metabolic activities, such Cyanobacterial strains and culture conditions
as photosynthesis. Usually chlorophyll a (Chl-a) can represent
the growth of cyanobacteria (such as increase of Chl-a indi- D. smithii NIES-824, a known geosmin producer, was obtain-
cates improved growth of cyanobacteria). Several researches ed from the Microbial Culture Collection at the National
have investigated the possible relationships between geosmin Institute for Environmental Studies (Japan). This strain was
production and isoprenoid synthesis using Chl-a as an indica- cultured at 28 ± 0.2 °C under an illumination of 60 ± 5 μmol
tor. Blevins et al. (1995) suggested that Chl-a concentration photons m−2 s−1 with a 12:12 light/dark cycle in CT liquid
might decrease at a high light intensity, which triggered more medium (150 mg Ca(NO3)2·4H2O, 100 mg KNO3, 50 mg β-
geosmin production, while the increase in temperature can Na2glycerophosphate·5H2O, 400 mg TAPS, 40 mg MgSO4·
stimulate the increase of Chl-a concentration at the expense 7H2O, 0.1 μg vitamin B12, 0.1 μg biotin, 10 μg thiamine HCl,
of geosmin synthesis. An inverse relationship between Chl-a 3 mL P IV metals per liter; P IV metals contain 100 mg
and geosmin production has been also reported by Saadoun Na2EDTA·2H2O, 19.6 mg FeCl3·6H2O, 3.6 mg MnCl2·
et al. (2001) and Wang and Li (2015). 4H 2O, 2.2 mg ZnSO4·7H 2O, 0.4 mg CoCl 2·6H 2O, and
In addition, the discovery of the synthase of geosmin and 2- 0.25 mg Na 2MoO 4·2H 2O per 100 mL) (Watanabe and
MIB in cyanobacteria has provided the basis for the develop- Hiroki 1997).
ment of detection and quantification strategies (Jüttner and
Watson 2007). Geosmin is synthesized using the cyclisation Experimental conditions and procedures
of farnesyl diphosphate (FPP) to geosmin by a bifunctional
terpene synthase gene called geoA encoding geosmin synthase The cultures of D. smithii NIES-824 at exponential growth
through the isoprenoid pathway (Watson et al. 2016). phase were inoculated for the different treatments. The origi-
Although the pathway of geosmin production has been re- nal CT medium was used as the normal N condition (control
vealed in actinomycetes (Jiang et al. 2006; Jiang and Cane group, TN/TP ratio of 8.1, calculated according to the concen-
2008), the trigger of its production under environmental con- trations of TN and TP in the medium, by mass). It is well-
ditions remains unclear. The knowledge of the trigger to pro- known that the TN/TP ratio in eutrophic water is generally
duce geosmin/2-MIB can contribute to the forecast of greater than 10 dues to high N concentration relative to P.
geosmin/2-MIB incident, playing an important role in the To simulate the elevated N loading and normal TN/TP ratio
maintenance of water treatment capacity, especially the usage (> 10) in eutrophic water bodies, the modified CT medium
Environ Sci Pollut Res

(TN/TP ratio of 13.1, by mass) by adding twice amount of Scientific ISQ Single Quadrupole GC-MS, 73 eV EI, positive
nitrate-N sources (Ca(NO3)2·4H2O and KNO3) was used in ion mode; 30 m × 0.25 mm Rxi-5 ms column). Temperature
this study. A cyanobacterial suspension (3 mL) of D. smithii program was started with hold time for 4 min under 60 °C and
NIES-824 was inoculated into 500-mL flasks containing 200 then the temperature was raised at a temperature gradient of 15
mL of the correspondingly CT medium with different N con- °C/min to 230 °C; after reaching 230 °C, the temperature was
centrations, respectively; and the final Chl-a concentration in held for 3 min. Helium gas was used as the carrier gas at a
each flask was set as approximately 19 μg/L. Cultures were constant flow rate of 1.0 mL/min. Different concentrations of
incubated at the conditions mentioned previously. All the ex- standard geosmin solution (0.05–10 μg/L) were diluted with
periments were carried out in quintuplicate. Experimental cul- methanol and used to verify the GC-MS analysis results, and
tures were sampled every 4 days during the whole test cycle then the geosmin concentrations were determined using an
(28 days) to analyze cell growth, geosmin production, and internal standard method with the diluted 2-isopropyl-3-
gene expression. methoxypyrazine (IPMP) in methanol as an internal standard
(0.5 μg/L). Both sample and standard solutions were prepared
Measurements of cell growth and some water with IPMP.
parameters
RNA extraction and reverse transcription
The growth of D. smithii NIES-824 was estimated by measur-
ing the biomass in terms of Chl-a concentration according to Cyanobacteria cells were collected into 2-mL collection tubes;
the modified method described by Suwanpakdee et al. (2016). then, the samples were centrifuged to discard culture media,
The cultures of 50 mL were collected and filtered using GF/C completely. Total RNA was subsequently extracted using the
glass microfiber filter (Whatman, USA). Chl-a was extracted NucleoSpin® RNA kit according to the manufacturer’s in-
from each sample with 10 mL of 90% methanol containing structions. RNA integrity was examined by electrophoresis
0.2% MgCl2 at 70 °C obtained by an incubator for 30 min in (1% agarose) and spectrophotometry. RNA samples were
the dark. The Chl-a concentration was determined at the quantified using the NanoVue Plus spectrophotometry (GE
wavelengths of 750, 665, 645, and 630 nm by a spectropho- Healthcare, USA) and Qubit Fluorometer (Invitrogen,
tometer, and calculated using the following formula: Carlsbad, CA, USA). RT-PCR was done using 50 ng of total
RNA to synthesize cDNA using the PrimeScript RT reagent
Chl−a ðmg=LÞ ¼ Kit (Takara Bio, Japan) with 10 μL reaction mix according to
the manufacturer’s manual. The mixture containing 1× buffer,
ð11:6  ðA665 −A750 Þ−1:31  ðA645 −A750 Þ−0:14  ðA630 −A750 ÞÞ
0.5 μL enzyme mix, and 50 pmol random 6 mers was incu-
 V m ðmLÞ=V f ðmLÞ bated at 37 °C for 15 min and heated to 85 °C for 5 s to stop
reaction. The cDNA samples were stored at − 20 °C.
where Vm and Vf are the volume of the methanol and filtered
culture, respectively.TN and TP were measured according to Standards for qPCR analysis and real-time RT-PCR
the standard method (APHA 2012). protocols

Geosmin extraction and analysis PCR amplification of the geoA fragments (2231 bp) and 16S
rRNA gene fragments (1468 bp) of D. smithii NIES-824 was
Geosmin was extracted and concentrated from the culture performed for sub-cloning to pUC-19 vector using In-Fusion
samples using solid-phase extraction (SPE) according to the HD cloning kit (Takara Bio Inc., Shiga, Japan) according to
standard methods (Japan Water Works Association, 2011). manufacturer’s protocol for qPCR standards. The geoA frag-
The sampled cultures of 100 mL were added 3 mL sodium ments were amplified using the primer sets Inf19geoA-1F (5′-
hypochlorite and then shaken violently to disrupt TCGAGCTCGGTACCCATGCAACCATTTAAACT
cyanobacteria cells. This homogeneous mixture of cell lysates GCCAGC-3′) and Inf19geoA-2231R (5′-GCAG
was used for the determination of total amount of geosmin. GTCGACTCTAGACCAAGGATCTGATATGTGCA-GC-
After filtration by GF/C glass microfiber filters (Whatman, 3′). The 16S rRNA gene fragments were amplified using
USA), the 50-mL filtrated culture samples were used for the primer sets Inf19-27F (5′-TCGAGCTCGGTACCCAGAGT
determination of extracellular geosmin. The intracellular TTGATCCTGGCTCAG-3′) and Inf19-1494Rc (5′-GCAG
geosmin concentration was obtained by subtracting the extra- GTCGACTCTAGTACGGCTACCTTGTTACGAC-3′).
cellular geosmin from the total amount of geosmin. All sam- The qPCR using an ABI 7500 system (Applied
ples extracted and concentrated by the SPE method were Biosystems, USA) with SYBR Green assay was performed
stored in sealed vials at -30 °C and then analyzed using a to investigate the expression levels of the target gene. The
GC-MS with the detection limit of 1 ng/L (Thermo housekeeping gene, 16S rRNA gene, was selected as the
 Environ Sci Pollut Res

control to normalize the expression of target gene. A primer geosmin concentration under the higher N condition were
set, geoA666F (5′-AAAAGACACATTTGCTGATGGTG- significantly lower than those under the control group from
3′) and geoA774R (5′-ATCACGCGGTCATCAGGCTT-3′) day 8 to day 28 during the cultivation (p < 0.05, Fig. 1b and d).
designed in this study was used to amplify the geosmin syn- The extracellular geosmin concentration was also lower under
thase gene, and a primer set for 16S rRNA gene was16S-RTF the higher N condition, but there was no significant difference
(5′-ACGGAGTTAGCCGATGCTTATTC-3′) and 16S-RTR from day 12 to day 28 (Fig. 1c). In addition, the extracellular
(5′-CGAAAGCCTGACG-GAGCAATA-3′) according to a geosmin concentration under the control group on day 4 was
previous report (Kakimoto et al., 2014). The 20 μL qPCR below the detection limit. The extracellular geosmin concen-
reaction mixture consisted of 0.4 μM of each primer, 10 μL tration ranged from 1.4 to 23.6 ng/L and 0 to 19.3 ng/L under
of hot start SYBR Premix Ex Taq II (Takara Bio, Japan), 0.4 the normal and higher N condition, respectively; and the in-
μL of ROX Reference Dye II, 1 μL cDNA as template, and 7 tracellular counterpart changed from 7.5 to 475.5 ng/L and 9.2
μL of ddH2O. The cDNA was diluted for 100 times for the to 343.5 ng/L, respectively (Fig. 1c and d). The proportion of
amplification of geoA and 16S rRNA gene. The PCR program intracellular geosmin (> 90%) was much higher than that of
was set as follows: preheating at 95 °C for 30 s, followed by extracellular geosmin (< 10%) under both two N conditions
40 cycles of 95 °C for 5 s and 60 °C for 34 s, respectively. The during the cultivation. Therefore, D. smithii NIES-824 pro-
following linear relationships between cycle threshold (CT) duces geosmin and mainly keep it in its intracellular part.
and the log of the gene copies were obtained: CT = − 3.361 The geosmin productivity (geosmin/Chl-a) under different
× + 40.055 (R2 = 0.9994) with an efficiency of 98.4% for N concentrations is shown in Fig. 2. Obviously, lower total
geoA qPCR assay and CT = − 3.1756 × + 37.838 (R2 = and intracellular geosmin productivities were obtained under
0.9991) with an efficiency of 106.5% for 16S rRNA qPCR the higher N condition, and significant difference between
assay, where x is the log of geoA and 16S rRNA gene copies these two scenarios was noticed after day 12 (p < 0.05, Fig.
respectively. Three replicates were conducted for each 2a and c). The extracellular geosmin productivity significantly
sample. decreased under the higher N condition except on day 12 and
day 20 (p < 0.05, Fig. 2b).
Statistical analysis
Effects of elevated nitrogen on the gene expression
All the data were presented as means ± standard deviation relating to geosmin synthesis
(SD). Since the Welch test is recommended when the vari-
ances are unequal and skewness values are less than 3 (Rasch The transcriptional response of geoA gene in D. smithii NIES-
et al., 2011), the Welch test was conducted for statistical anal- 824 under the test N concentrations was quantified using qRT-
ysis using the statistical analysis software “R”. The value of p PCR to know the relationship between geosmin productivity
< 0.05 was considered significantly different. and geoA expression level (Fig. 3). Under the higher N con-
dition, the transcription of geoA gene gradually decreased
during the growth period in the cells (4–16 days), and the
Results value was significantly different from the control group except
on day 12 (p < 0.05). The largest difference in the transcription
Effects of elevated nitrogen on cyanobacterial growth of geoA gene between two tested N concentrations occurred
and geosmin production on day 4 and day 16. In addition, both geoA expressions were
almost stably transcribed, resulting in the increase of total
The growth rate and geosmin concentration under different N geosmin concentration.
concentrations are shown in Fig. 1. There was no significant
difference in Chl-a concentration under the two N conditions
before day 12. The highest Chl-a concentration was detected Discussion
on day 16 (1.84 mg/L at normal N condition and 2.58 mg/L at
higher N condition), whereas Chl-a concentration decreased Relationship between Chl-a and geosmin concentra-
after the 16th day indicating cells entered the decline phase tion, and effects of nitrogen concentration on
caused by the possible limitation of P and/or accumulation of geosmin production
metabolites in the medium. As shown, a significant difference
in Chl-a content was noticed between the two scenarios after In this study, we investigated the effects of elevated N con-
day 12, in which much more Chl-a content was obtained at the centration on growth, geosmin production, and geoA gene
higher N condition (p < 0.05, Fig. 1a). Regarding the growth expression in D. smithii NIES-824. The present results sug-
rate, an inverse effect on the geosmin production was ob- gested a competitive relationship between Chl-a concentration
served (Fig. 1b–d). Both total geosmin and intracellular and geosmin production in geosmin-producing D. smithii
Environ Sci Pollut Res

Fig. 1 Effects of elevated N on

the growth (a), total geosmin (b),
extracellular geosmin (c), and
intracellular geosmin
concentrations (d) in the culture
of D. smithii

NIES-824. This observation was consistent with previous the transformation of germacradienol to geosmin (Jiang
findings (Saadoun et al. 2001; Giglio et al. 2011; Wang and et al. 2006; Jiang and Cane 2008).
Li 2015; Zhang et al. 2017). We found that elevated N con- In this study, the presumptive correlation between
centration promoted more Chl-a production, whereas the Chl-a and geosmin production (more Chl-a synthesis
geosmin synthesis decreased compared with the lower N con- with lower geosmin production) in cyanobacteria im-
centration (control group). It was suggested that the decrease plies a competitive relationship between the synthesis
of N availability may stimulate cyanobacteria to produce more of chlorophyll and geosmin. Naes and Post (1988) sug-
geosmin. Saadoun et al. (2001) suggested that increasing gested geosmin and the hydrophobic phytol tail of chlo-
nitrate-N concentrations promoted more Chl-a production rophyll are synthesized through the same isoprenoid
with geosmin synthesis decreased, and low nitrate-N favored pathway. Most evidence indicates that the isoprenoid
maximum geosmin production. Zhang et al. (2017) also found pathway can be used to produce phytol and carotenoids
higher geosmin productivity (geosmin/Chl-a) which was ob- in the cyanobacterial thylakoid membrane (Lange et al.
served at lower N concentration. A recent study by Perkins 2000; Ajikumar et al. 2008). The intermediate precursor
et al. (2019) also showed the decrease of TN: TP ratio caused FPP related to chlorophyll synthesis is also the imme-
by the decrease of nitrate concentration may promote the pro- diate precursor of geosmin. Combining the previous re-
duction of geosmin and 2-MIB in drinking water reservoirs. search works with the results from the present study,
the inverse correlation between Chl-a concentration and
Pathways of geosmin production geosmin production is most probably brought about by
competition for the essential common substrates. In this
Geosmin is a sesquiterpene which belongs to terpenoids, study, geosmin production was significantly higher at a
and its biosynthetic pathway has been thoroughly studied lower N condition (control group) while with a lower
in Streptomyces spp. The isopentenyl diphosphate (IPP) Chl-a content. When photosynthesis activity decreases,
is the universal precursors in the metabolic pathway for the precursor substance, FPP, may be used more to
geosmin, 2-MIB, and other terpenoids. It has been pro- synthesize geosmin than chlorophyll (Zimba et al.
posed that the MEP pathway is the major biosynthetic 1999). Geosmin concentration in surface waters is sig-
isoprenoid route in cyanobacteria (Proteau 1998; nificantly affected by the TN/TP ratio, and inorganic P
Ajikumar et al. 2008; Pattanaik and Lindberg 2015). might be the primary determinant of geosmin production
The geosmin synthase of Streptomyces coelicolor A3(2) (Dzialowski et al. 2009). Generally, direct monitoring of
is a Mg 2+ -dependent bifunctional enzyme. The N- the earthy-musty odor concentration and population of
terminal domain of the protein converts FPP to cyanobacterial producer can be a valuable means to pre-
germacradienol, while the C-terminal domain catalyzes dict and control the earthy-musty odor problems.
 Environ Sci Pollut Res

Fig. 3 Changes in expression of geoA gene (normalized against the16S

rRNA gene) at different N concentrations during 28 days’ cultivation of
D. smithii

concentration since geosmin was continuously produced un-

der the two N conditions (Fig. 1 and Fig. 3). Light condition is
also an important factor for the growth of cyanobacteria.
Giglio et al. (2011) investigated the geoA expression in
A. circinalis AWQC318 under different light conditions.
Their results showed that the expression of geoA gene was
constitutive under the light condition, which was also relating
to the growth stage and cell activity. Their results suggested a
decrease in isoprenoid synthesis may occur before a decrease
in the transcription of ribosomal units when cells enter a de-
cline phase. In this study, geoA expression level was lower
during the exponential growth phase of cyanobacterial cells at
the elevated N condition. Furthermore, the decrease of geoA
expression during the decline phase suggested that geoA tran-
scription was closely related to cell activity and isoprenoid
productivity (Fig. 3).

Concerns about the efficient removal of intracellular

and extracellular geosmin

The intracellular and extracellular geosmin and 2-MIB in sur-

Fig. 2 Effects of elevated N on the total geosmin (a), extracellular face water and their variations under different environmental
geosmin (b), and intracellular geosmin productivity (c) of D. smithii factors should be of concern more, because they can affect the
treatment efficiency and removal by the whole treatment pro-
Geosmin production and geoA gene expression in cesses, which is the key to earthy-musty odor management
D. smithii NIES-824 and treatment. The localization of geosmin at intracellular
and extracellular varies among different species, growth
PCR and qPCR methods used to target the genes coding for phases, and in response to different environmental conditions.
geosmin and 2-MIB synthase are useful to identify geosmin/2- In this study, most geosmin existed at the intracellular part of
MIB-producing cyanobacteria and investigate the relationship D. smithii NIES-824, including the decline phase, which is
between geosmin/2-MIB production and gene expression consistent with the findings from previous reports. The pro-
(Lee et al. 2017). However, few studies have analyzed the portion of intracellular geosmin was greater than 90% in this
expression of geosmin synthase in cyanobacteria. In the pres- study. Li et al. (2010) found that more than 85% of geosmin
ent study, the transcription of geoA gene was almost stably production was inside the cells in A. spiroides. Previous stud-
transcribed, resulting in the increase in total geosmin ies also suggest that a large portion of these odorous
Environ Sci Pollut Res

Fig. 4 The summary of the

findings on this study

compounds usually retain in the cells, while a small portion is geosmin (> 90%) was retained intracellularly during the 28
released to the culture medium. Moreover, these earthy-musty days of cultivation. The geoA-gene expression levels were
odor compounds would be released to the medium during the constitutive and decreased at the higher N concentration dur-
decline phase of cells, and their full release occurs when the ing the exponential growth phase of cyanobacterial cells.
cell dies and/or lyses (Saadoun et al. 2001; Alghanmi et al. Moreover, the expression of geoA gene gradually decreased
2018). Geosmin and 2-MIB can be successfully removed via along with the cultivation. Future analysis of other environ-
ozonation, activated carbon filtration, and biological sand fil- mental factors at the gene and metabolite levels is required to
tration during drinking water treatment (Ho et al. 2007; Hsieh understand deeper and to develop the forecast of earthy-musty
et al. 2010). However, since most of the water treatment plants odor incident using the mechanisms of geosmin biosynthesis
do not implement these facilities, powdered activated carbon in cyanobacteria.
can be used to remove the odor in the first sedimentation tank,
and cells are removed through the subsequent filtration facil- Funding information This study was supported by a Health Labor
Sciences Research Grant (H30-Kenki-Ippan-004) from the Ministry of
ities to meet the regulated water quality standards. Thus, the
Health, Labor and Welfare, Japan.
localization of geosmin and 2-MIB is important for the treat-
ment plants because the extracellular (dissolved) geosmin and
2-MIB are easy to remove using powdered activated carbon in
References
the first sedimentation tank, and the cells with intracellular
geosmin and 2-MIB are removed by both the sedimentation Ajikumar PK, Tyo K, Carlsen S, Mucha O, Phon TH, Stephanopoulos G
and filtration facilities. Blind use of disinfectants and oxidants (2008) Terpenoids: opportunities for biosynthesis of natural product
may cause more odorous substances to be released from lysed drugs using engineered microorganisms. Mol Pharm 5(2):167–190
odor-producing cyanobacterial cells. It is very important to Alghanmi HA, Fo’ad MA, Al-Taee MM (2018) Effect of light and tem-
perature on new cyanobacteria producers for geosmin and 2-
optimize the water treatment processes and scheme of the methylisoborneol. J Appl Phycol 30(1):319–328
whole plant for off-flavor removal. APHA (2012) Standard methods for the examination of water and waste-
The above results could shed light on the effects of elevated water, 22nd edn. American Public Health Association, Washington
N on the geoA expression and growth of cyanobacteria (Fig. DC
Blevins WT, Schrader KK, Saadoun I (1995) Comparative physiology of
4), which can give an important and fundamental knowledge geosmin production by Streptomyces halstedii and Anabaena sp.
of geosmin production and contribute to the development of Water Sci Technol 31(11):127–133
earthy-musty odor forecast in aquatic environment. Further Cai F, Yu G, Zhang K, Chen Y, Li Q, Yang Y, Xie J, Wang Y, Li R
studies are still needed to clarify the global regulation on (2017) Geosmin production and polyphasic characterization of
Oscillatoria limosa Agardh ex Gomont isolated from the open canal
geosmin production in the real aquatic environment for
of a large drinking water system in Tianjin City, China. Harmful
earthy-musty odor forecast. Algae 69:28–37
Cook D, Newcombe G, Sztajnbok P (2001) The application of powdered
activated carbon for MIB and geosmin removal: predicting PAC
doses in four raw waters. Water Res 35(5):1325–1333
Conclusions
Dolman AM, Rücker J, Pick FR, Fastner J, Rohrlack T, Mischke U,
Wiedner C (2012) Cyanobacteria and cyanotoxins: the influence
This study investigated the effects of elevated N on the of nitrogen versus phosphorus. PLoS One 7(6):e38757
growth, geosmin production, and gene expression in Dzialowski AR, Smith VH, Huggins DG, deNoyelles F, Lim NC, Baker
D. smithii NIES-824. Elevated N concentration promoted DS, Beury JH (2009) Development of predictive models for
geosmin-related taste and odor in Kansas, USA, drinking water res-
Chl-a production, whereas the geosmin synthesis decreased. ervoirs. Water Res 43(11):2829–2840
The competitive correlation was revealed between Chl-a con- Elser JJ, Bracken ME, Cleland EE, Gruner DS, Harpole WS, Hillebrand
tent and geosmin production in D. smithii NIES-824. Most of H, Ngai JT, Seabloom EW, Shurin JB, Smith JE (2007) Global
 Environ Sci Pollut Res

analysis of nitrogen and phosphorus limitation of primary producers water reservoirs: the importance of ammonium as a key nutrient
in freshwater, marine and terrestrial ecosystems. Ecol Lett 10(12): trigger. J Environ Manag 244:276–284
1135–1142 Proteau PJ (1998) Biosynthesis of phytol in the cyanobacterium
Frank D, Poole S, Kirchhoff S, Forde C (2009) Investigation of sensory Synechocystis sp. UTEX 2470: utilization of the non-mevalonate
and volatile characteristics of farmed and wild barramundi (Lates pathway. J Nat Prod 61(6):841–843
calcarifer) using gas chromatography−olfactometry mass spectrom- Rasch D, Kubinger KD, Moder K (2011) The two-sample t test: pre-
etry and descriptive sensory analysis. J Agric Food Chem 57(21): testing its assumptions does not play off. Stat Pap 52:219–231
10302–10312 Robin J, Cravedi JP, Hillenweck A, Deshayes C, Vallod D (2006) Off
Giglio S, Jiang J, Saint CP, Cane DE, Monis PT (2008) Isolation and flavor characterization and origin in French trout farming.
characterization of the gene associated with geosmin production in Aquaculture 260(1-4):128–138
cyanobacteria. Environ Sci Technol 42(21):8027–8032 Saadoun IM, Schrader KK, Blevins WT (2001) Environmental and nu-
Giglio S, Saint CP, Monis PT (2011) Expression of the geosmin synthase tritional factors affecting geosmin synthesis by Anabaena sp. Water
gene in the cyanobacterium Anabaena circinalis AWQC3181. J Res 35(5):1209–1218
Phycol 47(6):1338–1343
Smith JL, Boyer GL, Zimba PV (2008) A review of cyanobacterial odor-
Ho L, Hoefel D, Bock F, Saint CP, Newcombe G (2007) Biodegradation
ous and bioactive metabolites: impacts and management alternatives
rates of 2-methylisoborneol (MIB) and geosmin through sand filters
in aquaculture. Aquaculture 280(1-4):5–20
and in bioreactors. Chemosphere 66(11):2210–2218
Smith VH (2003) Eutrophication of freshwater and coastal marine eco-
Hsieh ST, Lin TF, Wang GS (2010) Biodegradation of MIB and geosmin
systems a global problem. Environ Sci Pollut Res 10(2):126–139
with slow sand filters. J Environ Sci Heal A 45(8):951–957
Japan Water Works Association (2011) Standard methods for the exam- Smith VH, Schindler DW (2009) Eutrophication science: where do we go
ination of water. Japan Water Works Association, Tokyo, Japan (in from here? Trends Ecol Evol 24(4):201–207
Japanese) Srinivasan R, Sorial GA (2011) Treatment of taste and odor causing
Jiang J, Cane DE (2008) Geosmin biosynthesis. Mechanism of the frag- compounds 2-methyl isoborneol and geosmin in drinking water: a
mentation−rearrangement in the conversion of germacradienol to critical review. J Environ Sci 23(1):1–13
geosmin. J Am Chem Soc 130(2):428–429 Suwanpakdee S, Gutierrez R, Pithakpol S, Jampeetong A, Pathom-aree
Jiang J, He X, Cane DE (2006) Geosmin biosynthesis. Streptomyces W, Nomura N, Itayama N, Whangchai N (2016) Earthy-musty odor
coelicolor germacradienol/germacrene D synthase converts farnesyl and off-flavour taints in Phayao Lake, Thailand. Chiang Mai J Sci
diphosphate to geosmin. J Am Chem Soc 128(25):8128–8129 43(1):22–31
Jüttner F, Watson SB (2007) Biochemical and ecological control of Tsao HW, Michinaka A, Yen HK, Giglio S, Hobson P, Monis P, Lin TF
geosmin and 2-methylisoborneol in source waters. Appl Environ (2014) Monitoring of geosmin producing Anabaena circinalis using
Microbiol 73(14):4395–4406 quantitative PCR. Water Res 49:416–425
Kakimoto M, Ishikawa T, Miyagi A, Saito K, Miyazaki M, Asaeda T, Wang Z, Li R (2015) Effects of light and temperature on the odor pro-
Yamaguchi M, Uchiyama H, Kawai-Yamada M (2014) Culture duction of 2-methylisoborneol-producing Pseudanabaena sp. and
temperature affects gene expression and metabolic pathways in the geosmin-producing Anabaena ucrainica (cyanobacteria). Biochem
2-methylisoborneol-producing cyanobacterium Pseudanabaena Syst Ecol 58:219–226
galeata. J Plant Physiol 171(3-4):292–300 Wang Z, Xu Y, Shao J, Wang J, Li R (2011) Genes associated with 2-
Kutovaya OA, Watson SB (2014) Development and application of a methylisoborneol biosynthesis in cyanobacteria: isolation, charac-
molecular assay to detect and monitor geosmin-producing terization, and expression in response to light. PLoS One 6(4):
cyanobacteria and actinomycetes in the Great Lakes. J Great Lakes e18665
Res 40(2):404–414 Watanabe MM, Hiroki M (1997) NIES-Collection. List of strains, algae
Lange BM, Rujan T, Martin W, Croteau R (2000) Isoprenoid biosynthe- and protozoa. National Institute for Environmental Studies,
sis: the evolution of two ancient and distinct pathways across ge- Environment Agency, Japan, 140.
nomes. Proc Natl Acad Sci U S A 97(24):13172–13177 Watson SB, Monis P, Baker P, Giglio S (2016) Biochemistry and genet-
Lee J, Rai PK, Jeon YJ, Kim KH, Kwon EE (2017) The role of algae and ics of taste-and odor-producing cyanobacteria. Harmful Algae 54:
cyanobacteria in the production and release of odorants in water. 112–127
Environ Pollut 227:252–262 Westerhoff P, Rodriguez-Hernandez M, Baker L, Sommerfeld M (2005)
Li Z, Yu J, Yang M, Zhang J, Burch MD, Han W (2010) Cyanobacterial Seasonal occurrence and degradation of 2-methylisoborneol in wa-
population and harmful metabolites dynamics during a bloom in ter supply reservoirs. Water Res 39(20):4899–4912
Yanghe Reservoir, North China. Harmful Algae 9(5):481–488
Zhang T, Li L, Song L, Chen W (2009) Effects of temperature and light
Naes H, Post AF (1988) Transient states of geosmin, pigments, carbohy-
on the growth and geosmin production of Lyngbya kuetzingii
drates and proteins in continuous cultures of Oscillatoria brevis
(Cyanophyta). J Appl Phycol 21(3):279–285
induced by changes in nitrogen supply. Arch Microbiol 150(4):
Zhang T, Li L, Zheng L, Song L (2017) Effects of nutritional factors on
333–337
the geosmin production of Lyngbya kuetzingii UTEX 1547
Olsen BK, Chislock MF, Wilson AE (2016) Eutrophication mediates a
(Oscillatoriales, Cyanobacteria). Phycologia 56(2):221–229
common off-flavor compound, 2-methylisoborneol, in a drinking
water reservoir. Water Res 92:228–234 Zimba PV, Dionigi CP, Millie DF (1999) Evaluating the relationship
Pattanaik B, Lindberg P (2015) Terpenoids and their biosynthesis in between photopigment synthesis and 2-methylisoborneol accumu-
cyanobacteria. Life 5(1):269–293 lation in cyanobacteria. J Phycol 35(6):1422–1429
Perkins RG, Slavin EI, Andrade TMC, Blenkinsopp C, Pearson P,
Froggatt T, Godwin G, Parslow J, Hurley S, Luckwell R, Wain DJ Publisher’s note Springer Nature remains neutral with regard to jurisdic-
(2019) Managing taste and odour metabolite production in drinking tional claims in published maps and institutional affiliations.