The Application of HACCP to Mycotoxin Control – Post-harvest

1. Introduction

For the purposes of this article, we will be considering the post-harvest stage in the commodity supply chain as those operations immediately following harvest and leading up to the first processing steps. This will typically involve storage, drying (if necessary) and transportation steps. In fact the post-harvest stage can become very complex as the commodity passes through the hands of a number of separate intermediaries from farm to primary processor. This is another area where in-depth knowledge and expertise is required in drawing up the commodity flow diagram (CFD).

The post-harvest situation will normally be characterized by mycotoxins produced by the "storage" fungi. These are a loose collection of fungi such as Aspergillus and Penicillium species that are able to grow in relatively dry conditions. Typical mycotoxins produced include ochratoxins (Aspergillus and Penicillium) and aflatoxins (Aspergillus), although aflatoxin contamination can also be a field event.

2. Codes of Practice

We have already seen that, in the pre-harvest stage, HACCP schemes can be simplified in part by invoking Good Agricultural Practice (GAP). GAP and the related Good Hygienic Practice (GHP) are also relevant to the commodity immediately after harvest and are concerned with the initial on-farm treatment of crop material. An example of GAP at the early post-harvest stage is:

• Cultivation of coconut for oil and animal feed (copra cake) in Southeast Asia: immediately following harvest, coconuts are split into halves or smaller pieces for drying. The prevention of the coconut material coming into contact with the soil (leading to the possibility of fungal contamination) is considered to be GAP.

Good Storage Practice (GSP) is an additional code very relevant to the post-harvest stage, which is primarily concerned with the main bulk storage step(s). GSP will also be relevant to transportation conditions. Some factors that are normally considered to be part of GSP are:

• Storage structure sanitation
• Storage structure soundness
• Prevention of moisture ingress (e.g. use of moisture barriers on floors)
• Storage conditions: temperature, aeration
• Prevention of invertebrate infestation
• Prevention of access by rodents and birds

GSP codes are set out in CODEX Food Hygiene Basic Texts and in a number of ISO publications for storage of cereals and pulses. As we have seen previously, adherence to these codes will normally simplify the HACCP plan by reducing the number of necessary CCP’s. GSP controls are typically not in place specifically for the purpose of mycotoxin control, but have this effect almost by serendipity. Of course, if a storage or drying procedure is designed and implemented in a process specifically for mycotoxin control, then this is a CCP.

Codes of practice such as GAP and GSP are usually represented in some form in the requirements of the crop proficiency schemes that were mentioned in the last article. These schemes relate to measurable quality and operating practices and can be audited and verified. One such scheme with particular relevance to post-harvest handling is the "Recommended Code of Practice for Mill intake" (The Incorporated National Association of British and Irish Millers).

3. Post-harvest control

Control at the post-harvest stage is dominated by one single parameter – moisture level. The importance of drying and moisture control during storage is generally well understood by the industry, in terms of the importance of prevention of fungal contamination. As in the case of pre-harvest controls, the concern has not been with the occurrence of mycotoxins, but the far more obvious implications of allowing the commodity to become visibly "mouldy" - namely the negative effects on acceptability and price.

The objective of drying and subsequent moisture control is to maintain the commodity at a "safe" level where fungal growth and mycotoxin production are not possible. It should be borne in mind that some growers (particularly in developing countries) may not have access to technologically advanced drying equipment. In many regions outdoor sun drying is the most commonly used method, and this may fail in unsuitable weather conditions.

In the storage environment the relevant fungi are usually those species capable of growing in relatively dry conditions. Fungi that are able to grow in these conditions are termed "xerophilic" or "xerotolerant", depending on whether they grow best at relatively dry conditions (xerophilic) or whether they grow best at "wetter" conditions but tolerate dry conditions (xerotolerant). One complicating factor is that once growth is initiated (perhaps in a localised region of sufficiently high moisture content) the metabolic water produced by the growing fungus can perpetuate and amplify the process, and even allow the development of other, more moisture dependent fungal species.

The important parameter when considering moisture and fungal growth is not water content, but water "activity" (a_w). Water activity is a measure of the fraction of the water content of a material which is "free"- since in any natural material water will partition into "bound" and "free" states. "Bound" water is attached chemically or physically to the material, and is therefore not available to support fungal growth. The remainder is "free" water, which is immediately available for fungal growth. The value of water activity of a sample is directly related to the equilibrium relative humidity (ERH) the sample can generate by the expression (a_w = ERH/100), and water activity is therefore usually measured using an electronic hygrometer or psychrometer, which actually measures ERH. Water activity is measured on a scale from 1.0 (pure water) to 0.0 (completely dessicated material). The growth limit for the most xerophilic fungal types is around 0.7. The relationship between water content and water activity for a given material can be determined by constructing "moisture sorption isotherms" which are graphs of moisture content plotted against water activity. Once the relationship between water content and water activity is known, "safe" water activity levels can be translated into "safe" moisture contents. This is important in the commodity supply situation because the concept of moisture content is far more readily understood, and can be measured without the rather specialized equipment needed to assess water activity accurately. It is important to note that water activity is a function of temperature, so moisture sorption isotherms are always constructed at a specified temperature. In fact, for a given water content, water activity increases with increasing temperature. The practical implication for commodity storage is that a temperature rise may lead to an increase in water activity to unsafe levels, and allow the initiation of fungal growth. Hence the importance of temperature control in the storage situation.

Safe water contents vary from one commodity to another. For maize this is around 14% (w/w, wet weight basis), and for wheat around 15% (both at 20^oC). For groundnuts the safe moisture content at the same temperature is much lower, at around 7%. This is a reflection of the chemical makeup of the commodity and the relative proportion of "free" water.

Prevention of pest damage is relevant to mycotoxin control because pests, feeding on stored material, can cause local heating and moisture generation due to their metabolic activity. These warmer, damper areas provide ideal conditions for the initiation and development of local fungal growth and mycotoxin production. Localised effects such as this lead to the development of heterogeneous conditions in large bulk stores. Heterogeneity, and the resultant difficulty in obtaining representative analytical samples is a central issue in storage technology.

In the post-harvest stage commodities will also be assessed directly for evidence of fungal contamination. Therefore, weighbridge checks will typically include visual examination of a sample for evidence of fungal damage e.g. pink stained grains indicating Fusarium contamination. Material showing this type of damage will normally be immediately rejected.

4. Post-harvest CCP's

The designation of CCP's at the post-harvest stage will normally follow from examination of the system in question, and the results of any additional surveillance studies and controlled experiments. Stages critical to mycotoxin control can then be identified and the appropriate control measures implemented. This is best illustrated with reference to a specific example. In this example we will also consider some pre-harvest issues for the sake of completeness. This example is adapted from the HACCP Manual for Mycotoxin control, FAO/IAEA.

Figure 1. CFD of yellow maize for animal feed produced in South East Asia.

Production of yellow maize for animal feed formulation in South East Asia: this example was the result of a project carried out in the mid-1980’s when exports of this commodity were threatened due to persistent aflatoxin levels above the regulatory limits (e.g. 20m g kg^-1 in the EU). The CFD is shown in Figure 1.

Yellow maize is produced in South East Asia as wet and dry season crops. Dry season maize is found to be more susceptible to aflatoxin contamination pre-harvest, due to the increased risk of drought damage, while the larger wet season crop is more likely to be subject to contamination post-harvest, due to problems with sun drying. The harvested crop was typically stored initially on farm, then passed through a local (primary) then regional (secondary) trader before passing to large feed mills or export silos. All of these represent storage sites. "Shelling" of the maize i.e. removal of the kernels from the cob, normally occurred after a period of on-farm storage of harvested cobs.

The following findings were made as a result of surveillance and field drying studies:

• Pre-harvest, harvest: levels of pre-harvest aflatoxin contamination were generally found to be low, although evidence of ear rot was usually apparent to some extent.
• On-farm accumulation: the usual practice was to take the commodity straight from the field and store for 1 to 6 months. This practice was shown to typically result in high aflatoxin production.
• Shelling: the shelling process was considered low risk providing there was a low incidence of damaged kernels.
• Primary trader: freshly shelled maize kernels were highly susceptible to increased aflatoxin contamination if not dried to a suitable moisture content.
• Secondary trader: further contamination was frequently encountered at this step, due to prevailing practices.
• Feed mills and export silos: conditions at these sites, and the mixing of commodities of varying quality frequently led to presence and increase in aflatoxin levels at this stage.

These findings led to the development of the following outline plan (see also Figure1.):

• Pre-harvest and harvest controls are designated as GAP. This includes cultivar selection and prevention of pest damage.
• CCP 1: Field drying prior to harvest allows the crop to be subsequently dried to a safe moisture content for storage, and to be shelled directly at this stage if desired. This typically involves field drying of the mature crop to bring the moisture content from 35% to less than 22%. The critical limit designated is a moisture content of kernels of less than 22%.
• CCP 2: on farm segregation of mouldy cobs prior to shelling is introduced as a check for any pre-harvest contamination. The critical limit is rejection of cobs exhibiting greater then 10% surface mould damage.
• CCP 3: this CCP involves drying cobs for storage to less than 16% moisture content within 2 days of harvest. Alternatively material not to be stored on-farm is shelled within 2 days (or no longer than 1 week), from harvest. Critical limits here relate to moisture levels and/or sun drying times.
• Shelling operations invoke GAP.
• CCP 4: (primary trader*) the CCP at this stage is to ensure that freshly shelled maize is dried to less than 16% moisture content. Primary traders rely on sun drying which can be unreliable, so, in some instances the control measure will be the rapid transfer of the commodity to secondary traders (who are more likely to have mechanical drying facilities).
• CCP 5: (secondary trader*) again the CCP here is drying and moisture control. In this case 14%, with no part greater than 15% moisture content.
• CCP 6: (feed mills and export silos*). the CCP here is aflatoxin analysis of representative samples, and segregation of material exceeding the set limit. The critical limit will be the legislative target level e.g. less than 20m g kg^-1 for the EU.

*These stages also invoke GSP.

It can be seen in the above example that the concern is overwhelmingly with drying and moisture control. It can also be seen that mycotoxin analysis itself does not figure importantly in the supply chain until the latter stages of bulk storage. This is very typical for many commodity types. In developing countries this probably arises from a lack of finance and relevant skills. Even in the western world the time required to carry out analyses limits their usefulness as a monitoring tool, although specific analysis can be used to advantage in verifying a HACCP plan.

References

HACCP Manual for Mycotoxin Control. FAO/IAEA Training and Reference Centre for Food and Pesticide Control. Joint FAO/IAEA Division, Sweden.