• Rural NZ
• Adverse Events
• People issues
• Profitability
• Research
  • Biotechnology
  • Pest control
  • Progress
  • Results
• Rural Proofing
• SLM Hill Country Erosion Programme
• Community Irrigation Fund
• Stats & forecasts
• Sustainability
• Legislation
• Media Centre
• Publications
• What's New

2. GM Material in Plant Parts Collected by Bees

For the purposes of this discussion, "GM material" is defined as either transgene DNA or novel proteins encoded by transgenes. (This is also the definition used in the Australian and New Zealand Food Authority (ANZFA) regulations concerning food labelling in relation to GM.) Plant parts collected by bees are pollen, nectar, resin or honeydew (sap that has passed through the digestive system of a sap-sucking insect).

In GM plants, transgene DNA may occur in any plant tissue which normally carries DNA, but the novel protein which it encodes will not necessarily be expressed in every tissue, since this process is directed by the promoter (switch) on the gene in question. There are many different promoters available to genetic engineers. Some switch on transgene expression in all parts of the plant, while others switch it off in all but one tissue (e.g. a root-specific promoter).

2.1 Potential for GM Material to be Collected by Bees.

2.1.1 Pollen.

Since pollen is a plant tissue composed of cells capable of protein synthesis, it is reasonable to expect to find transgene DNA in pollen grains and also novel protein, if the transgene's promoter permits it. Table 1 presents currently available data on novel protein expression levels in pollen from GM plants.

Compared to leaves, which are about 2% protein (J. Christeller, pers. comm.), pollen has a high protein content, estimated to be between 8 and 40% (Herbert, 1992), and for this reason transgenes may be expected to express novel proteins at reasonable levels in pollen. However, one of the most commonly-used promoters in GM plants, cauliflower mosaic virus (CaMV) 35S promoter, does not drive pollen expression particularly well. For example, in leaves, this promoter can drive expression of novel proteins such as protease inhibitors up to about 0.4% of total protein, but the same protein will be undetectable in the pollen of the those plants (Bonadé Bottino et al., 1998) (see also Table 1). The highest recorded level of a novel protein expressed in pollen via the CaMV 35S promoter is 0.6 mg per g pollen (fresh weight) for a Bt toxin in cotton (Greenplate, 1997). This is equivalent to 0.00024% w:w of total soluble protein, assuming the pollen is 25% protein (Table 1).

Higher levels of novel proteins may occur in pollen if the transgene includes a pollen-specific promoter such as that derived from maize (Kozeil et al., 1993). Expression levels as high as 0.0418% w:w of total soluble protein have been recorded for a Bt toxin with this promoter in maize pollen (Kozeil et al., 1993; Table 1). This type of GM Bt-maize (referred to as an event 176 hybrid) was used as the source of insecticidal pollen in the much-publicised early monarch butterfly studies of Losey et al., (1999) and Jesse and Obrycki (1999). Subsequent studies compared the effects of event 176 pollen with that of other Bt-maize hybrids in which the gene was controlled by the CaMV 35S promoter (referred to as MON810 or Bt11 hybrids). Pollen from the latter hybrids had negligible effects on monarch larvae (Stanley-Horn et al., 2001; Hellmich et al., 2001). The registration of maize hybrids derived from event 176 will terminate in the United States in 2001 (Stanley-Horn et al., 2001).

New biotechnological developments aimed at eliminating the problems associated with transgene DNA and expression of novel proteins in pollen are discussed in Section 4. The motivation for these developments centres mainly on concerns about gene flow to other plants and allergens in pollen, rather than for bees and bee products.

In summary, it is possible that GM material could occur in pollen harvested for human use from hives placed where bees may forage from GM crops. The actual concentration will depend on how much GM pollen is taken relative to pollen from other plants and how much transgene DNA or novel protein the GM plant produces in its pollen. Similarly, pollen containing GM material could also be present in honey harvested from bees foraging on GM crops.

2.1.2 Nectar

Nectar is a plant secretion, rather than a tissue, and has no cellular content. As such, transgene DNA is not likely to occur in nectar and there are no records of any RNA or DNA in nectar. Most nectars are also free of protein, being composed principally of sugars and sometimes free amino acids (Baker and Baker, 1973; R. Bieleski. pers. comm.). There are exceptions to this however. Recently, tobacco plants have been found to secrete a limited array of proteins to a concentration of about 0.024% protein of total nectar (Carter et al., 1999). Leek nectar has also been shown to contain two proteins, a lectin and an alliinase, comprising about 0.022% of the nectar (Peumans et al., 1997). Both of these proteins were inactivated or degraded when leek nectar was made into honey by bees, presumably by the action of enzymes in the bee's honey stomach. Incidentally, these authors also found two new proteins in the honey that had no equivalents in leek nectar and concluded that these must have come from the bees themselves (e.g. enzymes secreted by the honey stomach).

Consequently, it is theoretically possible that some GM plants could secrete novel proteins with their nectar, although their concentrations are likely to be very low. Consequently, bees could gather such nectar, but the honey they make from it may not necessarily contain the nectar proteins in active form.

Presumably because of nectar's low protein content, there has been only one record of an examination of nectar from a GM plant. Jouanin et al. (1998) noted that Bowman-Birk soybean trypsin inhibitor (BBI) could not be detected in the nectar (or pollen) of transgenic oilseed rape plants containing the BBI gene under the control of CaMV 35S promoter.

2.1.3 Plant resins and gums

There are no records of DNA or RNA being detected in plant gums or resins. Proteins, however, have been recorded from some of these plant secretions. For example, gum arabic (from Acacia species) is 2-4% protein (Menzies et al., 1996) and proteins have been recorded from pea root mucilage and rye root exudate (e.g. Knee et al., 2001; Siciliano et al., 1998). There are no references to the presence of either DNA or RNA in propolis. The composition of propolis varies from sample to sample due to the variety of plant resins and gums utilised by the bees and the collection techniques used by beekeepers to obtain propolis from the hive. One report describes a propolis with a protein content of about 2.5 % (Tuha and Simuth, 1991). Novel proteins from GM plants may conceivably find their way into the gums, exudates and resins that bees collect to make propolis, but there is, as yet, no published evidence to support this idea.

2.1.4 Plant sap and honeydew

Honeydew is excreted by sap-sucking insects such as aphids. As well as the sugars that predominate, phloem sap also contains free amino acids, small peptides and sometimes proteins (Salvucci et al., 1998). For example, phloem exudates from squash, cucumber and castor oil plant have been shown to contain proteins, some of which appear to be important in the transport of plant viruses (Christeller et al., 1998; Kruger et al., 2000; Schobert et al., 2000; Owens et al., 2001). Similarly plant mRNA (messenger RNA) is known to circulate in plants (translocate) via the phloem (Oparka and Santa Cruz, 2000). Thus, phloem sap from GM plants could conceivably contain both transgene mRNA and novel proteins. There are no published studies describing the fate of plant RNA after digestion by plant-sucking insects and no records of RNA or DNA in the honeydew they excrete. However, a DNA-tracking study has shown that squash leaf curl virus can pass intact through the guts and into the honeydew of whiteflies (Rosell et al., 1999). It is commonly thought that sucking insects lack digestive proteases and utilise only the free amino acids in sap for their nitrogen needs (e.g. Rahbé et al., 1995; Sandstrom and Moran, 2001). From this it might be concluded that sap proteins would pass into honeydew intact. However, when whiteflies were fed with labelled cotton leaf proteins, these were digested and excreted only as amino acids in the resultant honeydew (Salvucci et al., 1998). Thus it remains debatable whether or not novel proteins from transgenic plants could find their way into honeydew and if they would persist in honey made from it.

2.2 Records of GM Material in Bee Products

Currently two methods may be used to determine the GM status of foods. With the polymerase chain reaction (PCR) test for transgene DNA, a "primer" (consisting of a piece of DNA with a sequence that could only occur in the transgene) is added to a sample of the food to be tested. If the primer matches any DNA in the sample, then the PCR will cause this DNA to be "amplified". The amplified DNA can then be stained and visualised to give an indication that the sample contains some transgene DNA. With quantitative PCR, the concentration of transgene DNA in the sample may also be estimated. Theoretically this method is very sensitive and can detect even one or two pieces of transgene DNA in a sample. In practice, its sensitivity will depend on the nature of the food being tested and the transgene DNA sequence that is being sought. For example, a study of GM soyabeans and maize (Lin et al., 2000) showed that PCR using a CaMV 35S primer had a detection limit of 0.1% w:w of GM soyabeans, but with a nos (nopaline synthase) primer the test had a limit of 1%. With a CDPK-cry (maize calcium-dependent protein kinase promoter with Bt toxin) primer, PCR had a detection limit of 0.1% w:w for GM maize and with a cry1Ab (Bt toxin) primer the limit was 2%.

The second detection method uses enzyme-linked immunosorbent assay (ELISA) to detect novel proteins in food. With this method, an antibody to the novel protein in question is prepared and linked to an enzyme which catalyses a reaction resulting in a coloured end-product. The enzyme-linked antibody is added to the food sample and, if the novel protein is present, it links to the protein and cannot be washed away. The colour reagents are then added and the intensity of the coloured end-product gives a measure of the concentration of the novel protein in the food.

Pollen represents the most likely source of transgene DNA and novel proteins in bee products. It is also commonly present in the most widely-consumed bee product, honey. Because of this, the only attempts to measure GM material in bee products have focussed on honey and its pollen "contaminant". There are no records of attempts to detect GM material in honey from honeydew, pollen intended as human food or propolis.

Most New Zealand honeys (including clover honey) contain between 20,000 and 100,000 pollen grains per 10 g (Moar, 1985). This is considered the "normal range", but some honeys have pollen concentrations above or below this range. Molan (1998) states that concentrations as low as 500 grains and as high as 5 million per 10 g are possible. In the United Kingdom, shop-bought honeys were found to contain between 20,000 and 80,000 grains per 10 g (Anon A, 1998). Eady et al. (1995) found 100,000 pollen grains per ml in a commercial honey derived from garden flowers (in the UK).

If we assume that an "average" pollen grain weighs 0.03m g (Stanley and Linskens, 1974), then these figures translate as follows:

• 500 grains per 10 g is equivalent to 0.00015% w:w pollen in honey
• 20,000 grains per 10 g is equivalent to 0.006% w:w pollen in honey
• 80,000 grains per 10 g is equivalent to 0.024% w:w pollen in honey
• 100,000 grains per 10 g is equivalent to 0.03% w:w pollen in honey
• 100,000 grains per ml is equivalent to 0.3% w:v pollen in honey
• 5 million grains per 10 g is equivalent to 1.5% w:w pollen in honey.

The stability of transgene DNA and novel proteins in GM pollen stored in honey has been assessed using pollen from modified tobacco and Arabidopsis plants with marker genes on pollen specific promoters (Eady et al., 1995). A PCR test showed that transgene DNA remained "relatively intact" even after seven weeks in a commercial honey sample. Similarly, novel protein was detected unchanged after six weeks in honey. The authors pointed out that the experimental system they used represented a "worst-case scenario" for the presence of GM material in honey and that "the concentration of a given, potentially toxic pollen-borne protein is expected to be very low in natural honey made from nearby transgenic plants". However, they also pointed out that even vanishingly small quantities of some proteins may cause allergic reactions.

There are only two published studies of attempts to measure GM material in natural honey made by bees foraging near GM plants. The first was carried out by the UK Ministry of Agriculture, Fisheries and Food (now the Department for Environment, Food and Rural Affairs) (Anon A, 1998). In this study a sample of honey was taken from a "hive close to a transgenic oilseed rape field", pollen extracted from it and ELISA used to quantify the amount of npt II protein (which confers kanamycin resistance) present. Two readings from the single sample gave a mean of 1.61 ng per mg (0.00016% w:w) of total protein in the pollen sample. A control sample of non-transgenic oilseed rape honey was not included. Pollen samples taken from two GM tobacco plants containing the same transgene construct (nptII and nos promoter) gave mean readings of 35.1 pg and 1.39 ng of nptII protein per mg of total protein (3.51 x 10^–6% and 0.000139%, respectively). This report also stated that DNA could be extracted from a number of commercially available honey samples and from honey derived from transgenic oilseed rape (presumably the same sample used in the protein analysis), but the details of this part of the study are not given in the report.

The second study was commissioned by Friends of the Earth in the UK (Anon B, 2000; Anon C, 2000), who were concerned that pollen from unidentified GM-oilseed rape field research sites could occur in honey without beekeepers' knowledge. They purchased 11 jars of locally produced honey and honey comb from retail outlets in an area of England where GM herbicide-tolerant oilseed rape crops had been trialed. Each sample was checked for oilseed rape pollen content and nine of the samples found to contain significant quantities (actual amounts and detection limits not stated) of this pollen. Sub-samples were taken from these honeys and sent to the Austrian Federal Environmental Agency Laboratory. PCR was used to determine whether any of the samples contained DNA sequences corresponding to the herbicide-tolerance gene (bar or pat gene) and the nos gene promoter or terminator used in such GM oilseed rape plants. Two of the nine samples gave positive results for the pat gene and the nos promoter, suggesting that pollen from GM oilseed rape had found its way into honey. Unfortunately, Friends of the Earth could not afford to have a quantitative analysis conducted which would have shown how much GM pollen might be in the honey (Anon B, 2000). The apparent lack of appropriate controls in this study (honey from a region where only non-GM oilseed rape was grown) is also unfortunate. The bar, pat and nos genes are all derived originally from common bacteria (Streptomyces and Agrobacterium) (A. Gleave, pers. comm.; Wehrmann et al., 1996), suggesting that they may commonly contaminate natural products such as honey. Bar, pat or nos DNA from these bacteria would also give positive results with the PCR tests conducted. While it may be argued that this is unlikely, inclusion of suitable controls would have removed all doubt from this study.

The results of this honey study have subsequently been used by Friends of the Earth and UK beekeepers to call for a halt to field research on GM plants, particularly a series of "farm-scale" trials of GM herbicide-tolerant crops which commenced in the UK in 1999. Diamand (1999) gives further details of this campaign.

Contact for Enquiries

Dr Sharon Adamson
Manager, Innovation Policy
Ministry of Agriculture and Forestry
PO Box 2526
Wellington
NEW ZEALAND

Phone: +64 4 894 0618
Fax: +64 4 4 894 0741
Contact this person