Impact of mycotoxins on human health in developing countries

Abstract

Adverse human health effects from the consumption of mycotoxins have occurred for many centuries. Although mycotoxin contamination of agricultural products still occurs in the developed world, the application of modern agricultural practices and the presence of a legislatively regulated food processing and marketing system have greatly reduced mycotoxin exposure in these populations. At the mycotoxin contamination levels generally found in food products traded in these market economies, adverse human health effects have largely been overcome. However, in the developing world, where climatic and crop storage conditions are frequently conducive to fungal growth and mycotoxin production, much of the population relies on subsistence farming or on unregulated local markets. The extent to which mycotoxins affect human health is difficult to investigate in countries whose health systems lack capacity and in which resources are limited. Aflatoxin B₁, the toxin on which major resources have been expended, has long been linked to liver cancer, yet its other effects, such as immune suppression and growth faltering previously observed in veterinary studies, are only now being investigated and characterized in human populations. The extent to which factors such as immune suppression contribute to the overall burden of infectious disease is difficult to quantify, but is undoubtedly significant. Thus, food safety remains an important opportunity for addressing current health problems in developing countries.

Keywords:

• Aflatoxin
• health
• risk assessment
• mycotoxin
• fumonisin
• cancer
• developing countries
• Africa
• aflatoxicosis

Introduction

Although only recognized in recent times as a source of ill-health in humans, the agricultural problems associated with contamination of crops by fungal action have been noted for over two millennia (Campbell 1996). Consequently, human mycotoxicoses have probably also existed since the development of settled agricultural communities reliant on grain stores. Recent attempts to interpret the Biblical plagues of ancient Egypt over three millennia ago have suggested that the tenth plague, mentioned in the Book of Exodus and involving the death of the eldest sons, was due to macrocyclic trichothecene mycotoxins (possibly from Stachybotrys atra) occurring in the stored grains — the eldest being the first to access these stores (Marr & Malloy 1996). During the Middle Ages in Europe a common affliction known as St Anthony's Fire was prevalent and caused thousands of deaths over a period of many centuries (Marasas & Nelson 1987a). More recent understanding has identified this condition as ergotism, produced by the ingestion of the ergots of Claviceps purpurea, a fungus occurring on staple food grains such as rye and wheat. It has also been suggested that the ‘bewitchment’ displayed by persons in the then British colony of Massachusetts in North America in 1692 and which lead to the Salem witchcraft trials and executions were a result of convulsive ergotism (Woolf 2000). The deaths during the Second World War of thousands in the former Soviet Union from the haemorrhagic syndrome known as alimentary toxic aleukia, caused primarily by T-2 toxin produced by Fusarium sporotrichioides and F. poae contaminating cereal overwintered in fields (Marasas & Nelson 1987b), and the discovery in 1960 of aflatoxins, produced by Aspergillus flavus and A. parasiticus, focused attention on the adverse human health implications of the secondary metabolites of fungi.

Based on their known and suspected effects on human and animal health and their production by fungal pathogens of staple food crops, aflatoxin, fumonisin, deoxynivalenol (DON), ochratoxin A (OTA) and zearalenone (ZON)) are recognized as the five most important agricultural mycotoxins. A vast literature has arisen concerning aspects of the toxicology, chemical analysis and natural occurrence of these toxins. This has enabled risk assessments to be undertaken in various countries, which have highlighted the divide between the developed and developing worlds with respect to food safety. Firstly, mycotoxin exposures are generally lower in developed countries. The reasons for this are numerous. In the developed world, human diets are extremely varied, commercial food suppliers in these market economies compete with quality to meet the highest standards, while legislated maximum tolerated levels for mycotoxins are widely enforced and cover a majority of the population accessing food from retail markets. By contrast, diets consumed by the population of developing countries tend to be less varied and, when not grown on a subsistence basis, food items may be obtained from local markets with less attention to quality issues. In addition, there is less emphasis on legislating maximum tolerated levels and even when such legislation exists, the capacity to enforce it is frequently lacking. Food grown in these areas is frequently consumed irrespective of quality due to food scarcity problems. The problem of excessive consumption of a single cereal can readily be seen in many African diets which rely on maize consumed at levels between 400 and 500 g per person day^–1. Even moderate levels of mycotoxin contamination can result in exposure which exceeds the maximum tolerable daily intake (TDI) set by the Joint FAO/WHO Expert Committee on Food Additives (JECFA). In the case of fumonisin, a 60 kg person consuming 500 g maize day^–1 would exceed the provisional maximum TDI of 2 µg kg^–1 body weight day^–1 set by JECFA if the fumonisin contamination were to exceed 240 µg kg^–1 (Bolger et al. 2001). Yet contamination levels in home-grown subsistence maize frequently exceeds this level. By contrast, maize consumption within the countries of the European Union is approximately two orders of magnitude lower and fumonisin exposure is unlikely to approach the TDI (Bolger et al. 2001).

A further comparison of developing and developed countries reveals that for a human carcinogen such as aflatoxin B₁ (AFB₁), the carcinogenic potency used in calculating the population cancer risk is greater in developing countries. This is a consequence of AFB₁ being synergistic with hepatitis B virus (HBV) infection, which has a greater prevalence in the developing world. The overall potency of AFB₁ in risk characterization calculations is adjusted to reflect the proportion of hepatitis B surface antigen-positive individuals in the population (Henry et al. 1998). Thus for both parameters involved in risk characterization, namely population exposure and potency, values reflecting the population in developing countries are higher than typical values for developed countries.

Recognized effects on human health

The improvements in food safety in developed countries mentioned above have eliminated acute human mycotoxicoses such as ergotism, which was previously well known in the Middle Ages in Western Europe. However, such outbreaks still occur in rural communities in the developing world, as evidenced by documented cases in Ethiopia, East Africa, where there were outbreaks of gangrenous ergotism in 1978 after consumption of grain contaminated with Claviceps purpurea (Demeke et al. 1979; King 1979). The most tragic recent outbreaks of human mycotoxicosis have happened in Keny, where deaths due to aflatoxin exposure have occurred over a number of years. Aflatoxicosis is a toxic hepatitis leading to jaundice and, in severe cases, death. Repetitive incidents of this nature have occurred in Kenya during 1981, 2001, 2004 and 2005 (Shephard 2004; Lewis et al. 2005). During January to July 2004, in the eastern and central districts of Kenya, 317 cases were admitted to hospital with jaundice. Of these, 125 deaths were recorded. Maize collected from the affected areas had high AFB₁ levels with 55% of samples contaminated above the Kenyan legal limit of 20 µg kg^–1, 35% had levels above 100 µg kg^–1, and 7% above 1000 µg kg^–1. The maximum level found was as high as 8000 µg kg^–1. A similar outbreak in the eastern districts of Kenya during 2005 led to 75 cases being admitted to hospital (as of 21 June 2005) and 32 deaths (Centers for Disease Control (CDC) 2006). Of the maize samples collected in a survey of the affected region, 42% were contaminated with AFB₁ levels above 20 µg kg^–1, 24% were above 100 µg kg^–1, and 7% above 1000 µg kg^–1. Acute human aflatoxicosis has also been reported from Asian countries such as India (Bhat 1991) and Malaysia (Lye et al. 1995).

AFB₁ has been extensively linked to human primary liver cancer in which it acts synergistically with HBV infection and was classified by the International Agency for Research on Cancer (IARC) as a human carcinogen (group 1 carcinogen) (IARC 1993a). This combination represents a heavy cancer burden in developing countries. A recent comparison of the estimated population risk between Kenya and France highlighted the greater burden that can be placed on developing countries (Shephard 2006). Based on respective estimates for aflatoxin exposure of 133 and 0.12 ng kg^–1 body weight day^–1 and respective HBV prevalence of 25 and 1%, the liver cancer risk would be 11 vs. 0.0015 cancers per year per 100 000 population, respectively. Given recently published liver cancer incidence rates in the European Union of 10.0 per 100 000 for males and 3.3 per 100 000 for females (Bray et al. 2002), it is clear that aflatoxin plays a significant role in liver cancer in developing countries, but not in the developed world where other risk factors such as cirrhosis are more important.

Fumonisins have been implicated in one incident of acute food-borne disease in India in which the occurrence of borborygmy, abdominal pain, and diarrhoea was associated with the consumption of maize and sorghum contaminated with high levels of fumonisins (Bhat et al. 1997). Fumonisin B₁, the most abundant of the numerous fumonisin analogues, was classified by the IARC as a group 2B carcinogen (possibly carcinogenic in humans) (IARC 2002). Studies in the former Transkei region of South Africa and in Linxian and Cixian counties, China, have demonstrated an association between fumonisin exposure in rural subsistence farming areas and a high incidence of oesophageal cancer (Rheeder et al. 1992; Zhang et al. 1997). Fumonisins, which inhibit the uptake of folic acid via the folate receptor (Stevens and Tang 1997), have also been implicated in the high incidence of neural tube defects in rural populations known to consume contaminated maize, such as the former Transkei region of South Africa and areas of Northern China (Marasas et al. 2004).

The other three agriculturally important mycotoxins have also been associated with various outbreaks of human disease, mostly in developing countries. A number of occurrences of acute food-borne illness in India and China involving gastrointestinal symptoms have been attributed to the consumption of DON-contaminated cereals (Luo 1988; Bhat et al. 1989). OTA has long been associated with Balkan endemic nephropathy (BEN), a fatal renal disease with histopathological similarities to OTA-induced nephropathy in swine, and has been associated with the incidence of epithelial tumours of the upper urinary tract (Benford et al. 2001; Castegnaro et al. 2006). OTA was classified by the IARC as possibly carcinogenic to humans (group 2B carcinogen) (IARC 1993b). ZON is a naturally occurring endocrine-disrupting chemical and has been associated with clinical manifestations of hyper-oestrogenism in humans, including an outbreak of precocious pubertal changes in young children in Puerto Rico in the Caribbean (Saenz de Rodrigues et al. 1985) and gynecomastia with testicular atrophy in rural males in southern Africa (Campbell 1991).

Emerging effects on human health

The identified harmful effects of mycotoxin exposure on human health, as exemplified by acute aflatoxicosis and primary liver cancer, are increasingly being recognized as only the tip of the iceberg of morbidity associated with chronic mycotoxin exposure (Miller 1998; Williams et al. 2004). The WHO World Health Report 2002 (2002) identified the top ten health risks for developing countries with high mortality. These are listed in Table I, together with their associated disease burden as measured in disability-adjusted life years (DALYs) and the main type of affected outcome. The WHO report lists ten types of affected outcomes associated with a total loss of 833 million DALYs in developing countries with high mortality. Of these outcomes, infectious and parasitic diseases represent over 90% of the outcomes in the top five risk factors (underweight, unsafe sex, unsafe water, indoor smoke from solid fuels and zinc deficiency) and account for 300 million DALYs overall. Although mycotoxins are not specifically mentioned, they may be seen to play a modulating role in a certain number of these factors. According to the WHO estimate, being underweight caused 3.7 million deaths in 2000, mostly in children under five years of age in developing countries. The mortality and morbidity were due to the effect of poor nutrition on immune function, which led to diarrhoeal diseases, malaria, measles and pneumonia. Recently published results indicate that both being underweight and immune function are directly affected by aflatoxin exposure in developing countries. Studies in Benin and Togo, West Africa, have shown that aflatoxin exposure in infants, which increases at weaning, can lead directly to growth impairment and stunting (Gong et al. 2002). Aflatoxin has long been linked to kwashiorkor, a disease usually considered a form of protein energy malnutrition, although some characteristics of the disease are known to be among the pathological effects caused by aflatoxins in animals. It has been suggested that either aflatoxins could play a causal role in the disease (Hendrickse 1991) or children suffering from the disease are at greater risk to the hazards of dietary aflatoxin (Adhikari et al. 1994). Studies in Gambia and Ghana, also in West Africa, have begun to elucidate the role of aflatoxin in immune suppression in human populations. Aflatoxin exposure was associated with reduced levels of secretory immunoglobulin A (IgA) in Gambian children (Turner et al. 2003). Changes in differential subset distributions and functional alterations of specific lymphocyte subsets have been correlated with aflatoxin exposure in Ghanaian adults and indicate that aflatoxins could cause impairment of human cellular immunity that could decrease resistance to infections (Jiang et al. 2005).

Table I. Leading health risks for developing countries with high mortality, the corresponding contribution to the burden of disease in attributable disability-adjusted life years (DALYs) and the main type of affected outcome as given in the WHO World Health Report 2002 (2002).

Leading health risks	Burden in DALYs^1	Main component of outcome
Underweight	14.9	infectious and parasitic disease
Unsafe sex	10.2	infectious and parasitic disease
Unsafe water, sanitation and hygiene	5.5	infectious and parasitic disease
Indoor smoke from solid fuels	3.6	infectious and parasitic disease
Zinc deficiency	3.2	infectious and parasitic disease
Iron deficiency	3.1	maternal and perinatal conditions
Vitamin A deficiency	3.0	infectious and parasitic disease
Blood pressure	2.5	cardiovascular disease
Tobacco	2.0	cardiovascular disease^2
Cholesterol	1.9	cardiovascular disease
^1As a percentage of a total of 833 million DALYs.
^2Infectious disease, cancer and chronic respiratory disease are also significant outcomes.

Of the other health risk factors, the morbidity and mortality associated with unsafe sex, unsafe water and indoor smoke, arises from infectious diseases, such as HIV/AIDS, infectious diarrhoea and lower respiratory tract infection, respectively. The immunological suppression associated with aflatoxin and possibly DON could adversely affect all these outcomes. The modulating effect of aflatoxins in cases of zinc, iron and vitamin A deficiency in human health is less clear, but evidence from animal nutrition would suggest it could be significant (Williams et al. 2004).

Challenges and conclusions

Exposure to mycotoxins is a serious risk to human health, especially in developing countries where the effects of poverty and malnutrition lead to an exacerbation of the detrimental effects of these food-borne toxins by circumscribing biochemical detoxification mechanisms. Even where the health risks of mouldy grain to human health are recognized, the contaminated grain may be fed to livestock, decreasing animal productivity and the food supply, and increasing poverty (Wu et al. 2005). Apart from the fact that aflatoxin is a causative factor for primary liver cancer, strong epidemiological evidence for mycotoxins as causative factors of various diseases is still lacking. Nevertheless, exposure to these compounds should be addressed as an urgent food safety issue as they place a significant constraint on attempts to improve human health in developing countries. Measures to achieve the UN Millennium Development Goals are aimed at impacting directly on poverty in the developing world and hence on food choices and food safety.

In addressing mycotoxins as a food safety problem, the difficulty arises in prioritizing this issue in communities in which food sufficiency has not been attained. In the search for a means to address the question, it is not clear whether appropriate technologies addressing specific subsistence farmer needs (Turner et al. 2005) or developed world technologies such as improved varieties, genetically modified organisms and agricultural modelling (Gressel et al. 2004) (or a combination of both) will be more effective in improving food safety. Finally, the effects of climate change on fungal distribution and toxin production may present a series of new food safety challenges.