Skip to main content
Back to Earth and Space

POLLUTION · HUMAN HEALTH

Toxicology Dose-Response Lab

AP Environmental Science
File ▾

AP ENVIRONMENTAL SCIENCE · UNIT 8

How much is too much? Ask the brine shrimp.

“The dose makes the poison.” Toxicologists find out how much of a substance it takes to harm living things by exposing groups of small organisms to a range of concentrations and counting how many die. You will run that test on brine shrimp larvae (nauplii) and read the result from a dose-response curve.

A brine shrimp nauplius under the microscope: an orange, pear-shaped larva with one dark eye and two long, bristled swimming antennae.
  1. Dilute a stock solution step by step, halving the concentration each time.
  2. Expose 10 nauplii in each well of a 24-well plate, four wells per concentration, and let the clock run.
  3. Count the dead under the microscope, record the counts and read the curve.

Choose where to start:

What am I looking at?

The solutions are drawn tinted so you can see the dilution; many real toxicants are colorless.

About this model

The test. A brine shrimp lethality test: newly hatched Artemia nauplii are put in seawater with a range of concentrations of a substance, 10 per well with replicate wells, and the dead (no movement after several seconds of watching) are counted after a fixed time. It is a quick, cheap screen for acute toxicity.

The curve. The chance that a nauplius dies follows a log-logistic (Hill) curve: fraction dead = Cn ÷ (LC50n + Cn), where C is the concentration, LC50 the concentration that kills half the group, and n how steep the curve is. Each nauplius has its own tolerance, so the counts vary from well to well just as real counts do, and a nauplius that has died stays dead.

The substances are fictional. Substance A (steep curve), B (steep, about half as toxic as A) and C (shallow curve) have made-up LC50s and slopes chosen for teaching. They are not real toxicity data and must not be used for any real exposure. Their true LC50s are hidden: you find them from your own counts.

Time. In this model, the LC50 falls as exposure gets longer: the 24-hour LC50 is 1.6 times the 48-hour value, and the 96-hour LC50 is the 48-hour value ÷ 1.6. Real substances differ in how strongly their toxicity depends on time.

Reading the LC50. The line on your graph is fitted to your own counts (logit regression on the log of concentration), so your reading is checked against your data, not against the hidden value.

What is left out. No deaths in the controls (in real tests some die from handling; a test is usually accepted only if at least 90 % of the controls survive), no feeding, hatching age, temperature or light effects, and no sublethal effects such as slower swimming or growth. A 48-hour LC50 is an acute result: it cannot show what long exposure does. LC50 (a concentration in water or air, with a time) is not LD50 (a dose per kilogram of body mass).