Science of the Total Environment 408 (2009) 272–276

Contents lists available at ScienceDirect

Science of the Total Environment

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i t o t e n v

What has methylmercury in umbilical cords told us? — Minamata disease

Takashi Yorifuji a,⁎, Saori Kashima a, Toshihide Tsuda b, Masazumi Harada c
a
Department of Epidemiology, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 2-5-1 Shikata-cho, Kita-ku, Okayama, 700-8558, Japan
b
Environmental Epidemiology, Okayama University Graduate School of Environmental Science, 3-1-1 Tsushima-naka, Kita-ku, Okayama, 700-8530, Japan
c
Department of Social Welfare Studies, Kumamoto Gakuen University, 2-5-1 Ooe, Kumamoto, 862-8680, Japan

a r t i c l e i n f o a b s t r a c t

Article history: Severe methylmercury poisoning occurred in Minamata and neighboring communities in the 1950s and 1960s.
Received 23 April 2009 The exposed patients manifested neurological signs, and some patients exposed in utero were born with so-
Received in revised form 18 September 2009 called congenital Minamata disease. In a previous report, Nishigaki and Harada evaluated the methylmercury
Accepted 2 October 2009
concentrations in the umbilical cords of inhabitants and demonstrated that methylmercury actually passed
Available online 29 October 2009
through the placenta (Nishigaki and Harada, 1975). However, the report involved a limited number of cases
(only 35) and did not quantitatively evaluate the regional differences in the transition of methylmercury
Keywords:
Minamata disease
exposure. Therefore, in the present study, we evaluated the temporal and spatial distributions of
Methylmercury poisoning methylmercury concentrations in umbilical cords, with an increased number of participants and additional
Environmental pollution descriptive analyses. Then, we examined whether the methylmercury concentrations corresponded with the
Umbilical cord history of the Minamata disease incident. A total of 278 umbilical cord specimens collected after birth were
obtained from babies born between 1925 and 1980 in four study areas exposed to methylmercury. Then, we
conducted descriptive analyses, and drew scatterplots of the methylmercury concentrations of all the
participants and separated by the areas. In the Minamata area, where the first patient was identified in 1956,
the methylmercury concentration reached a peak around 1955. Subsequently, about 5 years later, the
concentrations peaked in other exposed areas with the expected exposure distribution corresponding with
acetaldehyde production (the origin of methylmercury). This historical incident several decades ago in
Minamata and neighboring communities clearly shows that regional pollution affected the environment in
utero. Furthermore, the temporal and spatial distributions of the methylmercury concentrations in the
umbilical cords tell us the history of the Minamata disease incident.
© 2009 Elsevier B.V. All rights reserved.

1. Introduction famous in a photo collection by Eugene Smith. Nishigaki and Harada

The episode of Minamata disease in the 1950s and 1960s in
Minamata, Japan, is a famous case of food poisoning caused by fish
contaminated with methylmercury (Harada, 1995). The exposed
patients manifested neurological signs, including sensory loss, ataxia,
visual field constriction, dysarthria and hearing difficulties. Further-
more, a considerable number of children with conditions resembling
cerebral palsy were born in the exposed areas (Kitamura et al., 1959).
Although many researchers did not initially believe that methylmer-
cury could pass through the placenta, subsequent epidemiological and
pathological investigations demonstrated that methylmercury expo-
sure affected the fetus in utero after maternal ingestion (Harada,
1979). A large incidence of fetuses affected by methylmercury in utero
was first reported in Minamata (Harada, 1979). Subsequently, these
cerebral palsy-like children started to be referred to as congenital
Minamata disease patients (Harada, 1978), and they were made
⁎ Corresponding author. Tel.: +81 86 235 7174; fax: +81 86 235 7178.
E-mail address: yorichan@md.okayama-u.ac.jp (T. Yorifuji).
previously evaluated the methylmercury concentrations in the
umbilical cords of inhabitants of the Minamata area (Nishigaki and
Harada, 1975). Then, they reported that the concentrations in the
umbilical cords reflected the chronological transition in methylmer-
cury pollution and also corresponded with acetaldehyde production,
in which methylmercury was produced as a byproduct. These findings
clearly suggested that methylmercury actually passed through the
placenta. However, that report involved a limited number of cases
(only 35) and did not quantitatively evaluate the regional differences
in the transition of methylmercury exposure.
Minamata comprises the southwestern part of Kumamoto Prefec-
ture in Japan and faces the Shiranui Sea (Fig. 1). In 1956, two young
sisters showing unknown neurological signs were officially referred to
their local public health center (Harada, 1995; McAlpine and Araki,
1958). Thereafter, numerous cases were identified and the first such
patient was traced back to 1942 (Nishigaki and Harada, 1975). The
causative agent was methylmercury, a byproduct of acetaldehyde
production that was discharged into Minamata Bay (a small part of
the Shiranui Sea) from a local chemical factory. In 1958, the factory
changed its waste drainage route from Minamata Bay to the Shiranui

0048-9697/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2009.10.011
 T. Yorifuji et al. / Science of the Total Environment 408 (2009) 272–276
Fig. 1. Map of the study areas.
Sea (Harada, 1995). Consequently, the exposure spread not only
within Minamata Bay but also along the entire coast of the Shiranui
Sea (Fig. 1) (Doi and Matsushima, 1996; Matsushima and Mizoguchi,
1996). As a result, patients who manifested similar neurological signs
began to be identified among the residents of not only the Minamata
Bay area, but also other villages around the Shiranui Sea (Ninomiya et
al., 1995; Yorifuji et al., 2008). At that time, the tide in the Shiranui Sea
was known to flow from north to south (Fujiki and Irukayama, 1979).
After methylmercury production ceased in May 1968, the Japanese
Government officially acknowledged a causal relationship between
wastewater from the Chisso factory and Minamata disease in the same
year (Health and Welfare Ministry of Japan, 1996). Although some
inhabitants of Minamata voluntarily stopped eating fish, consuming
fish has never been prohibited in Minamata and neighboring
communities. Therefore, the exposure continued and spread from
Minamata Bay area to the Shiranui Sea area during this period.
In the present study, we evaluated the temporal and spatial
distributions of methylmercury concentrations in umbilical cords,
with an increased number of participants and additional descriptive
analyses. Then, we examined whether the methylmercury concentra-
tions corresponded with the history of the Minamata disease incident
described above.
273
2. Materials and methods
2.1. Study site and participants
A total of 278 umbilical cord specimens collected after birth were
obtained from babies born between 1925 and 1980 in four areas
exposed to methylmercury, after informed consent was obtained.
Keeping umbilical cords after delivery is a traditional Japanese
custom. Generally, a piece of umbilical cord excised at delivery is
wrapped in gauze and stored in a small wooden box. Therefore,
umbilical cord samples were available for analysis as an indicator of
exposure to methylmercury in the past. The four study areas were the
areas around Minamata, Ashikita, Izumi and Goshonoura (Fig. 1). The
first patient was identified in the Minamata area where the causative
factory was located. All of the study areas face the Shiranui Sea.
2.2. Exposure assessment
Umbilical cord samples of <0.5 g were obtained in a dried state from
the early 1970s to 2008. One of the authors (M.H.) was responsible for
collecting all the umbilical cord samples. Since the span of this investi-
gation was so long (1970s to 2008), four institutes were mainly involved
274 T. Yorifuji et al. / Science of the Total Environment 408 (2009) 272–276

in measuring the concentrations, and the samples comprised of Table 1

32 samples from the Tokyo Metropolitan Research Laboratory Demographic characteristics of subjects (n = 278).

(Dr. Nishigaki), 50 samples from Tsukuba University (Dr. Fujiki), 10 n %

samples from the Tokyo Pollution Research Institute (Dr. Doi) and 110
Sex
samples from the National Institute for Minamata Disease (NIMD) (Dr. Men 132 47.5
Akagi and Dr. Sakamoto). Although 76 other samples were also Women 145 52.2
measured by these four institutions, we could not identify which Unknown 1 0.4
institutions were involved in the measurements owing to the very old
Area
data. In the early stages of the investigation (Dr. Nishigaki, Dr. Fujiki and Minamata area 85 30.6
Dr. Doi), the obtained specimens were analyzed by a modification of the Ashikita area 36 12.9
Westöö method (Westöö, 1968). Thus, Tokyo Metropolitan Research Izumi area 91 32.7
Laboratory, Tsukuba University, and Tokyo Pollution Research Institute Goshonoura area 6 2.2
Unknown 60 21.6
adapted this method (n = 92). In this method, the specimen was
digested with hydrochloric acid and the methylmercury concentration Measuring institutions
in the extract was determined by gas chromatography. More recently, Tokyo Metropolitan Research laboratory 32 11.5
(Dr. Akagi and Dr. Sakamoto at NIMD), the specimen was analyzed by a Tsukuba University 50 18.0
Tokyo Pollution Research Institute 10 3.6
revised analytical method (Ikingura and Akagi, 1996) (n = 110). In the
National Institute for Minamata Disease 110 39.6
revised method, the specimen was digested and then extracted with Unknown 76 27.3
dithizone solution in benzene. Subsequently, the methylmercury
concentration in the extract was determined by electron capture Periods
detector-gas chromatography. Born from 1925 to 1956 72 25.9
Born from 1957 to 1961 71 25.5
Because of historical data, information about performance of
Born from 1962 to 1965 61 21.9
quality assurance and quality control was not obtained regarding the Born from 1966 to 1980 63 22.7
method used in the early stages of the investigation (Dr. Nishigaki, Dr. Unknown 11 4.0
Fujiki and Dr. Doi). However, in the recent method conducted at
NIMD, the accuracy and precision of the method have been fully
evaluated internationally using a mussel homogenate (Mytilus edulis) No quality assurance information was available regarding the
reference material (IAEA-142) (Harada et al., 1999; Horvat et al., differences among the measuring laboratories. However, when we
1997). The detection limit for methylmercury was 0.5 ng/g dry wet restricted the analyses to participants in the Minamata area whose
(Harada et al., 1999). cords were measured at Tsukuba University and participants in the
The total mercury was not determined because each cord had been Izumi area whose cords were measured at the National Institute for
disinfected with mercurochrome. Minamata Disease, each institution measured the most umbilical
cords in the corresponding area, and the same tendencies were
2.3. Statistical analyses
Table 2
In the statistical analyses, we conducted descriptive analyses and Methylmercury concentrations (ppm) overall and separated by demographic char-
compared the methylmercury concentrations between sexes, areas, acteristics (n = 278).

institutions, and periods using t-tests or analysis of variance. We also n Mean SD Min Max p
drew scatterplots of the methylmercury concentrations of all the Overall
participants and separated by areas. SPSS software version 14.0J (SPSS Mean (standard deviation) 278 0.59 0.75 0.001 5.28
Japan Inc.) was used for all the analyses.
Sex
Men 132 0.60 0.72 0.001 4.65 0.99 ⁎
3. Results
Women 145 0.59 0.77 0.002 5.28
Unknown 1 0.44
The demographic characteristics of the subjects are shown in
Table 1. There were 60 participants whose residential areas were Area
unknown and 76 participants whose measuring institutions were Minamata area 85 0.49 0.72 0.020 5.28 0.00 †
Ashikita area 36 0.34 0.37 0.001 1.63
unknown. Table 2 shows overall methylmercury concentrations, sep- Izumi area 91 0.93 0.95 0.040 4.65
arated by the demographic characteristics. There was no difference in Goshonoura area 6 0.40 0.29 0.020 0.78
the concentrations by sex. The mean concentrations differed signif- Unknown 60 0.41 0.40 0.020 2.05
icantly among areas. The concentration in the Izumi area was the
Measuring institutions
highest. The mean concentration during the period of 1956–1961 was
Tokyo Metropolitan Research 32 1.06 1.22 0.020 5.28 0.03 †
the highest. laboratory
A scatterplot of all the participants is shown in Fig. 2a. Although Tsukuba University 50 0.68 0.86 0.030 4.65
the number of measurements was increased compared with the pre- Tokyo Pollution Research Institute 10 0.38 0.30 0.020 0.79
vious report (Nishigaki and Harada, 1975), the graph similarly de- National Institute for Minamata 110 0.59 0.68 0.010 3.37
Disease
monstrated that the concentration peaked in 1960 and corresponded
Unknown 76 0.38 0.34 0.001 1.85
with the acetaldehyde production. We also show concentrations of
mercury in shellfish in the Minamata area from 1960 to 1971 in Periods
Fig. 2b. The data was obtained from the paper written by Fujiki and 1925 thru 1956 72 0.66 0.86 0.010 5.28 0.00 †
1956 thru 1961 71 0.97 0.96 0.020 4.65
Irukayama (Fujiki and Irukayama, 1979). Similarly, concentrations
1961 thru 1965 61 0.44 0.44 0.020 2.20
decreased from 1960. Fig. 3 shows scatterplots classified by the four 1965 thru 1980 63 0.28 0.27 0.001 1.70
areas. Although the peak in the Minamata area was detected around Unknown 11 0.43 0.42 0.090 1.31
1955, the peaks in the Ashikita, Izumi and Goshonoura areas were Abbreviations: SD, standard deviation.
detected around 1960. In addition, the concentration seemed to be ⁎t-test were conducted between men and women.
higher in the Izumi area than in the Ashikita and Goshonoura areas. †Analyses of variance were conducted except for unknown groups, respectively.
 T. Yorifuji et al. / Science of the Total Environment 408 (2009) 272–276
Fig. 2. (a) Amounts of acetaldehyde production (tons) by the chemical factory during the
period 1932–1968 are shown as square dots and plotted by the line. Methylmercury
concentrations in the umbilical cords of participants (ppm) are shown as diamond-shaped
dots. (b) Concentrations of mercury in shellfish in the Minamata area from 1960 to 1971.
The data was obtained from Fujiki and Irukayama (Fujiki and Irukayama, 1979). Circular
dots show concentrations in Hormomya mutabilis. Triangular and diamond-shaped dots
show concentrations in Venus japonica in two difference points in the Minamata area.
Fig. 3. Methylmercury concentrations in the umbilical cords of the participants, according to study area. (a) Minamata area. (b) Ashikita area. (c) Izumi area. (d) Goshonoura area.
275
observed. Specifically, the peak in methylmercury concentrations in
the Minamata area was detected around 1955 and the peak in the
Izumi area was detected around 1960.
4. Discussion
The present study supplemented the findings in the previous
report (Nishigaki and Harada, 1975) with an increased number of
participants and additional descriptive analyses. Similar to the pre-
vious report, the present study showed that the methylmercury con-
centration corresponded with acetaldehyde production. Furthermore,
the analyses clearly demonstrated that the temporal and spatial dis-
tributions of methylmercury corresponded with the known history of
Minamata disease.
As mentioned above, after the first patient was officially identified
in 1956, some residents in the Minamata area voluntarily stopped
eating contaminated fish. As a result, the peak of the methylmercury
concentration was detected around 1955. After the factory changed its
drainage route in 1958, acetaldehyde production peaked in 1960,
inducing consequent peaks in methylmercury concentrations in the
Izumi, Ashikita and Goshonoura areas around 1960. Furthermore,
since the tide flowed from north to south, the concentration was
higher in the Izumi area.
Umbilical cord methylmercury concentrations can be used as a
valid measure of prenatal exposure to methylmercury (Grandjean
et al., 2005). Indeed, congenital Minamata disease patients and chil-
dren diagnosed with mental retardation in exposed areas had higher
methylmercury concentrations in their umbilical cords than control
children (Akagi et al., 1998; Harada et al., 1999). Therefore, con-
sidering that it is not complicated to keep umbilical cords after
delivery, the present findings remind us of the utility of umbilical
cords for pollution assessment, as previously suggested (Miller, 1976).
In the present study, umbilical samples were collected voluntarily.
Consequently, there is a possibility that the sampling method may
have affected methylmercury concentrations detected in each area.
276 T. Yorifuji et al. / Science of the Total Environment 408 (2009) 272–276
Thus, observed mean methylmercury concentrations might have
overestimated the real concentrations in each area because of par-
ticipants who were interested in this poisoning. However, such sam-
pling would not affect the relative concentrations or peaks in the
individual areas. In addition, since a higher abnormal pregnancy rate
was reported in the exposed areas (Itai et al., 2004), severely exposed
fetuses may have been aborted. However, this would not affect the
relative concentrations either.
This historical incident several decades ago in Minamata clearly
shows that regional pollution affected the environment in utero. The
temporal and spatial distributions of the methylmercury concentra-
tions in the umbilical cords tell us the history of the Minamata disease
incident.
Acknowledgment
We appreciate the contributions of S. Nishigaki, M. Fujiki, R. Doi, H.
Akagi and M. Sakamoto in measuring the samples. We also thank
Youichi Tani, Kimiyo Ito, and Akiko Yabe who helped us conduct the
study.
References
Akagi H, Grandjean P, Takizawa Y, Weihe P. Methylmercury dose estimation from umbil-
ical cord concentrations in patients with Minamata disease. Environ Res 1998;77:
98-103.
Doi S, Matsushima Y. Investigations on mercury content from hair samples regarding
Minamata disease. Report 2. In: Research Group on Minamata Disease, editor.
Collection of materials about Minamata Disease. Fukuoka: Ashi publisher; 1996.
p. 1508–13. (in Japanese), 1508–1513 pp.
Fujiki M, Irukayama K. Causes of Minamata disease and methylmercury pollution in
Shiranui-Sea. In: Arima S, editor. Minamata disease — researches during 20 years
and problems at present. Tokyo: Seirinsha; 1979. p. 121–52. in Japanese.
Grandjean P, Budtz-Jørgensen E, Jørgensen P, Weihe P. Umbilical cord mercury concen-
tration as biomarker of prenatal exposure to methylmercury. Environ Health Perspect
2005;113:905–8.
Harada M. Congenital Minamata disease: intrauterine methylmercury poisoning. Teratology
1978;18:285–8.
Harada M. Congenital Minamata disease — methylmercury intoxication during fetal
life. In: Arima S, editor. Minamata disease — researches during 20 years and problems
at present. Tokyo: Seirinsha; 1979. p. 345–70. in Japanese.
Harada M. Minamata disease: methylmercury poisoning in Japan caused by environmen-
tal pollution. Crit Rev Toxicol 1995;25:1-24.
Harada M, Akagi H, Tsuda T, Kizaki T, Ohno H. Methylmercury level in umbilical cords from
patients with congenital Minamata disease. Sci Total Environ 1999;234:59–62.
Health and Welfare Ministry of Japan. An opinion regarding Minamata disease and
future measures. In: Research Group on Minama Disease, editor. Collection of
materials about Minamata Disease. Fukuoka: Ashi publisher; 1996n. p. 1412–3.
(in Japanese), 1412–1413 pp.
Horvat M, Liang L, Azemard S, Mandic V, Villeneuve JP, Coquery M. Certification of total
mercury and methylmercury concentrations in mussel homogenate (Mytilus edulis)
reference material, IAEA-142. Fresenius J Anal Chem 1997;358:411–8.
Ikingura J, Akagi H. Monitoring of fish and human exposure to mercury due to gold
mining in the Lake Victoria goldfields, Tanzania. Sci Total Environ 1996;191:59–68.
Itai Y, Fujino T, Ueno K, Motomatsu Y. An epidemiological study of the incidence of
abnormal pregnancy in areas heavily contaminated with methylmercury. Environ
Sci 2004;11:83–97.
Kitamura S, Hirano Y, Noguchi Y, Kojima T, Kakita T, Kuwaki H. [Supplementary results of
epidemiological studies regarding Minamata disease — second report ] (in Japanese).
Kumamoto Igakkai Zasshi 1959;33(suppl. 3):569–71.
Matsushima Y, Mizoguchi S. Investigations on mercury content from hair samples
regarding Minamata disease. Report 1. In: Research Group on Minamata Disease,
editor. Collection of materials about Minamata disease. Fukuoka: Ashi publisher; 1996.
p. 1500–7. (in Japanese), 1500–1507 pp.
McAlpine D, Araki S. Minamata disease: an unusual neurological disorder caused by
contaminated fish. Lancet 1958;2:629–31.
Miller R. Dried human umbilical cores and study of pollutants. The Lancet 1976;307:315–6.
Ninomiya T, Ohmori H, Hashimoto K, Tsuruta K, Ekino S. Expansion of methylmercury
poisoning outside of Minamata: an epidemiological study on chronic methylmer-
cury poisoning outside of Minamata. Environ Res 1995;70:47–50.
Nishigaki S, Harada M. Methylmercury and selenium in umbilical cords of inhabitants of
the Minamata area. Nature 1975;258:324–5.
Westöö G. Determination of methylmercury salts in various kinds of biological material.
Acta Chem Scand 1968;22:2277–80.
Yorifuji T, Tsuda T, Takao S, Harada M. Long-term exposure to methylmercury and neurologic
signs in Minamata and neighboring communities. Epidemiology 2008;19:3–9.