|

Toxicity research report final-

Fireweed Toxicity
Facts and
Steven M Colegate BSc(Hons), PhD Photos: Forest&Kim Starr Fireweed Toxicity Facts and Perspectives Executive Summary
• HPLC-MS analysis of fireweed collected in the Bega Valley (NSW) in the spring of 2006 and 2008 showed the presence of dehydropyrrolizidine alkaloid esters at levels up to about 220 milligrams per kilogram of plant. • Due to the adverse effects of dehydropyrrolizidine alkaloid esters, a comprehensive assessment of livestock health, welfare and productivity in the infested areas should be undertaken. • Animal-derived food for humans, especially bee-products and honey produced by bees foraging on fireweed, require monitoring for the presence of fireweed pyrrolizidine alkaloids. • All possible measures should be undertaken to manage and control the infestation and potential spread of fireweed in Australia. Fireweed Toxicity Facts and Perspectives • Senecio madagascariensis (fireweed) is a non-indigenous invasive weed that is expanding its range in Australia. • It is closely related to the indigenous Senecio lautus in both morphology and its distribution. • The dehydropyrrolizidine alkaloid esters have been associated with acute and chronic liver toxicity, pneumotoxicity, carcinogenicity and genotoxicity. • Toxic dehydropyrrolizidine alkaloids have been detected in samples of fireweed collected from the Bega Valley in the spring of 2006 and 2008. • The alkaloid profiles of young plant and mature green-stemmed plant and mature red-stemmed plant are very similar. • In this limited survey, young plants contain higher levels (up to about 220 milligrams per kilogram of plant) of the toxic pyrrolizidine alkaloids than the mature plant (up to about 110 milligrams per kilogram of plant). • The presence of alkaloids was confirmed using ion trap mass spectrometry and time of flight mass spectrometry. Additionally a sample was analyzed by another laboratory experienced in Senecio madagascariensis. • The suspected presence of a low level of a toxic pyrrolizidine alkaloid, associated with fireweed, in a single, blended wild honey sample of unclear heritage remains unconfirmed. Fireweed Toxicity Facts and Perspectives CONTENTS
Executive Summary . 1 Executive Summary . 2 Summary . 3 List of Figures . 5 Introduction . 7 Potential Toxicity Issues . 8 Toxicology of Dehydropyrrolizidine Alkaloids (DHPAs) . 8 Potential Toxicity of DHPAs for Livestock . 11 Potential Toxicity of DHPAs for Humans . 12 Relevance of DHPA Content and Toxicity to Fireweed . 13 General Approach and Examples . 16 Analysis of Fireweed . 19 Ion trap mass spectrometry . 19 Time-of-flight mass spectrometry . 24 Analysis of Wild Honey from the Bega Valley . 27 Conclusions and Recommendations . 28 References . 30 Fireweed Toxicity Facts and Perspectives List of Figures
Figure 1:
Common structural cores of dehydropyrrolizidine alkaloid esters Figure 2:
Structures of pyrrolizidine alkaloids isolated from Senecio madagascariensis sourced from Hawaii and Australia (Gardner et al., 2006). The bold number is the mass of the molecular ion adduct
(MH+) for each alkaloid. Figure 3:
HPLCesiMS ion chromatograms showing alkaloidal profiles for Echium plantagineum flowers (A) and leaves (B). Each of the peaks
represents a different pyrrolizidine alkaloid. In this case the alkaloids are in their N-oxide form. Figure 4:
Base ion (m/z 200 – 500) HPLC-ESI-MS chromatograms for (A)
pollen loads from bees foraging on Echium vulgare; (B) pollen loads
from bees foraging on Eupatorium cannabinum; (C) pollen, hexane-
washed from anthers of Senecio jacobaea; and (D) pollen loads from
bees foraging on Senecio ovatus. The internal standard (IS) is used to standardize the chromatograms. The presence of different suites of alkaloids from each sample is evident. The alkaloids are in their N- Figure 5:
The HPLC-MS ion chromatogram of an extract of honey produced by bees that foraged in the vicinity of Echium vulgare. Each of the six "arrowed" peaks is a hepatotoxic pyrrolizidine alkaloid that is also found in the plant itself. Figure 6:
HPLC-esiMS base ion (m/z 200-500) chromatograms of the alkaloidal components of Senecio madagascariensis (fireweed): A. fresh young
plant; B. fresh mature plant, and C. dried and milled mature plant.
Figure 7:
HPLC-esi-ion trap MS base ion (m/z 200-500) chromatograms of: A.
fresh young fireweed; B. fresh mature fireweed; and C. dried and
milled mature fireweed. Also shown are example mass spectra for selected peaks: a. Unidentified alkaloid, m/z 424 (1); b. N-oxide of
senecionine type alkaloid, m/z 352 (2); c. N-oxide of a retrorsine type
alkaloid, m/z 368 (3).
Fireweed Toxicity Facts and Perspectives Figure 8:
HPLC-esi-ion trap MS base ion chromatograms of: A. an extract of
young Senecio madagascariensis spiked with; B. integerrimine-N-
oxide, C. usaramine, D. retrorsine, and E. senecionine-N-oxide.
Figure 9:
An HPLC-MS comparison of a zinc/acid-reduced extract of fireweed collected in the Bega Valley, New South Wales in about October 2006 with that collected in northern New South Wales (Gardner et al., 2006). The bold numbers are the MH+ values for each peak (see
Figure 2). The analyses were done by Dr Dale Gardner of the USDA
Poisonous Plants Research Laboratory in Logan, Utah, USA Figure 10:
Direct infusion, ESI-TOF MS analysis of fresh Senecio madagascariensis collected in the Bega Valley, October 2008. A.
young plants; B. mature, red-stemmed plants; and C. mature, green-
Figure 11.
HPLC-TOF mass spectrometric analysis of fresh, green-stemmed mature samples (green trace) and fresh, red-stemmed mature samples
(red trace) of Senecio madagascariensis. The numbers shown are the
masses of the ions observed. Fireweed Toxicity Facts and Perspectives Introduction

In North America, "fireweed" is a trivial name used for the purple-flowered Epilobium angustifolium and Erechtites hieracifolia. In Australia, in addition to "fireweed" being used to describe a toxic blue-green algae (Lyngbya majuscule), "fireweed" can be a generic, trivial name applied to a number of weed species in the Senecio genus of the Asteraceae (daisy/thistle) Family. In the context of this report "fireweed" refers to Senecio madagascariensis Poiret that is spreading prolifically along the east coast of Australia and specifically, in the particular context of this report, the Bega Valley in New South Wales where it is a declared plant under the Noxious Weeds Act 1993 (www.dpi.nsw.gov.au/agriculture/ pests- Senecio madagascariensis, a native of South Africa, is closely related to S. lautus which is native to Australia (www.esc.nsw.gov.au/weeds/Sheets/herbs/H%20Fireweed.htm) and is one of over 1200 Senecio species world-wide. It was first observed in the Hunter Valley region of Australia in about 1918. Apparently, it is not a declared noxious plant in this region Since Senecio spp. produce alkaloids that are toxic to livestock and humans, and since previous phytochemical analyses of S. madagascariensis in particular have shown the presence of hepatotoxic (liver damaging) alkaloids, it was deemed essential to complete a phytochemical assessment of the fireweed collected in the Bega Valley of New South Wales. This report, and the research described within, have been completed at the request of the New South Wales, Bega Valley Fireweed Committee and is intended to be complementary to related reports (also commissioned by the Bega Valley Fireweed Committee) that describe other aspects of this plant such as its identification, growth habits, dispersal rates and distribution, and methods of control (if not eradication) including biological options for control. Fireweed Toxicity Facts and Perspectives Herein, potential problems associated with the expected presence of dehydropyrrolizidine alkaloid (DHPA) esters in the fireweed are discussed. The results of a limited phytochemical research project aimed at identifying the complement of toxic DHPAs in the fireweed and in wild honey gathered in the vicinity of fireweed are presented and discussed. Potential Toxicity Issues
The potential toxicity issues for livestock and humans associated with fireweed are due to the likely presence of DHPA esters. This diverse class of natural chemicals comprises over 400 different entities occurring in over 6000 species of plants world- The basic structural variations of the alkaloids include monoesters, open chain diesters and macrocyclic diesters of the free base and corresponding N-oxide forms of dehydropyrrolizidine-based and the otonecine-based alkaloids (Figure 1).
There exists a high level of variation in the suite of DHPAs found in different plant genera and different species. There can even be variation in the DHPA complement between different populations of the same species. Toxicology of Dehydropyrrolizidine Alkaloids (DHPAs)
Following absorption into the blood stream, the DHPA esters are rapidly metabolized by the liver to form the didehydropyrrolizidine alkaloid form commonly referred to as
the "pyrrolic" form. It is this pyrrolic form that then reacts with macromolecules within the liver (eg., proteins and DNA) to initiate the chain of events culminating in clinical disease (Stegelmeier et al., 1999). The DHPAs are primarily liver toxins (hepatotoxic) but can also "spill out" of the liver to damage the lungs (pnuemotoxic). In some animals, some DHPAs have been shown to cause cancers (carcinogenic) and damage genes (mutagenic). Fireweed Toxicity Facts and Perspectives Not all DHPAs are as toxic as each other. Not all animal species are equally susceptible to the DHPAs. This is due, in the main, to the relative ease (or difficulty) with which the DHPAs are metabolized to the reactive pyrrolic forms. For example, sheep are relatively resistant to the effects of DHPAs and, as such, have been recommended as grazing control options (Schmidl, 2006; www.dpi.nsw.gov.au / data/assets/pdf_file/0007/49840 /fireweed _-_primefact_126-final.pdf). On the other hand, cattle and horses are relatively susceptible to the toxic effects of ingested Even without knowing the reasons why, the toxicity of Senecio spp. to livestock was first recognized in the early 20th century. Similarly, the danger to humans was first recognized in 1920 when cirrhosis of the liver was attributed to senecio poisoning (Willmot and Robertson, 1920). Subsequently there has been extensive research into the detection of DHPAs and the determination of their mechanism of action and their toxicities. A relatively recent review states: "Pyrrolizidine alkaloids are the leading plant toxins associated with disease in humans and animals. The PAs present a serious risk to human populations . Some PA adducts are persistent in animal tissues and the metabolites may be released and cause damage long after the initial period of ingestion. " (Prakash et al.,1999). Fireweed Toxicity Facts and Perspectives Macrocyclic diester Otonecine diester Figure 1: Common structural cores of dehydropyrrolizidine alkaloid esters
Fireweed Toxicity Facts and Perspectives Potential Toxicity of DHPAs for Livestock The acute and chronic effects of exposure of livestock to dietary DHPAs have been well-documented (for example: Stegelmeier et al., 1999). The major effect is on the liver and many clinical signs are related to this initial insult. For example, chronic poisoning of horses by dietary DHPAs causes a neurological syndrome that is related to increased levels of systemic ammonia (particularly in the brain) as a result of the damaged liver's decreased capacity to process ammonia to urea. While cattle, horses, pigs and chickens are quite susceptible to the poisonous effects of the DHPAs, sheep and goats are fairly resistant. However, sheep are very susceptible to copper poisoning and exposure to DHPAs can exacerbate the accumulation of copper in the liver of the sheep. Eventually an acute, copper-induced haemolytic crisis occurs, as the red blood cells are disrupted, causing the death of the animal. Clinical signs of chronic exposure to DHPAs vary depending upon the animal exposed but can be attributed to the loss of liver function. Signs of exposure can include a rough, unkempt appearance, oedema of gastro-intestinal tract, ascites (fluid- filled body cavities eg the peritoneal cavity), diarrhea, prolapsed rectum, dullness and lethargy, photosensitization and general abnormal behaviour. As the liver disease progresses then jaundice will be observed. A summary of clinical signs resulting from exposure to DHPAs is presented in the Merck Veterinary Manual On post-mortem examination, the livers in most species are shrunken, hard to the touch and fibrotic in nature following chronic exposure to the DHPAs. In sheep, however, the livers can be soft and mushy rather than hard. Microscopic indicators of exposure to the DHPAs include the observation of enlarged liver cells (hepatocytes) referred to as megalocytosis. The long term prognosis for affected livestock is not good due to the primary effect on the liver. It has also been shown that the fatal effects can be delayed with no immediate signs of liver damage following the initial exposure. For example, steers given a single sub-lethal dose of Senecio riddellii, S. jacobaea or S. longilobus died from a progressive liver failure several months later (Molyneux et al., 1988). Fireweed Toxicity Facts and Perspectives Using sheep as grazing control agents for the DHPA-producing plants is not without its risks to the sheep since, although relatively resistant to the effects of the DHPAs, they are not totally resistant. As a result, the use of sheep as biocontrol agents may be unacceptable on animal welfare grounds (www.bbc.co.uk/dna/h2g2/A22548774). It has not been positively confirmed that such "biocontrol" animals do not accumulate toxic metabolites of the DHPAs in meat destined for the human food supply. The lethal effects (and economic losses associated with them) of acute or chronic exposure of livestock to the DHPAs are not expected to be the major source of loss to producers. Low-level, sub-clinical exposures can lead to severe losses in productivity that may include poor feed utilization and reduced weight gains, reduced milk yields and potential reproductive/breeding adverse effects. Potential Toxicity of DHPAs for Humans A review published in 1989 clearly demonstrated the exposure of humans to the DHPAs and the adverse consequences (Huxtable, 1989). Since then, there have been continuing reports of humans being poisoned by dietary DHPAs, including exposure via medicinal herbs preparations, cooking spices and flour made from grain contaminated with seeds from DHPA-producing plants. The major, reported clinical symptoms of pyrrolizidine alkaloid intoxication in humans result from a veno-occlusive disease of the liver (Prakash et al., 1999). In addition to outbreaks of "bread poisoning", when the seeds of pyrrolizidine alkaloid- producing plants were co-harvested with milling grains (Ahmad, 2001), pyrrolizidine alkaloids in herbs and spices have been associated with many instances of fatal hepatic disease in adult, neonatal and prenatal humans. Whilst the effects of acute exposure to pyrrolizidine alkaloids on humans and animals have been well documented, the effects of long term, low level exposure to DHPAs via the diet (grain, milk, meat, honey, and related products) (Colegate et al., 1998) are a relatively unknown factor. In addition, and for many reasons, adverse clinical effects can be very difficult to unambiguously associate with the sources of dietary DHPAs. After a comprehensive risk assessment of toxic pyrrolizidine alkaloids, the German Federal Health Bureau established regulations that restrict oral exposure to Fireweed Toxicity Facts and Perspectives pyrrolizidine alkaloids or their N-oxides in herbal preparations to 0.1 microgram per day with the exclusion of pregnant and lactating women for which zero exposure is recommended (German Federal Health Bureau, 1992). In the Netherlands, levels of pyrrolizidine alkaloids and their N-oxides are restricted to 0.1 microgram per 100 g of food (H van Egmond, The Netherlands National Institute for Public Health and the Environment, personal communication). In Australia, the Food Standards Australia New Zealand (FSANZ, 2001) has set a limit of exposure to 1 microgram per kilogram bodyweight per day. For example, a 30 kg child would be allowed 30 micrograms of DHPAs (or their N-oxides) per day. In contrast to the European deliberations, the FSANZ did not consider DHPA-induced carcinogenicity or genotoxicity as a major factor with humans and therefore developed their recommendations on the occurrence (or prevention of the occurrence) of hepatic veno-occlusive disease. In addition to the clearly proven contamination of grains, and the use of cooking spices and medicinal herbs either prepared or contaminated with DHPA-producing plants, the potential exists for humans to be exposed to DHPAs (and their N-oxides) via honey and bee products (Edgar et al., 2001; Beales et al., 2004; Betteridge et al., 2005; Boppré et al., 2005), milk (James et al., 1994), eggs (Edgar and Smith, 2000) and meat (Seawright, 1994). Relevance of DHPA Content and Toxicity to Fireweed
In Australia, grazing of fireweed (Senecio madagascariensis)-contaminated pasture has previously been associated, but not proven, with livestock poisonings in New South Wales (Seaman and Walker, 1985). Another report describes pyrrolizidine alkaloidosis in a 2 month-old foal and associated it with in utero exposure of the foetus to DHPAs ingested by the mare grazing a pasture heavily infested with fireweed (Small et al., 1993). An intoxication of cattle was associated with S. lautus (Noble et al., 1994) but, apparently, subsequent re-examination of the herbarium specimens identified the plant as S. madagascariensis (referred to in Gardner et al., Despite the previous associations of fireweed with overt disease of livestock, it was only recently that the DHPAs, in fireweed collected in Hawaii and in Australia, were Fireweed Toxicity Facts and Perspectives identified and compared (Gardner et al., 2006). This report, that used gas chromatography-mass spectrometry (GCMS) to identify 12 macrocyclic DHPAs (Figure 2) at levels that were potentially toxic to livestock (217 – 1990 micrograms
per gram of plant material). The DHPA profile for the fireweed collected in Hawaii was essentially identical to the profile determined for a composite sample of fireweed collected from northern New South Wales. Citing this latter observation the authors supported the contention that fireweed in Hawaii arrived via Australia. The research by Gardner et al (2006) on the fireweed in Hawaii highlighted the high variation in DHPA content and profile from plants collected at different locations. Therefore, to help complete the understanding of the fireweed population, and problems posed by it, in the Bega Valley area of New South Wales, a limited phytochemical analysis of the plants collected in the area was conducted. Fireweed Toxicity Facts and Perspectives Desacetyldoronine Acetylsenkirkine Figure 2: Structures of pyrrolizidine alkaloids isolated from Senecio
madagascariensis sourced from Hawaii and Australia (Gardner et al., 2006). The bold number is the mass of the molecular ion adduct (MH+) for each alkaloid.
Fireweed Toxicity Facts and Perspectives Analysis for Dehydropyrrolizidine Alkaloids
In mid to late spring of 2006, samples of mature, flowering plants and juvenile plants were collected in the Bega Valley and immediately sent to Dr Steven Colegate, Leader of the CSIRO Plant-associated Toxins Research Laboratory in Geelong, Victoria, for processing and analysis. In mid spring of 2008 other samples of plants and samples of wild honey collected in the same area were sent to Dr Steven Colegate for analysis at Deakin University once the CSIRO had closed down its plant toxins research group. The plants included mature flowering plants, some with a red hue to the stems and some just green, and juvenile plants. There is no botanical separation at this stage of the green-stemmed fireweed and the red-stemmed fireweed. All samples were received in a fresh condition and were immediately processed for The earlier samples processed at CSIRO were analysed using high pressure liquid chromatography (HPLC)-electrospray ionization (ESI) ion trap mass spectrometry (MS) while the samples analyzed at Deakin University were processed using HPLC- ESI-time of flight (TOF) MS. General Approach and Examples
The plant and honey samples were analyzed for the presence of DHPAs using a sequential combination of solvent extraction, solid phase concentration and subsequent high pressure liquid chromatography-mass spectrometry (HPLC-MS) in a manner previously described for: • plants eg., Echium plantagineum (Paterson's Curse, Salvation Jane) (Colegate et al., 2005) (Figure 3), Echium vulgare, Senecio jacobaea (ragwort) and
Senecio ovatus (Boppré et al., 2008) • pollen collected from plants or bees eg., Echium vulgare (Boppré et al., 2005), Eupatorium cannabinum, Senecio jacobaea and Senecio ovatus (Boppré et al., 2008) (Figure 4)
• honey (Betteridge et al., 2005) (Figure 5)
Fireweed Toxicity Facts and Perspectives In all examples shown, and in the analyses of fireweed in the following section, the individual DHPAs are represented by peaks in an ion chromatogram. Identities are rationalized using the mass spectrometric data collected for each peak and subsequent comparison with authentic standards of those alkaloids or with mass data reported in the scientific literature. In general, because the alkaloids within a species of plant are structurally similar, the size of the peak reflects the relative abundances of the individual pyrrolizidine alkaloids in the sample. Figure 3:
HPLCesiMS ion chromatograms showing alkaloidal profiles for Echium plantagineum flowers (A) and leaves (B). Each of the peaks represents a
different pyrrolizidine alkaloid. In this case the alkaloids are in their N-oxide form.
In contrast to the GC-MS method employed by Gardner et al (2006), this HPLC-MS method of analysis allows the simultaneous extraction, recovery and analysis of the pyrrolizidine alkaloids and their N-oxides. Since the N-oxides are also toxic (Chou et al., 2003) it is important that they are fully accounted.
The ion chromatograms shown in Figure 3 demonstrate how different parts of the
same plant can have slightly different alkaloid profiles. In this case the flowers produce acetylated derivatives of the alkaloid-N-oxides detected in the leaves (Colegate et al., 2005). However, Figure 4 clearly shows the different profiles of
Fireweed Toxicity Facts and Perspectives Time
(min)
Figure 4:
Base ion (m/z 200 – 500) HPLC-ESI-MS chromatograms for (A)
pollen loads from bees foraging on Echium vulgare; (B) pollen loads from bees
foraging on Eupatorium cannabinum; (C) pollen, hexane-washed from anthers of
Senecio jacobaea; and (D) pollen loads from bees foraging on Senecio ovatus. The
internal standard (IS) is used to standardize the chromatograms. The presence of
different suites of alkaloids from each sample is evident. The alkaloids are in their N-
oxide forms.
alkaloids that might be expected from different plant genera i.e., Echium versus Senecio versus Eupatorium, and even from different species of the same genus i.e., Senecio jacobaea versus Senecio ovatus (Boppré et al., 2008). As a final example of the application of this methodology, Figure 5 shows the
detection of hepatotoxic pyrrolizidine alkaloids detected in honey produced by bees that foraged in the vicinity of the DHPA-producing plant Echium vulgare (Betteridge et al., 2005). The levels detected in this commercial honey would clearly exceed European and Australian regulations. Fireweed Toxicity Facts and Perspectives 22/05/2004 12:04:15 Honey N ov.2002-5(Arataki clover blend) elute 1 (1Eho2) R T: 0.00 - 19.99 m /z= 200.0-500.0 MS Figure 5:
The HPLC-MS ion chromatogram of an extract of honey produced by bees that foraged in the vicinity of Echium vulgare. Each of the six "arrowed" peaks
is a hepatotoxic pyrrolizidine alkaloid that is also found in the plant itself.
Analysis of Fireweed
Samples of fireweed were processed and analysed using two different mass spectrometry approaches: 1. ion trap mass spectrometry, and 2. time of flight (TOF) mass spectrometry. Ion trap mass spectrometry Qualitatively, there were no major differences observed in the alkaloid profiles of the mature plant and the juvenile plants collected in 2006 and analyzed using HPLC-ESI- ion trap MS i.e., the chromatograms looked very similar. Within the profiles, several, potentially toxic DHPAs were identified (Figure 6) in
extracts of the Bega Valley fireweed. The structures of the alkaloids were indicated by comparison of their mass spectrum characteristics (for example, Figure 7) with the
Fireweed Toxicity Facts and Perspectives 500.0 MS yu260906lc 500.0 MS yu260906lc Figure 6:
HPLC-esi-ion trap MS base ion (m/z 200-500) chromatograms of the alkaloidal components of Senecio madagascariensis (fireweed): A. fresh young plant;
B. fresh mature plant, and C. dried and milled mature plant.
literature report (Gardner et al., 2006) and using authentic standards (Figure 8) from
the collection of the CSIRO Plant Toxins Research Group. The major peaks with MH+ ions at m/z 368 and 352 all showed a significant dimer ion signal at m/z 735 and 703 respectively. This is good evidence for the N-oxide character (Figure 1) of the alkaloids (Colegate et al., 2005), in this case the N-oxides
of the usaramine-type alkaloids (MH+ 352) and senecionine-type alkaloids (MH+ 336) respectively (see Figure 2). On the contrary, the major peaks with MH+ ions at m/z
382 and 424 did not show the presence of dimeric ions and this is supportive of their structures being of the otonecine type (Figure 1) such as otosenine and florosenine
respectively (see Figure 2).
Fireweed Toxicity Facts and Perspectives
Figure 7:
HPLC-esi-ion trap MS base ion (m/z 200-500) chromatograms of: A.
fresh young fireweed; B. fresh mature fireweed; and C. dried and milled mature
fireweed. Also shown are example mass spectra for selected peaks: a. Unidentified
alkaloid, m/z 424 (1); b. N-oxide of senecionine type alkaloid, m/z 352 (2); c. N-oxide
of a retrorsine type alkaloid, m/z 368 (3).
Fireweed Toxicity Facts and Perspectives 460.4 461.3 461.4 Base Peak m/z= 200.0-500.0
Figure 8:
HPLC-esi-ion trap MS base ion chromatograms of: A. an extract of
young Senecio madagascariensis spiked with; B. integerrimine-N-oxide, C.
usaramine, D. retrorsine, and E. senecionine-N-oxide.
The alkaloid profile observed for the Bega Valley fireweed samples was similar but seemed sufficiently different from the report on the profiles of both the Hawaiian and Australian (northern NSW) samples previously analyzed (Gardner et al., 2006). It was the apparent close similarity in these latter samples that prompted the conclusion that the Hawaiian fireweed was originally from Australia. Therefore, a sample of the extract of the Bega Valley plants was sent to Dr Dale Gardner of the USDA Poisonous Plants Laboratory in Logan, Utah, USA for comparative analysis with their sample of Australian fireweed collected in northern New South Wales (Figure 9).
Fireweed Toxicity Facts and Perspectives RT: 0.00 - 20.00 SM: 3B Northern NSW
0.40 1.73 2.43 2.87 17.04 17.28 19.69 Bega Valley
350 10.07
350 8.26 8.80
3.08 4.65 5.44 5.55 6.97 Figure 9: An HPLC-MS comparison of a zinc/acid-reduced extract of fireweed
collected in the Bega Valley, New South Wales in about October 2006 with that
collected in northern New South Wales (Gardner et al., 2006). The bold numbers are
the MH+ values for each peak (see Figure 2). The analyses were done by Dr Dale
Gardner of the USDA Poisonous Plants Research Laboratory in Logan, Utah, USA.
The major differences evident between the two Australian samples include the greater levels of the senecionine-type alkaloids (MH+ 336, Figure 2), the relative imbalance
of the retrorsine-type alkaloids (MH+ 352, Figure 2) and the apparent appearance of
dehydroretrorsine-type alkaloids (MH+ 350) in the Bega Valley sample. The latter are possibly related to seneciphylline and have not been previously reported in this plant. Fireweed Toxicity Facts and Perspectives In a similar way to the quantitation of alkaloids in honey and pollen (Betteridge et al., 2005; Boppré et al., 2005; Boppré et al., 2008), the approximate concentrations of the DHPAs and their N-oxides were estimated by comparison to calibration curves generated using the authenticated DHPA lasiocarpine and lasiocarpine-N-oxide respectively. Thus, fresh young plant was estimated to contain between 100 and 220 ppm (milligrams of alkaloids per kilogram of plant) and mature fresh plant between 70 and 110 ppm. Another sample of mature plant that was dried and milled to a fine powder was estimated to contain between 140 and 170 ppm of pyrrolizidine alkaloids. In this very limited survey, it appears that young fresh plant contains more pyrrolizidine alkaloids than the fresh mature plant. The higher levels estimated in the dried and milled mature plant reflects the increased efficiency of extraction. Significantly, there was no apparent selective degradation of most of the alkaloids during the drying process however, significant loss of the alkaloid with MH+ 424 did seem to occur. This observation would require further checking for confirmation. The levels of pyrrolizidine alkaloids detected in the Bega Valley samples was at the lower end of the scale (217 – 1990 ppm) for the levels detected in samples collected from various sites in Hawaii (Gardner et al., 2006). It is unlikely that this results from differences in extraction efficiency but may reflect the relatively limited sampling procedure employed for the Bega Valley samples, combined with inter-plant differences in levels. Time-of-flight mass spectrometry Samples of fresh fireweed (entire plant) were extracted in the usual way and analyzed using TOF mass spectrometry to complement the results obtained using ion trap mass spectrometry. Three plant samples, collected in spring of 2008, were analyzed: 1. mature fireweed plants with green stems; 2. mature fireweed plants with red stems; and 3. small (juvenile) fireweed plants. Fireweed Toxicity Facts and Perspectives + Scan (0.182 min) JFW100.d Counts vs. Mass-to-Charge (m/z) + Scan (0.194 min) RMFW100.d Counts vs. Mass-to-Charge (m/z) Figure 10:
Direct infusion, ESI-TOF MS analysis of fresh Senecio madagascariensis collected in the Bega Valley, October 2008. A. young plants; B.
mature, red-stemmed plants; and C. mature, green-stemmed plants
The direct infusion TOF mass spectrometric analysis (Figure 10) of the three 2008
samples showed the presence of ions compatible with the major pyrrolizidine alkaloids reported for S. madagascariensis (Gardner et al., 2006) and that were observed using the ion trap mass spectrometric analysis on plants collected two seasons earlier in 2006. In particular the ion profiles of the juvenile and mature-red- stemmed plants were very similar. The ion profile for the mature, green-stemmed plants was also qualitatively similar to the other samples but it showed a relative increase in the presence of the ions at 352 and 368. This apparent relative increase could be a result of the sampling technique i.e., accidentally choosing plants with slightly different alkaloid profiles, or it could reflect the presence of other alkaloids with the same molecular weights. The first explanation can be addressed by collecting more samples and completing seasonal alkaloid profiles of plant populations. The Fireweed Toxicity Facts and Perspectives Figure 11.
HPLC-TOF mass spectrometric analysis of fresh, green-stemmed mature samples (green trace) and fresh, red-stemmed mature samples (red trace) of
Senecio madagascariensis. The numbers shown are the masses of the ions observed.
second explanation was addressed by examining the high pressure liquid chromatographic (HPLC) separation of the alkaloids (Figure 11) as with the ion trap
mass spectrometry analysis described for the 2006 samples. The HPLC-TOF mass spectrometric analysis showed that the alkaloid with a molecular ion adduct (MH+) of 424 is predominant. In the samples analyzed, the red- stemmed variation of the plant shows an additional peak with an MH+ of 382 that is not observed in the green-stemmed variation. The red-stemmed variation also shows an additional 2 peaks with an MH+ of 368. The green-stemmed variation shows a larger relative amount of the free alkaloid at 336 (derived from the N-oxide at 352). Despite these minor differences, the two variations are very similar in the profile of Fireweed Toxicity Facts and Perspectives Analysis of Wild Honey from the Bega Valley
Three small samples of honey-in-the-comb, taken from a wild hive in the vicinity of previous fireweed flowerings in the Bega Valley, were each extracted with dilute acid. The three separate extracts were then combined to improve the sensitivity of the detection method. The dilute acid extract was processed using solid phase extraction in a slight modification of the method used to process the plants. The resultant sample was analyzed using HPLC-TOF-MS. There was no unambiguous detection of fireweed pyrrolizidine alkaloids in this honey sample. Nonetheless, the very low level presence in the honey of toxic pyrrolizidine alkaloids derived from Senecio spp. was indicated by the observation of a peak with an MH+ of 352 suggestive of senecionine-N-oxide. The lack of convincing evidence for toxic pyrrolizidine alkaloids in this sample should not be taken as an indication that fireweed honey generally does not contain such alkaloids. The actual history of the honey sample presented is unknown and it may well be that fireweed contributed very little to the foraging bees. Fireweed Toxicity Facts and Perspectives Conclusions and Recommendations

The results of this very limited phytochemical survey of plants collected in the Bega
Valley of New South Wales, clearly demonstrate the presence of hepatotoxic dehydropyrrolizidine alkaloids. There is increasing international concern about these alkaloids in animal feed and in the human food supply (European Food safety Authority, 2007). On the basis of potential toxicity to livestock and to humans it would seem prudent to control the spread of fireweed. In the first instance this should involve the rigorous containment of current populations thereby preventing incursions into new areas. Where small, invading populations are discovered, they should be immediately removed. The success of this approach has been reported on Hawaii where, in some areas, regular monitoring and hand-pulling of fireweed plants has resulted in a dramatic decrease in the plants observed (www.hawaiiinvasivespecies.org/iscs/kisc/pdfs/wow3text.pdf). The example set, and the concern shown, by the Hawaiian Invasive Species Council and the Kauai Invasive Species Committee (www.hawaiiinvasivespecies.org/iscs/kisc/wow.html) should be examined and emulated in Australia. Thorne et al (2005) describe an adaptive management approach to the control of fireweed. The essential component is an integrated weed management plan comprising three levels of activity i.e., prevention, control and immediate response. In concert with controlling fireweed where it is established, the fireweed should be prevented from invading new areas and, when it does, immediate action should be taken to eradicate it from the new area. Thorne et al (2005) describe six steps to adaptive management control of fireweed i.e., establish goals; set management priorities; identify appropriate methods; develop and implement an integrated weed management plan; monitor results; and adaptive
modification to improve the plan. Livestock health and productivity should be carefully monitored (taking into consideration the delayed onset of some clinical signs) in fireweed-endemic areas. Animal-derived food products, including meat, milk and bee products such as honey and pollen, should be checked for the presence of hepatotoxic pyrrolizidine alkaloids. Fireweed Toxicity Facts and Perspectives Because of the potential for higher exposures, locally-produced and consumed honey and pollen products should be specifically monitored for hepatotoxic pyrrolizidine alkaloid content. Fireweed Toxicity Facts and Perspectives References

Ahmad K (2001). Health and safety of Afghans hangs in the balance. Lancet 358:
Beales KA, Betteridge K, Colegate SM, and Edgar JA (2004). Solid phase extraction and LCMS analysis of pyrrolizidine alkaloids in honeys. Journal of Agriculture and Food Chemistry 52: 6664-667. Betteridge K, Cao Y, and Colegate SM (2005). Improved method of extraction and LC-MS analysis of pyrrolizidine alkaloids and their N-oxides in honey: Application to Echium vulgare honeys. Journal of Agriculture and Food Chemistry 53: 1894-1902. Boppré M, Colegate SM, and Edgar JA (2005). Pyrrolizidine alkaloids of Echium vulagre honey found in pure pollen. Journal of Agriculture and Food Chemistry 53: Boppré M, Colegate SM, Edgar JA, Fischer OW (2008) Hepatotoxic pyrrolizidine alkaloids in pollen and drying-related implications for commercial processing of bee pollen. J Agric Food Chem 56: 5662-5672. Chou MW, Wang Y-P, Yan J, Yang Y-C, Beger RD, Williams LD, Doerge DR, and Fu PP (2003). Riddelliine-N-oxide is a phytochemical and mammalian metabolite with genotoxic activity that is comparable to the parent pyrrolizidine alkaloid riddelliine. Toxicol. Lett. 145, 239-247. Colegate SM, Edgar JA, Knill AM, and Lee ST (2005). Solid phase extraction and LCMS profiling of pyrrolizidine alkaloids and their N-oxides: A case study of Echium plantagineum. Phytochemical Analysis 16: 108-119. Fireweed Toxicity Facts and Perspectives Colegate SM, Edgar JA, and Stegelmeier BL (1998). Plant-associated Toxins in the Human Food Supply. In Environmental Toxicology: Current Developments, Rose J. (ed.) Gordon and Breach Science Publishers, Amsterdam, pp. 317-344. Edgar JA and Smith LW (2000). Transfer of pyrrolizidine alkaloids into eggs: Food safety implications. In Natural and Selected Synthetic Toxins : Biological Implications, Tu AT, and Gaffield W (eds). American Chemical Society, Washington, DC, pp. 118-128. Edgar JA, Roeder E, and Molyneux RJ (2001). Honey from plants containing pyrrolizidine alkaloids: A potential threat to health. Journal of Agriculture and Food Chemistry 50: 2719-2730. European Food Safety Authority. (2007). Opinion of the scientific panel on contaminants in the food chain on a request from the European Commission related to pyrrolizidine alkaloids as undesirable substances in animal feed. The EFSA J., 447, 1- Food Standards Australia New Zealand. 2001. Pyrrolizidine alkaloids in food: a toxicological review and risk assessment. (http://www.foodstandards.gov.au/_ srcfiles/TR2.pdf ). Gardner DR, Thorne MS, Molyneux RJ, Pfister JA and Seawright AA (2006). Pyrrolizidine alkaloids in Senecio madagascariensis from Australia and Hawaii and assessment of possible livestock poisoning. Biochem. Sys. Ecol. 34: 736-744. German Federal Department of Health. (1992). Bundesanzeiger June 17: 4805; cited by: Dtsch Apoth Ztg 132: 1406-1408.
Huxtable RJ (1989). Human health implications of pyrrolizidine alkaloids and herbs containing them. In Toxicants of Plant Origin, Volume 1 Alkaloids, Cheeke PR (ed). CRC Press, Boca Raton, Fl., pp. 41-86. Fireweed Toxicity Facts and Perspectives James LF, Panter KE, Molyneux RJ, Stegelmeier BL and Wagstaff DJ (1994). Plant toxicants in milk. In Plant-associated Toxins: Agricultural, Phytochemical and Ecological Aspects. Edited by Colegate SM and Dorling PR. CAB International, pp Molyneux RJ, Johnson AE and Stuart LD. (1988). Delayed manifestation of Senecio- induced pyrrolizidine alkaloidosis in cattle: case reports. Vet. Hum.Toxicol. 30: 201- Noble W, Crossley JdeB, Hill BD, Pierce RJ, Mckenzie RA, Debritz M and Morley AA (1994). Pyrrolizidine alkaloidosis of cattle associated with Senecio lautus. AVJ Prakash AS, Pereira TN, Reilly PEB, and Seawright AA (1999). Pyrrolizidine alkaloids in human diet. Mutation Research 443: 53-67. Seawright, AA (1994). Toxic plant residues in meat. In Plant-associated Toxins: Agricultural, Phytochemical and Ecological Aspects. Edited by Colegate SM and Dorling PR. CAB International, pp 77-82. Seaman JT and Walker KH (1985). Pyrrolizidine alkaloid poisoning of cattle and horses in New South Wales. In: Plant Toxicology Edited by Seawright AA, Hegarty MP, James LF and Keeler RF. Queensland Poisonous Plants Committee, Yeerongpilly. pp. 235-246. Small AC, Kelly WR, Seawright AA, Mattocks AR and Jukes R (1993). Pyrrolizidine alkaloidosis in a two month old foal. Zentralbl. Veterinarmed A 40: 213-218. Stegelmeier BL, Edgar JA, Colegate SM, Gardner DR, Schoch TK, Coulombe RA, Molyneux RJ. (1999). Pyrrolizidine alkaloid plants, metabolism and toxicity. J. Nat. Toxins 8, 95 – 116. Fireweed Toxicity Facts and Perspectives Schmidl L (2006). Biology and control of ragwort, Senecio jacobaeaL., in Victoria, Australia. Weed Res. 12: 37-45. Thorne MS, Powley JS and Fukumoto GK (2005). Fireweed control: an adaptive management approach. Pasture and Range Management Oct 2005. Cooperative Extension Service, College of Tropical Agriculture and Human Resources, University of Hawaii at Manoa. Willmot FC and Robertson GW. (1920). Senecio disease, or cirrhosis of the liver due to senecio poisoning. Lancet 2, 848–849.

Source: http://fireweed.org.au/wp-content/blogs.dir/55/files/2014/04/Toxicity_Research_Report_Final-1_01.pdf