Past, present, and future of environmental specimen banks

Abstract

Environmental specimen banks are an essential part of the infrastructure of environmental sciences. They have various functions: (1) evaluation of governmental environmental policy-making and regulations; (2) a resource for animal health evaluation; (3) research tools to investigate time trends in ecosystems; (4) detection of newly emerging chemicals in the time trends; (5) validations of computer models for environmental phenomena; (6) source identification of contaminants; (7) a tool for food safety; (8) evaluation of genetic selection pressure due to environmental changes. In this review paper, we present a detailed description of the Kyoto University Human Specimen Bank (history, protocol and questionnaires) and provide brief outlines of other representative environmental specimen banks. We then review two illustrative cases in which environmental specimen banks have unveiled insidious contaminations of polybrominated diphenyl ethers and perfluorooctanoic acids. Finally, we give a perspective of new functions for environmental specimen banks in the next 20 years.

Introduction

The rapid development of new materials, new production methods, and new pharmaceuticals and commercial products in the 21st century has resulted in the release and/or emission of a myriad of chemicals into the environment. The environmental fates of only a few of the estimated 70,000 chemicals commonly used in industry have been characterized. Since monitoring lags far behind the rate of new development, regulatory decisions should be made as soon as possible to minimize the effects on ecological systems, including wild animal health.

An environmental specimen bank (ESB) is an organization and facility that is engaged in the systematic long-term preservation of representative environmental specimens. Specimens from ESBs have been used for retrospective analysis and evaluation for regulatory decision-making. As such, a well-designed ESB can be a valuable resource of specimens for real-time and retrospective monitoring. Specimens maintained in ESBs have enabled investigators to extend their current research on present-day situations into the past as well as extrapolate it to the future. They enable future exposure assessments for given chemicals to be made under various scenarios.

In the last 30 years, formal ESBs have been constructed in many countries, including the USA, Germany, Sweden, and Japan. A human specimen bank has also recently been established in Kyoto University (i.e., the Kyoto University Human Specimen Bank) [1]. An important aspect of this specimen bank is that it provides the means to reconstruct human exposures from the 1970s through to 2008. The bank contains human samples collected not only in Japan but also in various Asian countries, including China, Korea, Thailand, Vietnam, Malaysia and the Philippines, and is thus expected to provide a means of monitoring temporal as well as geographic trends in environmental contamination.

The major aim of this review is to introduce the Kyoto University Human Specimen Bank. Features of representative ESBs—in Japan, the USA, Germany and Sweden—are also compared to demonstrate individual functions of the ESBs. Finally, we discuss the future functions of ESBs in the environmental sciences.

Kyoto University Human Specimen Bank and other representative environmental specimen banks throughout the world

As of 2008, there are more than a dozen ESBs in the world, with one also currently under construction in France. Here, we provide a brief description of a number of these as well as their protocols (Table 1).

Table 1 Current environmental specimen banks in 2008

Kyoto University Human Specimen Bank

The Kyoto University Human Specimen Bank was established in 2004 at the Kyoto University Graduate School of Medicine [1]. The stored samples originate from four research activities. The first group of samples was collected in Japan as part of the nation-wide heavy-metal monitoring projects led by Prof. Ikeda [2–6] during the late 1970s up to the 1990s. Beginning in 1980s, samples were systematically collected in Japan and other Asian countries within the framework of a consistent sampling design in which participants donated blood, urine and duplicate 24-h food samples. Personal information and biochemical data were obtained by questionnaire and biochemical analysis and included data on age, gender, blood pressure, past and present illnesses, medication use, aspartate aminotransferase, alanine, gamma-glutamyl transferase, total cholesterol, triglycerides, high-density lipoprotein cholesterol, urinary protein, and red blood cells in urine [8]. In the duplicate food sampling protocol, the foods are cooked and meal menus recorded. The samples are then transferred to the laboratory within 48 h and stored at −30°C until processed. In the processing of food samples, each food composite homogenate is weighed and homogenized together with the drinking water. One-liter (approx. 1 kg) portions of the total homogenate are stored in ten 100-ml bottles at −30°C. Concurrent and long-term trends in Pb and Cd exposures are reported in detail [2–6].

The second group of samples comprises samples collected in Akita prefecture during the 1980s. The samples consist of breast milk, blood, and serum samples donated by the Hiraga General Hospital in the rural area of Akita. These samples were originally collected to monitor farmers’ exposure to pesticides [7].

The third group of samples was collected by Prof. Koizumi and his colleagues from 2004 to 2006 [1]. Blood and breast milk samples were collected nationwide in Okinawa, Kochi, Hyogo, Kyoto, Takayama (Gifu), Fukui, Tokyo, Miyagi, Akita, and Shizunai (Hokkaido). In this project, commercially available packed breakfast, lunch, and dinner samples were collected from those sampling sites. Breast milk sampling was conducted until 12 weeks post-partum. Blood and breast milk donors also submitted self-reported questionnaires (Table 2). Within the framework of this study, food samples were homogenized as a set of breakfast, lunch, and dinner samples, and drinking water was collected at the sampling sites in the same manner as in the first study.

Table 2 Self-reported questionnaire

The fourth group of samples was collected in 2007 and 2008 in Japan (Miyagi, Takayama, and Kyoto), Beijing in China, Seoul, and Busan in Korea and Hanoi in Vietnam. In this project, blood or breast milk samples were collected domestically. However, blood and meals were sampled in the same way as in the third group of samples mentioned above. The donors of blood and food completed self-reported questionnaires (Table 2) and food record sheets (Table 3). All breast milk donors, irrespective of nationality, followed the same protocol: samples were collected up to 12 weeks post-partum, and donors filled out questionnaires (Table 2).

Table 3 An example of a food record sheet

The total quantities of samples are shown in Table 1. Meta data describing the donor’s personal information are shown in Table 1.

The Kyoto University Human Specimen Bank was designed so that human exposure assessments can be made on samples taken from the 1980s to the present. When distribution requests are received, the protocol will be reviewed by the committee of our sample bank. If the request is approved, our sample bank will release the specimens requested to the researcher(s) without any fees other than shipping.

es-Bank

Ehime University (Matsuyama City, Japan) began collecting environmental specimens in 1965 [9]. At that time, Ehime University focused on collecting specimens for studying local environmental contamination with pesticides that had been used by regional farmers. Samples were systematically collected and stored by the staff of Ehime University, and these samples later became seeds for subsequent collections of samples on larger scales. Thousands of samples from all over the world have been collected by the research group of the center for marine environmental studies over the past three decades. A large portion of these globally collected ecological samples was upgraded to form the es-Bank in 2002. The unique scientific merit of this collection, which cannot be matched by other environmental specimen banks, is its global scope, with a large number of specimens from the Asia–Pacific region (Table 1).

Time capsule ESB (http://www.nies.go.jp/)

The National Institute for Environmental Studies (NIES) in Japan started a pilot ESB in 1979. The Environmental Specimen Time Capsule program was extended and has started storing environmental specimens and genetic resources of endangered species (Table 1). The aim of this specimen bank is to store specimens for a long period (50–100 years) to await future needs and analyses. The bank has compiled atmospheric samples as well as samples of bivalves, fish, and human breast milk. The bivalve archives are very comprehensive and very important as environmental samples. Specifically, they are expected to provide information on long-term changes in genetic diversity or the natural selection of these species due to climate change.

This ESB, which is supported nationally, is characterized by long-term storage under the strong initiative of the NIES; as such, it does not allow the distribution of samples upon request by researchers. At the present time, the time capsule ESB does not seem to have a systematic sampling design; rather, it seems to be aimed at covering a large variety of research needs in the future.

The U.S. National Biomonitoring Specimen Bank and the Marine Environmental Specimen Bank

These two banks are very well designed and have a very clear protocol [10]. There are two national ESBs that have very similar designs. The first sample bank is the CASPIR [The CDC (center for disease control) and ATSDR (Agency for Toxic Substances and Disease Registry) Specimen Packaging, Inventory and Repository], which has collected various human specimens as part of public health activities by the CDC and ATSDR. The second sample bank is maintained by the National Institute of Standards and Technology (NIST) and consists of two separate facilities: the National Biomonitoring Specimen Bank and the Marine Environmental Specimen Bank. While CASPIR maintains specimens for human health research, the NIST banks are designed for environmental research (Table 1).

The relevant ESB in the USA continuously monitors animals living in diverse environments covering Texas desert areas, Hawaii, and Alaska [11]. Activities also include monitoring endangered species. The collections cover fish, mammals, avian species, and plants. This sample bank was designed to consider the transfer of contamination through the food web and the health status of wild animals and as such, it plays a key role in quality assurance. Stored samples are presently being analyzed using newer and more sensitive analytical methods.

German Environmental Specimen Bank for human tissues (ESBHum: http://www.umweltprobenbank.de)

The ESBHum was established as part of the German Environmental Specimen bank, and it focuses on human exposure assessments by real-time and retrospective monitoring [12, 13]. Samples are processed annually to measure 20 inorganic (Sb, Th, As, Ba, Cd, Pb, Hg, Ag, Tl, Sn, U, Cu, Ca, Fe, Mg, K, Se, Na, Sr and Zn) and five organic (hexachlorobenzene, pentachlorophenol, PCB-138, PCB-153 and PCB-180) chemicals. Samples are donated annually by 500 voluntary students aged 20–29 years, who live in four cities (Munster, Halle, Griefswald, and Ulm). The participants provide 24-h urine, blood, and other human specimens. Detailed personal information is attached to the samples. Given this context, the ESBHum can be said to be designed for health-related environmental monitoring.

The Swedish Specimen Bank, Swedish Museum of Natural History

This ESB was initiated in 1980 by the Swedish Environmental Protection Agency to study residue levels of pollutants and their effects on biota in terrestrial, freshwater, and marine environments [14]. The aim of this sample bank is to collect, prepare, store, and supply specimens for a variety of tasks in order to provide information for updating environmental agendas.

At the present time, the Swedish ESB stores specimens on 260,000 organisms. Approximately, 8,000–9,000 specimens are collected annually. The Swedish monitoring programs are tightly linked to banking and monitoring, and 3,500 specimens are consumed annually to investigate time trends, spatial monitoring, and screening of new substances.

In concert with ESB activities, the Swedish EPA established a program in 1989 for the bio-monitoring of top marine predators. This program aims to monitor the population, reproduction, development, and health status of three types of seals and a white-tailed sea eagle. To support this program, the ESB stores tissues and organ samples from these animals. The Swedish ESB is also currently collecting plants, mosses, sediments, sludge, and human foodstuffs.

Lessons taught by the use of ESB specimens in modern environmental problems

The ESBs have become an essential part of the infrastructure of modern environmental sciences and decision-making and have played key roles in a wide range of aspects related to the environmental sciences, such as (1) evaluations of governmental environmental policy-making and regulations; (2) as a resource for animal health evaluation; (3) as research tools to investigate time trends in ecosystems; (4) detection of newly emerging chemicals in time trends; (5) validations of computer models for environmental phenomena; (6) source identification of contaminants; (7) as a tool for food safety; (8) evaluation of genetic selection pressure due to environmental changes. Here, we briefly outline the roles of modern ESBs in recent environmental issues.

Polybrominated diphenylethers

It is known that, contrary to the case with organochlorine compounds, the use of diphenylethers (PBDEs) increased in the European Union (EU) during the 1980s. These products were widely used as flame retardants, especially in polymers used in electronics and textiles. Similar to the organochlorine compounds, PBDEs were found in ecological biota [15]. Thus, the first screening was conducted using archived breast milk samples in 1997 [16, 17], and the first astonishing evidence that emerged revealed an exponential increase in PBDEs in Swedish breast milk from 1972 to 1997 [18]. This increasing trend of PBDEs in human breast milk [19, 20] and serum [1, 21] was subsequently confirmed in several other countries.

The unique feature of our sampling design is that serum and duplicate food samples were collected from the same person. This enabled us to obtain definitive evidence that dietary intakes of PBDEs estimated from duplicate food samples in 1995 did not differ from those collected in 1980 [22], while PBDE levels in serum were significantly higher in 1995 than in 1980 [1]. These results suggest the importance of inhalation as a primary route of exposure.

The initial alarming evidence generated by Swedish researchers showed the importance of continuous monitoring using breast milk [23] and raised concerns internationally, resulting in new regulations in many countries, since PBDEs are suspected to have a variety of toxic effects on wildlife and humans [24].

Perfluorooctane sulfonate and perfluorooctanoate

Perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) are two classes of chemicals that have been used in a variety of applications, such as in lubricants, paints, cosmetics, and fire-fighting foams. The former has been an important perfluorinated surfactant, but in 2002, after 50 years of production, The 3M company phased out its manufacture. Once released into the environment, PFOS is postulated to be stable and persistent due to its resistance to degradation in ecological systems and its bioconcentration in food webs. As postulated, PFOS and PFOA were found in a variety of wildlife [25–28]. In Japan, nationwide surveys have demonstrated high-level contamination of PFOS in an airport and extremely intense PFOA contamination in Osaka Bay and the Kanzaki River [29, 30].

There have been few studies on PFOS and PFOA levels in humans. Data from early studies in the USA demonstrate that PFOS and PFOA serum levels have not changed over the past 20 years, although they did increase up to the 1980s [31].

In 2004, we investigated the time trend and special distribution of PFOS and PFOA in Japan using specimens stocked in the Kyoto University Human Specimen Bank [32, 33]. The analyses revealed an exponential increase in serum PFOA concentrations in Japan from 1980 to 2000, while PFOS levels reached a plateau during that time (Fig. 1, cited from Harada and Koizumi [34]). Experiments to reconstruct time trends and spatial differences have been made in China [35] and other countries and have confirmed increasing levels over the past 30 years [34].

Fig. 1

Time trends in perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) serum levels in Japan and the USA. Cited from Harada and Koizumi [34]. Data are geometric means and geometric standard errors

In our previous studies, we had found that there was a local emission source of PFOA in the Osaka region [29, 36]. The human serum levels of PFOA in the Osaka region were significantly higher than those in other regions [33]. We thus conducted a study to determine to what extent dietary exposure can explain the serum levels in residents of a highly PFOA-contaminated area (Osaka) and a non-contaminated area (Sendai) by duplicate food samples and paired serum samples stocked in the Kyoto University Human Specimen Bank [37]. The result revealed that the dietary route, including drinking water, cannot explain the high levels of serum PFOA in Osaka residents, suggesting that inhalation should also be taken into account when explaining excess PFOA exposures. These results showed the usefulness of paired sampling of food duplicates and blood samples to reconstruct human exposures.

It should be mentioned that there are several reports on the decline of PFOA and PFOA concentrations in human blood following the withdrawal of production of PFOA and PFOS by the 3M Company [38, 39]. We are currently planning to test whether such declines are global trends or not.

Other studies

There have also been several studies that have reconstructed long-term exposures to persistent organic compounds other than routine monitoring substances, such as PCBs and organochlorine pesticides or insecticides. For example, there is a report on phthalate [40].

Future perspectives of ESB functions

Environmental specimen banks have become an essential part of the fundamental research infrastructure for environmental sciences. In the next 20 years, further breakthroughs in technologies will occur. In terms of environmental studies, two of these will have a large impact. The first one is a technology which enables us to analyze isotopic separation, and the second is a high-throughput pyrosequencing technology. Those advances in technologies will create new functions for ESBs.

Fine isotopic profiling

¹³C and ¹⁴C are natural isotopes that are incorporated in CO₂ by plants. Labeled isotopes will be transformed to glucose via photosynthesis in plants. However, photosynthetic enzymes prefer to utilize ¹²C and radioactive isotope ¹⁴C will be degraded to ¹⁴N in fossil fuels. Thus, the greatest anthropogenic source of CO₂ production, i.e., the incineration of fossil fuels, will yield ¹³C- or ¹⁴C-depleted CO₂. This in turn results in the production of ¹²C glucose and other biological products. Accordingly, ¹³C versus ¹²C or ¹⁴C versus ¹²C ratios in diets or human compositions are variable according to the extent to which anthropogenic CO₂ was absorbed in recent years [41, 42]. Other isotopic analyses also give us interesting information. For example, lead from a smelter emission from a local smelter plant had ²⁰⁶Pb versus ²⁰⁷Pb ratios of 0.993, which is significantly smaller than the ratio in natural lead [43]. Such mineralogical signatures will provide information for identifying emission sources in transboundary contaminant transfers. Isotopic ratios of ²⁰⁶Pb and ²⁰⁴Pb have given a clear demarcation for the separation of geochemical signatures of authoritarian lead from other lead [44]. However, such clear signatures are now becoming less clear [44]. In terms of lead measured in Chinese studies, there are several overlapping geochemical signatures of isotopic ratios. Thus, the dominance of coal combustion as a source of lead has made it difficult to perform geological identification of the sources in China [45].

In the next few decades, isotopic analysis linked with the banked samples will provide a new area of research for environmental sciences.

DNA profiling

In recent times, many genetically modified organisms (GMOs) have become commercially available in many countries. The rapid progress of GMOs has enabled the conferring of new characteristics, such as herbicide tolerance, resistance to insects, among others into plant genomes. The foreign pieces of DNA consist of a transcription promoter, a coding sequence, and an expression terminator. Examples of transgenic plants include soybeans and maize. In recent years there has been an ongoing debate on the risks associated with the introduction of GMOs into agriculture. Consequently, research evaluating the effects of GMOs has become increasingly important. Such GMO assessments are carried out by detecting inserted foreign DNA in transgenic plants. DNA is the preferred analyte for both raw ingredients and processed food. A long-term time trend of the environmental fate of foreign DNA needs to be traced using food samples.

The testing of samples in specimen banks will be very informative in determining such long-term trends. Ecological samples are especially useful when GMOs are being monitored—i.e., the ecological fate and influences of GMOs on ecological biota can be assessed rigorously [46]. In particular, rapid advances in high-throughput sequencing technology enable large-scale sequencing without any prior assumptions, and horizontal or vertical transmission of the genetic elements of genetically engineered genomes can be traced.

Another monitoring protocol would be to investigate the selection pressure posed by global environmental changes. Rapid environmental changes will increase natural selection pressures and induce alterations at genome levels, as previously reported [47]. Such effects can be tested by real biota samples collected over the long term. Ecological environmental specimen banks are suitable for conducting such studies.

Conclusions

In the last 20 years, ESBs have emerged as part of the fundamental research infrastructure required for environmental sciences. In their early phase of development, ESBs were expected to monitor local ecological or human exposures. The expansion of environmental problems on both geographical and time scales has resulted in ESBs differentiating into purpose-oriented groups, with some becoming more oriented to the ecological environment while others trending towards human exposure determination.

The increase in environmental problems has led to a need for global environmental monitoring. Increases in the demands for ESBs now require that sample exchanges, supplies, banking and other relevant activities associated with ESBs be standardized by an internationally accredited guideline. Such guidelines, including those on legal issues, ethical issues (especially for human samples), and technical issues, have recently been proposed [48].