Be careful about misapplication of residue analysis results.
ng of pesticide in a sample of bees, and LD50 per bee are NOT the same
thing, can not be directly compared.
Any analysis lab will give you a result expressed as conc of
pesticide/weight of bee. A common sample size is 5 gm of bees. So, if the instrument
reads out 10 ng in a 5 gm sample, that's going to be reported as 2 ng/gm of
bee.
A live bee weighs approximately 0.1 gm WET Weight, but only about 28 mg dry
weight. So, you need to know the condition (moisture content) of the bee
sample in order to convert 2 ng to a per bee amount (wet/dry explained
below).
LD50s are calculated based on dose delivered per bee, not per weight of bee
tissue.
A caveat here: The following is an over-simplified example, and there are
exceptions, such as a pesticide that is repellant.
LD50 assays are often contact or ingestion. For a typical contact
toxicity assay, that means how much chemical was in the drop applied to the bee.
And, in that case, the amount per bee should be consistent.
Many LD50s, however, are based on dose from food (syrup, pollen,
vegetation), which is often calculated from the amount of chemical in the source
material (food), the amount of food consumed, and the number of bees in the
trial (e.g., bees in a cup).
In the best situation, the investigator uses the amount of chemical
measured in the food source, the amount of food consumed, and then divides by
number of bees in the treatment (bees in cup/cage). In a non-GLP assay, the
investigator may simply calculate the amount of chemical in the food based
on the amount of chemical mixed in to the food item, or applied to
vegetation. That only works IF the investigator got the mix right. Misread your
pipette, miscalculate, get stratification in/on the food source, and this
number will be wrong. GLP (Good Laboratory Practice) will always require
confirmatory chemical analysis of the spiked food.
But note, even in the best case, the LD50 is an average. A cup of say 30
bees is not going to consume the same (equal) amount per bee. So, in
essence, the test is often skewed by those bees that consumed the most food.
They get a higher dose than the calculated dose.
But, as Randy notes, pesticides in a bee are not necessarily stable. Some
persist, some break down rapidly. And that where the ng of pesticides in a
sample of bees gets to be an issue.
Now, ng per bee sample, is NOT ng/bee. It is the measured amount of
chemical remaining in a sample of bees, but the number of bees in the sample is
often unknown
Even if you have instrumentation that can measure pesticide residues in a
single bee, you don't want to do that, because one bee may not be
representative of the thousands of bees in a colony. As a general rule, you need
25-30 bees or more to get a REPRESENTATIVE sample.
So, labs like the USDA facility at Gastonia, prefer samples of 5 gms or
more bees for two reasons:
1) it provides a better, more representative, value for the colony, and
2) it provides sufficient material for them to analyze without having to
push their instrument detection limits.
So, back to our 2 ng example. If the resultant value is 2 ng of
pesticide/1 gm of bee, the dry weight of a single bee is about 28 mg (1/1,000 gm).
That means that the ng amount in that bee is very low. A 5 gm bee sample,
on the other hand, is likely to consist of 50 or more bees. So, its easier
to detect 2 ng pesticide in 50 bees, since there should be 2 ng/gm of bee,
or at least 10 ng of pesticide in the 50 bees.
Unfortunately, many labs do not adjust for moisture content of the bee
sample. Since bees are 60-70% water, a sample of just dying, quickly frozen
bees will be three times heavier per bee, than a sample of dried bees (in an
oven or picked up from the ground in front of the hive after baking in the
sun). So, one needs to know the condition of the samples, the moisture
content, or the number of bees in the weight of bees analyzed, before the
amount of the suspect chemical expressed as an amount per gm of bee can be
determined.
Now, if you have ALL of this information at hand, you can then calculate
the per bee dose from the residue analysis. When I was first working with
bee kills, I ended up in a court case on the opposite of the issue from Larry
Atkins, who testified for the company (in this case, it was an industrial
pollution poisoning incident) that the residue analysis that the beekeepers
had from their dead bees could not be compared to the LD50 values, that he
himself had determined for the poison, in this case arsenic. For those
of you too young to remember, Larry is the source of much of the information
about LD50s for many of the pesticides used in the USA - up to the time
that he retired.
That statement by Larry, in court, launched the thesis of my first
graduate student, who worked out and proved that one could compare the two, but
only if the residue analysis was thorough and well documented - we routinely
either take bee samples to dry weight, or if the chemical is not stable
during drying, count the number of bees per sample weight, so that we can
calculate amount per bee.
That problem persists to this day. Whenever I send a bee sample to Roger
Simmonds at the USDA lab, I ask him to provide the number of bees in the
sample analyzed, since the USDA lab grinds the samples that they receive with
dry ice, use a set weight of bee sample (often 5 gm) - but makes no
allowance for moisture content of the bees as received. It is cost prohibitive
for them to 'dry' the bee samples.
Not to belabor the point, but a ballpark estimate of the number of bees in
a one gm sample of live bees, would be 10 bees. But, if the sample is of
dried/out, sun-baked bees, then the number of bees in a one gm sample of
bees would be closer to 36. So, 2 ng/gm bee would be 0.2 ng to 0.05 ng per
bee.
So, the next best thing is to count then number of bees in the sample
analyzed. If they have a known number of bees per weight of bee sample
analyzed (e.g, 47 bees/5 gm), then I can calculate the ng amount per bee, which
is what you need if you want to compare to LD50, expressed as dose/bee.
Jerry
***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html
Guidelines for posting to BEE-L can be found at:
http://honeybeeworld.com/bee-l/guidelines.htm
|