This article was downloaded by: [Laurentian University]

On: 02 November 2013, At: 11:19

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Food Additives & Contaminants: Part B: Surveillance

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tfab20

Trimethylamine oxide, dimethylamine, trimethylamine

and formaldehyde levels in main traded fish species in
Hong Kong
a a
S.W.C. Chung & B.T.P. Chan
a
Food Research Laboratory, Food and Environmental Hygiene Department , 4/F Public
Health Laboratory Centre, 382 Nam Cheong Street, Hong Kong, SAR, China
Published online: 10 Nov 2009.

To cite this article: S.W.C. Chung & B.T.P. Chan (2009) Trimethylamine oxide, dimethylamine, trimethylamine and
formaldehyde levels in main traded fish species in Hong Kong, Food Additives & Contaminants: Part B: Surveillance, 2:1,
44-51, DOI: 10.1080/02652030902858921

To link to this article: http://dx.doi.org/10.1080/02652030902858921

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Food Additives and Contaminants: Part B
Vol. 2, No. 1, June 2009, 44–51

View Dataset

Trimethylamine oxide, dimethylamine, trimethylamine and formaldehyde levels in main traded

fish species in Hong Kong
S.W.C. Chung* and B.T.P. Chan
Food Research Laboratory, Food and Environmental Hygiene Department, 4/F Public Health Laboratory Centre, 382 Nam
Cheong Street, Hong Kong SAR, China
(Received 1 October 2008; final version received 2 March 2009)

Levels of trimethylamine oxide (TMAO), dimethylamine (DMA), trimethylamine (TMA) and formaldehyde (FA)
were studied in 266 different fishes, including fresh/frozen raw whole fishes of 89 different species that traded in
Hong Kong, China. Determination of TMAO can confirm the source of DMA and FA if present in the sample.
These samples were purchased from different commercial outlets between April and August 2007. All samples
Downloaded by [Laurentian University] at 11:19 02 November 2013

of raw whole fish were identified for their species by the Agriculture, Fisheries and Conservation Department.
The content of TMAO was determined by high-performance liquid chromatography (HPLC) coupled with a
chemiluminescent nitrogen detector. The possible decomposition products of TMAO, DMA and TMA
were analysed by headspace solid-phase micro-extraction gas chromatography-mass spectrometry (HS-SPME-
GC-MS), while FA was conducted by steam distillation then quantified by a HPLC. The range for TMAO of all
samples was 55–3800 mg kg1 with median of 970 mg kg1, while the endogenous enzymatic cleavage products
DMA, TMA and FA were in the range of 52–320, 51–190 and 51–160 mg kg1, respectively. These cleavage
products were mainly found in three fish species, Harpadon nehereus, Saurida elongata and Saurida tumbil, that
belong to the family Synodontidae (Lizardfishes) and subfamily Harpadontinae. Besides, freshwater fish species,
namely, Micropterus salmoides, Oreochromis niloticus niloticus and Siniperca chuatsi, were found to contain TMAO
in the range of 510–760, 85–720 and 400–640 mg kg1, respectively.
Keywords: chromatography – high-performance liquid chromatography (HPLC); chromatography – headspace;
market basket survey; natural toxicants; fish

Introduction demonstration for such endogenous enzymatic

Trimethylamine oxide (TMAO) constitutes a charac-
teristic and important part of the non-protein nitrogen
fraction in marine species which is responsible for
osmoregulation in marine fish and is found in all
marine fish species in quantities from 1% to 5% of the
muscle tissue (dry weight) but is virtually absent from
freshwater species and from terrestrial organisms
(Anderson and Fellers 1952; Hebard et al. 1982).
Huss (1995) and other scientists reported that dimethy-
lamine (DMA) and formaldehyde (FA) in some species
of fishes originated from the breakdown of TMAO.
During storage, TMAO is demethylated to equimolar
FA and DMA by catalysing with an endogenous
enzyme, trimethylamine oxide aldolase (TMAOase).
The enzymatic reaction occurs even under frozen
storage. However, Nielsen and Jorgensen (2004)
reported enzymatic activity of TMAOase in white
muscle of gadiform fish showed large variations
between species as well as between individuals.
A popular marine fish in Hong Kong called
Bombay duck (Harpadon nehereus) provides good
reaction (Bai 2003; Vong et al. 2005; Zhong et al.
2006). Indeed, the increase of FA level in H. nehereus
was closely related to its DMA as well as TMAO level
since FA is the bi-product generated from the break-
down of TMAO with endogenous enzyme, TMAOase.
Furthermore, during storage of fish of gadoid species
such as Cadous macrocephalus, FA is also formed by
enzymatic cleavage of TMAO present in the fish (Liu
et al. 2005). Since FA is reported to be a carcinogenic
substance by the media, the natural occurrence of FA
in some fish species was concerned by general public.
FA may also be formed during ageing and
deterioration of the flesh of marine fish and crusta-
ceans, from the enzymatic reduction of TMAO. It has
been shown that the toughening of frozen hake muscle
is correlated to the amount of FA produced, and that
the rate of FA production is greatest at high frozen-
storage temperatures (Gill et al. 1979). However, high
levels do not accumulate in fish tissues due to
subsequent conversion of the FA formed to other
chemical compounds (Tsuda et al. 1988). FA formed in

*Corresponding author. Email: swcchung@fehd.gov.hk

ISSN 1939–3210 print/ISSN 1939–3229 online

ß 2009 Taylor & Francis
DOI: 10.1080/02652030902858921
http://www.informaworld.com

fish reacts with protein and subsequently causes muscle
toughness, which suggests that fish containing
the highest levels of formaldehyde (for example,
10–20 mg kg1) may not be considered palatable as
a human food source (Yasuhara and Shibamoto 1995).
Since TMA is a pungent volatile amine, it is often
associated with the typical ‘fishy’ odour of spoiling
seafood. Its presence in spoiling fish is due to the
bacterial reduction of TMAO which is naturally
present in the living tissue of many marine fish species.
However, the indictor of seafood’s freshness would rely
on the total volatile basic nitrogen instead of just
TMA. The European Union has set up an indicator
value of 25–35 mg of nitrogen per 100 g of flesh for
fitness for consumption of different unprocessed
fishery products in Directive 95/149/EC.
Downloaded by [Laurentian University] at 11:19 02 November 2013
Though a number of studies on the background
level of FA have been published by different scientists
in Mainland China (An et al. 2005; Ma et al. 2004;
Shentu et al. 2006), a comprehensive study on the
relationship of TMAO, DMA, TMA and FA in
different fish species was not observed. In Hong
Kong under the Food Surveillance Programme,
individual data on the levels of TMAO, DMA, TMA
and FA in local fish at the species level were also not
available currently. In view of the above, it was decided
to investigate the levels of TMAO, DMA, TMA and
FA in the different species of edible fish commonly
consumed by the local population.
Materials and methods
Samples
Fish species analysed were confined to those com-
monly consumed in Hong Kong where whole fish was
available for species identification during the sampling
period of April–August of 2007. Fish species com-
monly consumed in Hong Kong were sampled
including marine fish and freshwater fish:
. Local fish from wet markets and other fish-
selling premises.
. Imported fish for preparation by restaurant.
The length (which correlates to the age of the fish) and
weight of each fish were recorded when received. The
samples were separated into the edible portion includ-
ing skin and flesh, homogenized, and the extraction/
distillation were conducted immediately. The extracts/
distillates were stored at –20 C, if deemed necessary,
until analysis.
Analytical determination of trimethylamine oxide
The determination of TMAO was based on the method
published in Metrohm’s information issue (Metrohm
2005). A total of 10 g homogenized sample were mixed
Food Additives and Contaminants: Part B 45
and homogenized with 15 ml of 0.05% H3PO4 twice.
The extracts were combined and centrifuged for 20 min
at 8500 rpm at 5 C. The supernatant was then filtered
and made up to 50 ml with 0.05% H3PO4.
The TMAO content was quantified by HPLC with
chemiluminescent nitrogen detection. The separation
was achieved with a Dionex IonPac CS12A column
(250 mm  4 mm  8 mm) using a mobile phase com-
posed of 4 mM aqueous methanesulfonic acid at a flow
rate of 1.25 ml min1.
Analytical determination of dimethylamine and tri-
methylamine (TMA)
The analysis of DMA and TMA were based on the
method developed by Chan et al. (2006). A total of 10 g
homogenized sample were mixed and homogenized
with 15 ml of 0.5 N HCl twice. The extracts were
combined and then filtered. After addition of 0.25 ml
of 1000 mg l1 1-propylamine solution (internal stan-
dard), the mixture was made up to 50 ml with 0.5 N
HCl. Aliquot of 0.5 ml sample solution was pipetted
into a 20 ml SPME vials equipped with a magnetic
stirrer bar, 1 ml of 15 N sodium hydroxide solution was
then added. DMA and TMA were then determined by
headspace solid-phase micro-extraction gas chromato-
graphy-mass spectrometry with mass selective detec-
tion operating in the selective-ion monitoring mode.
In brief, DMA and TMA were extracted onto
carboxen/divinylbenzen/polydimethylsiloxane fibre
(Supelco, Bellefonte, PA, USA) at 35 C for 5 min,
then injected into a gas chromatography-mass spectro-
meter with a PTA-5 column (30 m  0.32 mm internal
diameter, 0.5 mm). The oven was hold at 35 C for
3 min, then ramped to 200 C at a rate of 40 C min–1
and held for another 5 min. The qualifier ion for
DMA, TMA and n-propylamine (internal standard)
were m/z 45, 59 and 58, respectively, while the
corresponding identifier ion were m/z 44, 58 and 59,
respectively.
Analytical determination of formaldehyde
The determination of FA was mainly based on the
method published in the literature (Ma et al. 2005).
The homogenized sample (20 g) was first mixed with
45 g sodium chloride and 40 ml 5% phosphoric acid.
FA was steam distilled out (Food and Agricultural
Organization (FAO) 1997) and then derivatized by 2,4-
dinitrophenylhydrazine in a 60 C water bath for 1 h
before quantification by LC (Ernes 1985). The test
solution was filtered before LC analysis.
The quantification of FA was performed using a
liquid chromatographic system equipped with a visible
detector at 352 nm (Agilent 1100, Santa Clara, CA,
USA). The separation was achieved with an Alltima
C8 column (250 mm  2.1 mm  5 mm) using a mobile
 46 S.W.C. Chung and B.T.P. Chan
phase composed of Methanol/Water (60 : 40) at a flow
rate of 0.2 ml min1.
Results and discussion
A total of 266 whole fish samples (89 species) were
identified for their species and analysed for the levels of
TMAO, DMA and FA. The number of fish species and
the number of samples tested for each species were
limited by the availability of fish during the period of
sampling and available resources. Ideally, more sam-
ples for each species, more fish species and samples in
different seasons could better reflect the levels of target
analytes for each species. The TMAO, DMA and FA
content of these fish samples are summarized in
Table 1.
Downloaded by [Laurentian University] at 11:19 02 November 2013
Since the fishes were purchased from local markets,
not freshly caught, the results may be affected by the
storage period, condition at the point of sale, live/
frozen of the fishes, etc. Therefore, the TMA content
was also quantified for reference of its freshness and
the determined values of TMAO, DMA and FA may
not fully reveal the original level of target analytes if
the TMA level was found to be higher than
1000 mg kg1 (equivalent to 25 mg of nitrogen per
100 g of flesh).
Analytical method performance
Method performance data are summarized in Table 2.
Laboratory and cross-sample contamination was
monitored by analysing procedural blanks. The
average spiking recovery percentages for TMAO,
DMA, TMA and FA were 91%, 89%, 99% and
93%, respectively. Corrections based on recovery
percentages were not applied to the test results.
Trimethylamine oxide in whole fish
This study provided a comprehensive study of TMAO
in different fish species marketed in Hong Kong. For
the content of TMAO in the samples tested, all marine
and diadromous fishes in this study contained a
significant amount, which ranged from tens to several
thousands of mg kg1. However, two diadromous fish
species, Anguilla japonica and Mugil cephalus, were
found to contain TMAO of less than the detection
limit of 5 mg kg1.
Since freshwater organisms are hyperosmtic to
fresh water (they tend to take in water through
osmosis) while marine water organisms are hypertonic
to seawater (they tend to lose water by osmosis), the
osmoregulation mechanism in freshwater fish is
expected be different from that of marine fish.
According to Stroem (1984), it is now generally
believed that TMAO has an osmoregulatory role in
marine fish. Furthermore, Hebard et al. (1982)
described that TMAO is virtually absent from fresh-
water species and from terrestrial organisms. In the
literature there was only one study of Nile perch and
tilapia from Lake Victoria, where as much as
15–20 mg kg1 TMAO of freshwater fish was found
(Gram et al. 1989). In this study, TMAO was not
detected in six out of nine freshwater fish species.
Besides Oreochromis niloticus niloticus (tilapia), two
other freshwater fish species, namely, Micropterus
salmoides and Siniperca chuatsi, were found to contain
TMAO at hundreds of mg kg1. Furthermore, the
existence of TMAO in these freshwater fish was
confirmed by liquid chromatography-mass spectro-
metric analysis.
Regarding the variation of the contents of TMAO
in the same species, most of the fish species were within
30% from its mean. However, notable variation were
still observed in ten species: Cynoglossus bilineatus,
Eleutheronema tetradactylum, Epinephelus lanceolatus,
Harpadon nehereus, Lates calcarifer, Oreochromis
niloticus niloticus, Pampus argenteus, Platycephalus
indicus, Trachinotus blochii and Trichiurus lepturus, in
which the highest sample was at least two-fold more
than the lowest one. Similar findings was also observed
by Hebard et al. (1982) that the amount of TMAO in
the muscle tissue depends on the species, season,
fishing ground, etc.
In general, the highest mean concentration was
found in Epinephelus quoyanus (3200 mg kg1), Beryx
splendens, Cromileptes altivelis, Epinephelus awoara,
Choerodon schoenleinii, Lutjanus malabaricus, Pennahia
argentata, Lutjanus russelli and Plectropomus leopardus
(2500 mg kg1). Other fish species have mean TMAO
content of less than 2500 mg kg1.
Dimethylamine and formaldehyde in whole fish
Amongst the 18 fish samples that FA was detected,
twelve were found with very low level of 55 mg kg1
and FA was detected in less than half of each species.
The remaining six fish samples belonged to three fish
species under the family Synodontidae (Lizardfishes)
and subfamily Harpadontinae, and were found to
contain FA of detectable amount and FA was detected
in each samples of same species. As discussed
previously, Harpadon nehereus could generate FA
during frozen storage. All three Harpadon nehereus
samples were found to contain more than 100 mg kg1
FA. The remaining three fish samples belong to
Saurida elongata and Saurida tumbil with the levels of
5, 7 and 66 mg kg1.
Regarding DMA in fish samples, about half of the
tested samples were found to contain a detectable
amount of DMA. DMA was almost not detected in
freshwater fish samples except for two fish samples of
 Downloaded by [Laurentian University] at 11:19 02 November 2013

Table 1. Overview of fish samples collected in the study: formaldehyde (FA), dimethylamine (DMA), trimethylamine oxide (TMAO) and trimethylamine (TMA) concentrations
(mg kg1) in various fish species.

Number Mean Mean Mean Mean

of TMAO DMA FA TMA
Fish species Common name Habitat samples (range) (range) (range) (range)

Acanthopagrus australis Black bream, Surf bream Marine 2 1400 (1200–1500) 52 (52) 51 (51) 10 (9–12)
Acanthopagrus latus Yellowfin seabream Dia 4 240 (190–310) 52 (52) 51 (51) 18 (3–52)
Acanthopagrus schlegeli Black porgy, Blackhead seabream Marine 1 570 (570) 12 (12) 51 (51) 47 (47)
Anguilla japonica Japanese eel Dia 5 55 (55) 52 (52) 2 (51–4) 52 (52–2)
Ariomma indica Indian ariomma, Indian driftfish Marine 3 340 (170–470) 18 (12–27) 51 (51) 240 (170–320)
Aristichthys nobilis Bighead carp Fresh 3 55 (55) 52 (52) 51 (51–2) 52 (52)
Beryx splendens* Splendid alfonsino Marine 3 3100 (2600–3500) 4 (3–6) 51 (51) 69 (42–97)
Branchiostegus albus White horsehead Marine 4 1700 (1400–2000) 52 (52) 51 (51) 12 (4–19)
Cephalopholis urodeta Darkfin hind Marine 3 1500 (1300–1700) 52 (52–3) 51 (51) 21 (3–57)
Channa maculata Snakehead, Blotched snakehead Fresh 3 55 (55) 52 (52) 51 (51–3) 52 (52)
Choerodon schoenleinii Green wrasse, Blackspot tuskfish Marine 4 2600 (2100–3400) 52 (52–6) 51 (51) 12 (3–30)
Cirrhinus molitorella Mud carp Fresh 3 55 (55) 52 (52) 51 (51) 6 (5–8)
Clarias fuscus Catfish, Hong Kong catfish Fresh 3 55 (55) 52 (52) 51 (51) 52 (52–5)
Cololabis saira* Pacific saury Marine 3 610 (570–660) 14 (6–16) 51 (51) 200 (37–290)
Cromileptes altivelis Humpback grouper Marine 3 2800 (2500–3200) 52 (52) 51 (51) 52 (52–2)
Ctenopharyngodon idellus Grass carp Fresh 3 55 (55) 52 (52) 51 (51) 52 (52)
Cynoglossus arel Largescale tonguesole, Tonguefish Dia 3 1700 (1600–1900) 7 (2–13) 2 (51–4) 400 (29–1000)
Cynoglossus bilineatus Fourlined tonguesole Marine 3 760 (280–1100) 6 (2–12) 51 (51) 340 (19–690)
Dentex tumifrons Golden tail, Yellowback seabream Marine 4 1800 (1400–2600) 7 (4–13) 51 (51) 230 (11–890)
Eleutheronema tetradactylum Fourfinger threadfin, Blind tasselfish Dia 3 660 (320–1300) 6 (2–10) 51 (51) 99 (6–280)
Epinephelus areolatus Areolate grouper, Green-spotted rock cod Marine 3 2200 (1900–2800) 2 (52–5) 51 (51) 92 (8–250)
Epinephelus awoara Yellow grouper, Banded grouper Marine 3 2700 (2500–3000) 5 (52–13) 51 (51) 340 (52–1000)
Epinephelus bleekeri Duskytail grouper Marine 3 1600 (1300–1900) 52 (52) 51 (51) 12 (52–31)
Epinephelus coioides Green grouper, Orange-spotted grouper, Marine 3 1400 (1100–1600) 52 (52) 51 (51) 2 (52–4)
Estuary grouper
Epinephelus fasciatomaculosus Rock grouper, Banded reef-cod Marine 3 2000 (1600–2300) 52 (52–2) 51 (51) 22 (52–62)
Epinephelus hexagonatus Starspotted grouper Marine 2 1700 (1700) 52 (52) 51 (51) 52 (52–2)
Epinephelus lanceolatus Giant grouper Marine 4 1000 (250–2900) 52 (52–4) 51 (51) 5 (52–13)
Epinephelus merra Honeycomb grouper Marine 1 1500 (1500) 52 (52) 51 (51) 52 (52)
Epinephelus quoyanus Longfin grouper Marine 1 3200 (3200) 52 (52) 51 (51) 52 (52)
Epinephelus trimaculatus Threespot grouper Marine 1 1500 (1500) 52 (52) 51 (51) 52 (52)
Gymnothorax favagineus Laced moray Marine 1 1600 (1600) 52 (52) 51 (51) 7 (7)
Gymnothorax reevesii Reeve’s moray Marine 3 1600 (1100–2100) 3 (52–6) 51 (51) 11 (6–21)
Hapalogenys nitens Skewband grunt, Grunt Marine 3 900 (810–970) 52 (52–3) 51 (51) 14 (3–34)
Food Additives and Contaminants: Part B

Harpadon nehereus Bombay duck Marine 3 230 (120–420) 270 (200–320) 150 (110–190) 64 (7–160)
Katsuwonus pelamis* Skipjack tuna Marine 3 160 (130–170) 11 (10–11) 51 (51) 36 (30–39)
47
 Downloaded by [Laurentian University] at 11:19 02 November 2013

Larimichthys croceus Yellow croaker, Croceine croaker, Marine 3 1600 (1400–1800) 52 (52–2) 51 (51) 13 (52–37)
48

Large yellow croaker

Lateolabrax japonicus Japanese seaperch, Common sea bass, Dia 5 290 (180–490) 52 (52–2) 51 (51) 10 (52–48)
Japanese seabass
Lates calcarifer Barramundi Dia 6 1200 (160–1700) 52 (52–10) 51 (51–3) 8 (52–34)
Lethrinus obsoletus Orange-striped emperor Marine 2 2000 (1800–2300) 52 (52) 51 (51) 3 (2–4)
Lutjanus argentimaculatus Mangrove red snapper Dia 5 1300 (840–1700) 7 (52–15) 51 (51) 550 (5–1100)
Lutjanus malabaricus Red snapper, Malabar blood snapper Marine 3 2600 (2100–2700) 8 (5–10) 51 (51–2) 400 (320–680)
Lutjanus russelli Russell’s snapper, fingermark bream Marine 3 2500 (1900–2900) 3 (52–4) 51 (51) 9 (6–11)
Lutjanus stellatus Star snapper Marine 3 1400 (1400) 6 (3–9) 51 (51) 18 (4–42)
Megalobrama terminalis Black amur bream Fresh 1 55 (55) 52 (52) 51 (51) 52 (52)
Micropterus salmoides Large mouth bass, Largemouth black bass Fresh 3 660 (500–760) 52 (52) 51 (51) 52 (52–5)
Mugil cephalus Grey mullet, Flathead grey mullet Dia 3 55 (55) 52 (52–4) 51 (51–2) 7 (3–10)
Mulloidichthys flavolineatus Yellowstripe goatfish Marine 1 1200 (1200) 52 (52) 51 (51) 7 (7)
Nemipterus japonicus Japanese golden thread, Japanese Marine 4 1100 (1000–1300) 4 (52–8) 51 (51) 72 (6–240)
threadfin bream
S.W.C. Chung and B.T.P. Chan

Nemipterus virgatus Golden threadfin bream, Golden thread Marine 3 1400 (1000–1700) 5 (52–14) 51 (51) 310 (12–920)
Oreochromis niloticus niloticus Tilapia, Nile tilapia Fresh 4 370 (85–720) 52 (52–3) 51 (51) 3 (3–4)
Pagrus major Red pargo, Japanese seabream, Red seabream Marine 3 1300 (980–1800) 52 (52) 51 (51) 12 (9–17)
Pampus argenteus Silver pomfret, Butterfish, Pomfret Marine 5 1000 (470–1800) 11 (6–17) 51 (51–1) 260 (9–730)
Pampus nozawae Swallow tail pomfret Marine 3 1200 (970–1600) 4 (3–5) 51 (51) 22 (7–44)
Paralichthys olivaceus* False halibut, Bastard halibut Marine 3 1800 (1600–2100) 52 (52) 51 (51) 52 (52)
Parupeneus barberinus Dash-and-dot goatfish Marine 1 960 (960) 7 (7) 51 (51) 550 (550)
Parupeneus indicus Indian goatfish Marine 2 1000 (720–1300) 9 (6–12) 51 (51) 370 (47–690)
Pennahia argentata White croaker, White Chinese croaker, Marine 5 2600 (2000–3100) 3 (52–15) 51 (51) 250 (2–1200)
Silver croaker
Platycephalus indicus Flathead, Bartail flathead Marine 4 1100 (510–1600) 4 (52–10) 51 (51) 68 (6–200)
Platycephalus spp. Flathead Marine 1 1400 (1400) 52 (52) 51 (51) 16 (16)
Plectorhinchus cinctus Cresent sweetlips, Grunt Marine 3 850 (670–1000) 52 (52) 51 (51) 4 (4)
Plectropomus areolatus Squaretail coralgrouper Marine 2 1900 (1400–2300) 52 (52) 51 (51) 3 (2–3)
Plectropomus leopardus Leopard coralgrouper Marine 4 2500 (2300–2600) 52 (52–4) 51 (51) 78 (52–270)
Pomadasys kaakan Javelin grunter Marine 4 1500 (1100–1800) 3 (52–7) 51 (51) 129 (4–500)
Priacanthus macracanthus Red bigeye, Bulls-eye perch Marine 2 1400 (1400–1800) 7 (5–12) 51 (51) 180 (37–330)
Priacanthus tayenus Purple-spotted bigeye, Big-eye perch Marine 4 1500 (1400–1500) 3 (52–5) 51 (51) 12 (9–16)
Psenopsis anomala Butter fish, Pacific rudderfish Marine 4 1100 (900–1200) 13 (6–18) 51 (51) 230 (15–680)
Pseudocaranx dentex* White trevally Marine 3 340 (260–500) 52 (52) 51 (51) 4 (3–5)
Rachycentron canadum Black bonito, cobia Marine 1 840 (840) 3 (3) 51 (51) 13 (13)
Salmo Salar* Atlantic salmon Dia 3 380 (310–470) 52 (52) 51 (51) 5 (4–6)
Sarda orientalis* Striped bonito Marine 3 180 (150–230) 4 (4–6) 51 (51) 16 (12–22)
Sardinops sagax* South American pilchard Marine 3 570 (160–870) 12 (12–13) 51 (51) 20 (20–21)
Saurida elongata Slender lizardfish Marine 2 710 (500–930) 140 (64–210) 36 (5–66) 73 (39–110)
Saurida tumbil Greater lizardfish Marine 1 280 (280) 180 (180) 7 (7) 380 (380)
Scatophagus argus Spotted scat, Butter fish, spade fish Dia 3 310 (170–440) 52 (52) 51 (51) 26 (6–42)
Scomber japonicus* Chub mackerel Marine 3 400 (280–490) 16 (11–24) 51 (51) 40 (30–50)

( continued)
 Downloaded by [Laurentian University] at 11:19 02 November 2013

Table 1. Continued.

Number Mean Mean Mean Mean

of TMAO DMA FA TMA
Fish species Common name Habitat samples (range) (range) (range) (range)

Scomberomorus commerson Narrow-barred Spanish mackerel, Marine 1 330 (330) 2 (2) 51 (51) 9 (9)
Albacore, Banded tuna
Scomberomorus guttatus Indo-pacific king mackerel Marine 5 700 (400–970) 6 (52–12) 51 (51) 160 (11–440)
Sebastiscus marmoratus Rockfish Marine 3 1300 (1100–1800) 52 (52) 2 (51–5) 8 (7–9)
Seriola dumerili* Purple amberjack, Greater amberjack Marine 3 1100 (950–1200) 52 (52) 51 (51) 10 (9–11)
Seriola lalandi* Yellowtail kingfish, Yellowtail amberjack Marine 6 310 (220–360) 4 (2–6) 51 (51) 14 (6–27)
Siganus canaliculatus Rabbitfish, pearl-spotted spinefoot, Marine 3 540 (340–740) 52 (52–2) 51 (51) 5 (3–9)
white-spotted spinefoot
Sillago japonica Japanese sillago Marine 3 340 (250–390) 5 (52–9) 51 (51) 200 (13–350)
Siniperca chuatsi Freshwater grouper, Mandarin fish Fresh 3 520 (400–640) 4 (52–11) 51 (51) 17 (52–51)
Sphyraena flavicauda* Yellowtail barracuda, Barracudas Marine 3 1700 (1500–1900) 2 (2–3) 51 (51) 13 (10–18)
Trachinotus blochii Snubnose pompano Marine 3 470 (260–860) 52 (52–2) 51 (51–2) 11 (7–17)
Trachurus japonicus* Japanese jack mackerel, Marine 3 2000 (1800–2200) 3 (2–3) 51 (51) 12 (9–15)
Atlantic horse mackerel
Trichiurus lepturus Largehead hairtail, Hairtail Marine 5 1300 (500–2000) 2 (52–3) 51 (51) 37 (5–73)
Trichiurus nanhaiensis Largehead hairtail, South China Sea hairtail Marine 1 1800 (1800) 2 (2) 51 (51) 17 (17)
Variola albimarginata White-edged lyretail Marine 3 2200 (1600–2700) 52 (52) 51 (51) 2 (2–3)

Note: *Imported fish; Fresh: freshwater fish; Marine: marine fish; Dia: diadromous fish; TL: total length; SL: standard length; n.a.: not available.
Food Additives and Contaminants: Part B
49
 50 S.W.C. Chung and B.T.P. Chan

Table 2. Summary of analytical performance data for the tested analytes.

Limit of Limit of Relative standard

detection quantification Recovery, deviation
Analyte Matrix (LOD) (mg kg1) (LOQ) (mg kg1) range (%) (RSDr) (%)

TMAO Fish 5 25 81–104 9% (n ¼ 22)

DMA Fish 2 10 77–105 9% (n ¼ 25)
TMA Fish 2 10 81–123 12% (n ¼ 25)
FA Fish 1 5 81–100 5% (n ¼ 28)

Note: DMA, dimethylamine; FA, formaldehyde; TMA, trimethylamine; TMAO; trimethylamine oxide.

species of Oreochromis niloticus niloticus and Siniperca Conclusion

chuatsi that contained 3 and 11 mg kg1 DMA,
respectively. The results matched well with previous
findings that these fish species contained TMAO.
However, only two of the nine freshwater fish samples
Downloaded by [Laurentian University] at 11:19 02 November 2013
This study provided a comprehensive database of
TMAO, DMA and FA levels of different fish species
marketed in Hong Kong. Amongst the sampled 89
species of fresh/frozen whole fish, most marine and

that contained TMAO were found to contain a
detectable amount of DMA but not FA which
indicated that the reduction of TMAO to DMA and
FA is not favourable.
Among the marine and diadromous fishes, DMA
was not detected in half of these fish samples. The
remaining marine and diadromous fish samples were
found to contain less than 30 mg kg1 DMA except
three Harpadon nehereus, two Saurida elongata and one
Saurida tumbil samples. These six fish samples con-
tained a relatively high level of DMA which ranged
from 64 to 324 mg kg1, which also agrees with the
findings of a relatively high level of FA in these fish
species.
The FA and DMA in Harpadon nehereus have been
repeatedly reported to originate from the breakdown
of TMAO by endogenous enzyme. Based on the above
findings for the levels of FA and DMA in these fishes,
fishes of species of Saurida elongata and Saurida tumbil
might also generate FA under frozen storage.
Therefore, more studies should be conducted to
establish the relationship between these analytes and
the frozen storage time for fish species under the family
Synodontidae and subfamily Harpadontinae.
diadromous fish samples were found to contain
TMAO in the range 55–3500 mg kg1. However, the
contents of TMAO of two diadromous fish species,
Anguilla japonica and Mugil cephalus, were found to be
55 mg kg1. Regarding freshwater fishes, TMAO was
not detected in most fish species except Oreochromis
niloticus niloticus, Micropterus salmoides and Siniperca
chuatsi. The latter two freshwater fish species were
found to contain relatively high level of TMAO in the
range 400–760 mg kg1.
For endogenous enzymatic cleavage products
of TMAO, DMA and FA, they were basically only
detected in three fish species in family of Synodontidae
(Lizardfishes) and subfamily of Harpadontinae. The
levels of DMA and FA in these fish species marketed
in Hong Kong ranged from 64–320 and 5–190 mg kg1,
respectively.
Acknowledgements
Special thanks are made to Ms Valerie C. M. Ho and Dr
Albert W. Y. Leung, Agriculture, Fisheries and Conservation
Department, who identified fish species.

Trimethylamine in whole fish
Around 85% of the tested fish samples were found
to contain a detectable amount of TMA, some even up
to the thousand mg kg1 level. In terms of fish species,
TMA was almost not detected in Anguilla japonica,
Aristichthys nobilis, Channa maculate, Cromileptes
altivelis, Ctenopharyngodon idella, Epinephelus
quoyanus, Epinephelus trimaculatus and Paralichthys
olivaceus, and most of these fish were live when sold. It
was only noted that the content of TMAO in fish was
consistently higher than that of TMA; no other
relationship could be observed between TMA and
TMAO/DMA in this study.
References
An LH, Sun Q, Zheng WY. 2005. Investigation of
formaldehyde background of marine products from East
China Sea. Chin J Food Hyg. 17(6):524–527.
Anderson DW, Fellers CR. 1952. The occurrence of
trimethylamine and trimethylamine oxide in fresh water
fishes. Food Res. 17:472–474.
Bai YL. 2003. Determination of formaldehyde from
Harpadon nehereus. Chin Trop Med. 3(5):670–671.
Chan ST, Yao MWY, Wong YC, Wong T, Mok CS,
Sin DWM. 2006. Evaluation of chemical indicators for
monitoring freshness of food and determination of volatile
amines in fish by headspace solid-phase microextraction
and gas chromatography-mass spectrometry. Eur Food
Res Technol. 224:67–74.

Ernes DA. 1985. Formaldehyde determination in aqueous
systems. Text Chem Color. 17:24–26.
European Commission. 2008. Commission decision 95/
149/ECi [cited 2008 Jan 30]. Available: http://eur-lex.
europa.eu/LexUriServ/LexUriServ.do?uri¼CELEX:31995
D0149:EN:HTML.
Food and Agricultural Organization (FAO). 1997. FAO
Manuals of Food Quality Control Number 14/7, 7 Food
analysis: general techniques, additives, contaminants and
composition. Rome: Food and Agricultural Organization
(FAO). p. 65–69.
Gill TA, Keith RA, Smith Lall B. 1979. Textural deteriora-
tion of red hake and haddock muscle in frozen storage as
related to chemical parameters and changes in myofibrillar
proteins. J Food Sci. 44:661–667.
Gram L, Oundo J, Bon J. 1989. Storage life of Nile perch
(Lates niloticus) dependent on storage temperature and
initial bacteria load. Trop Sci. 29:221–236.
Downloaded by [Laurentian University] at 11:19 02 November 2013
Hebard CE, Flick GJ, Martin RE. 1982. Occurrence and
significance of trimethylamine oxide and its derivatives in
fish and shellfish. In: Martin RE, Flick GJ, Hebard CE,
editors. Chemistry and biochemistry of marine food
products. Westport, CT: AVI. p. 149–304.
Huss HH. 1995. Quality and quality changes in fresh fish.
Rome: Food and Agriculture Organization (FAO); [cited
2008 Jun 30]. Available from: http://www.fao.org/docrep/
V7180E/V7180E00.HTM/.
Liu SF, Du YF, Zhu Wen-hui, Ma JJ, Zhou DQ. 2005.
Initial study on the intrinsic formaldehyde in edible fish.
Mar Fish Res. 26(6):77–82.
Ma JJ, Zhou DQ, Liu SF, Ma HM. 2005. Determination
of formaldehyde in aquatic products by HPLC with
dinitrophenylhydrazine derivation. Mar Fish Res.
26(1):28–31.
Food Additives and Contaminants: Part B 51
Ma JJ, Zhou DQ, Zhang SL. 2004. Research progress on the
intrinsic amount of formaldehyde and mechanisms of
formaldehyde-producing in aquatic products. Mar Fish
Res. 25(4):85–89.
Metrohm. 2005. How fresh is the fish? Ion chromatography
provides the answer! Metrohm information issue 2/2005;
[cited 2008 Jan 31]. Available from: http://www.metrohm.
com/company/metrohminfo/2005/pdf_e/mi_2_2005e_07.pdf/.
Nielsen MK, Jorgensen BM. 2004. Quantitative relationship
between trimethylamine oxide aldolase activity and for-
maldehyde accumulation in white muscle from gadifrom
fish during frozen storage. J Agric Food Chem.
52(12):3814–3822.
Shentu JK, Zhong HY, Zhu LH, Liang QQ, Shu SP. 2006.
Investigation of concentration of formaldehyde in frozen
fresh aquatic products and processed aquatic products.
Chin J Food Hyg. 18(6):549–551.
Stroem AR. 1984. Mikrobiologiske og biokemiskeforhold
ved lagring affisk6. Lecture notes, Tromsoe University,
Tromsoe.
Tsuda M, Frank N, Sato S, Sugimura T. 1988. Marked
increase in the urinary level of N-nitrosothioproline after
ingestion of cod with vegetables. Can Res. 48:4049–4052.
Vong WT, Ung WH, Lei WH, Ngan HL, Sio HM. 2005.
Formaldehyde content of Harpadon nehereus found
in different storage condition. Macao, SAR: Macao
IACM Laboratory; [cited 2008 Jan 31]. Available from:
http://www.iacm.gov.mo/lab/22-Formaldehy_PDF.pdf/
Yasuhara A, Shibamoto T. 1995. Quantitative analysis of
volatile aldehydes formed from various kinds of fish flesh
during heat treatment. J Agri Food Chem. 43:94–97.
Zhong HY, Shentu JK, Liang QQ, Shiu SP, Wang JP. 2006.
Study on content of formaldehyde in Harpadon nehereus.
Chin J Food Hyg. 18(5):441–443.