Chapter 3 – Making Light Work in Biology  107

detector with pixel edge length 18.4 μm and a magnification of 300 between the sample

and the camera, was used to monitor the localization of the protein with millisecond

sampling, suggesting the protein was freely diffusing with a mean of two fluorescent

spots detected at the start of each acquisition if the focal plane of the microscope was

set to the midpoint of each cell. Estimate the intracellular concentration in nanomolar

of the protein.

3.7 What do we mean by the Stokes shift, and where does it originate from? Why does the

normalized excitation spectrum of a fluorophore look like a rough mirror image of

the normalized emission spectrum?

3.8 A fluorescence microscopy experiment using an objective lens of numerical aper­

ture 0.9 was performed on a zebrafish embryo (a “model organism” used to investi­

gate multicellular tissues, see Chapter 7) to investigate a single layer of GFP-​labeled

cells of 10 μm diameter, focusing at the midheight level of the cells. These cells were

expected to express a mean number of 900 GFP-​tagged protein molecules per cell

with a standard deviation of 500 molecules per cell. Typical cells had a measured

intensity during fluorescence microscopy of ~107 counts per cell integrated over the

whole of the data acquisition prior to cells being completely photobleached. Data

were acquired from a high-​efficiency camera detector whose magnification per pixel

was equivalent to 200 nm at the sample, gain was 300. Similar cells from another

zebrafish in which there was no GFP had a total measured intensity during the same

fluorescence microscopy imaging of ~2 × 106 counts.

a

Assuming that captured fluorescence emissions come from a cell slice whose

depth is equivalent to the depth of field of the objective lens, estimate the number

of fluorescence emission photons detected from a single molecule of GFP.

b

The quantum efficiency of the camera was ~90%, the transmission function of

the dichroic mirror transmits a mean of 85% of all GFP fluorescence emissions,

and an emission filter between the dichroic mirror and camera transmits 50%

of all GFP fluorescence emissions. The transmission losses due to other optical

components on the emission pathway of the microscope resulted in ~25% of all

light not being transmitted. Estimate the mean number of photons emitted in

total per GFP molecule, stating any assumptions.

3.9 A similar experiment was performed on the cells from Question 3.5, but on a thicker

tissue section 80 μm thick, consisting of a 70 μm layer of similar cells not labeled with

GFP close to the microscope’s coverslip surface, above which are the single layer of

GFP-​labeled cells.

a

Assuming that the rate at which light is absorbed when it propagates through

a homogeneous tissue is proportional to its intensity and to the total number

of molecular absorption event, derive the Beer–​Lambert law, stating any

assumptions.

b

If each layer of cells attenuates, the excitation beam by 4% calculates the total

integrated emission signal due to GFP from a single cell using the same micro­

scope and camera detector and of focusing at mid-​cell height for the GFP-​labeled

cells, acquiring data until the cells are completely photobleached as before.

c

Estimate the total noise per cell detected during these data acquisitions.

d

To detect a given cell reliably above the level of noise, the effective SNR needs

to be above ~2. If the tissue area is ~100 × 100 μm, estimate how many cells you

would expect to detect, stating any assumptions you make.

3.10 Describe the technique of TIRF microscopy and give an example of its application in

biophysics. Most TIRF in vivo studies investigate membrane complexes; why is this?

Would TIRF still be effective if there was a high concentration of auto-​FPs in the cyto­

plasm? Can TIRF be applied to monitoring the nucleus of cells?

3.11 The wave equation for a plane wave of light has solutions of the form

E

E

i kx

ky

t

=

+

[

]

{

}

0exp

sin

cos

θ

θ

ω