random walk plays a fundamental role in biology

  • pantarhei (Heraklit): everything flows

Diffusion in general is the interpenetration between two and more substances without any chemical reaction.
Diffusion causes the flow rate at which matter moves across a unit area. It is sensitive to the temperature and the viscosity of the medium wherein it occurres.
Diffusion of liquid into another, the transfer of heat or electricity from one point in space to another, they all are examples of “fluxes” induced by “forces”. These forces are gradients of concentration, differences in temperature, or electrical potential, respectively.
For details see Cussler 1997, Kleeman 1920 and Reid et al. 1977.
Berthollet (1803) was the first, who pointed out the analogy between heat flow and diffusion, when he discussed the mechanism of the dissolution of a salt crystal in water:
“The crystal dissolves, and the removal of the dissolved solute from its surface may involve pure diffusion, occuring without visible movement of the solution at a whole, and, in addition, a macroscopic flow of the denser parts of the solution relative to the lighter parts. Similarly, heat flow by conduction may be accom-
panied by convection;”  (cited by Tyrrell 1961).

Fourier (1822) found  that in the case of heat conduction the flow of heat is in linear relationship to the gradient of temperature. Five years later it has been shown by Ohm that the electric current flowing in a conductor was linear to the potential difference between the ends of the conductor. A simple linear function dominates both flow of heat and electricity.
It was Fick (1855) who assumed upon Berthollet’s analogy that the force responsible for diffusion is the gradient of concentration, and he formulated the first and second law of diffusion.
This linear relationship is given by the differential equation (in a one-dimensional system)
                                    J =  - (constant) df/dx.

J represents the transport or flow of heat, matter or electricity in direction x perpendicular across a reference plane of unity, df/dx is the corresponding “force” namely gradient of temperature, concentration of dissolved molecules or electrical potential. And so the flow of heat, molecules and electricity can be written as

                            J = - k dT / dx                 (Fourier’s Law)
                            J = - D dC / dx                (Fick’s Law)
                            J = - κ dΨ / dx                (Ohm’s Law)
 
where T, C and ψ are temperature, concentration and electrical potential, respectively. and k, D and κ are the corresponding coefficients. If the conductor is a uniform bar of lenght L with a cross-section of unity across which a constant potential (-ΔΨ) is applied than Ohm’s law looks like

                             I = (- κ/L) (-ΔΨ) = ΔΨ / R,

where R is called the electrical resistance regulating the electrical current I caused by the electrical potential ΔΨ.

This is one of the most frequently used equations on earth.

Coming back to diffusion of molecules or larger particles it should be mentioned that two men have been pioneers in this field: Thomas Graham and Adolf Fick.
Graham (1829, 1833) built the first instruments to study the diffusion in gases and liquids, figures 1a,b.















Fig.1a:
Graham’s diffusion tube for gases. This apparatus was used in the best early study of diffusion. A gas like hydrogen diffuses out through the plug, the tube is lowered to ensure that there will be no pressure difference.















Fig.1b:

Graham’s diffusion apparatus for liquids. The equipment in (a) is the ancestor of free diffusion experiments ; that in (b) is a forerunner of the capillary method.

 He could show that diffusion in gases was several thousand times faster than in liquids and he concluded that “the quantities diffused appear to be closely in propotion to the quantity of salt in the diffusion solution” Graham (1850 p.6). Here he statet that the diffusion rate is proportional to the concentration drop.

Adolf Fick brought the next major advances to modern understanding of diffusion. He was born in 1829 when Graham etsablished his first apparatus for gases. His first intention was to become a mathematician but he has been persuaded by his older brother, a professor of anatomy, to change to medicine.
Major topics in his scientific life did not depend on diffusion at all, but on general studies of physiology. He was particularly interested in the function of muscles and in the visual and thermal functioning of the human body.

It was 1855 when he published his first diffusion paper and introduced it with: “Diffusion of dissolved material is left completely to the influence of the molecular forces basic to the same law ... for the spreading of warmth in a conductor and which is already been applied with such great success to the spreading of electricity” (Fick 1855, p. 65). In analogy to Fourier’s heat equation he developed the laws of diffusion. Fick called the quantity D “the constant depending of the nature of the substances” which we call the diffusion coefficient. The dimension of D is square centimeter divided by time in seconds. The values of D differ by several magnitudes depending on the state of aggregation. Some representative examples of diffusion coefficients:

in gases:                           CO2 in air             D = 0,14             cm2/sec
in liquids:                           NaCl in water        D = 0,14 • 10-5   cm2/sec
                                       glucose in water    D = 0,6   •  10-5   cm2/sec
                                       urea in water        D = 1,4   •  10-5   cm2/sec
in biomolecular environment: H2O in lipid bilayer  D = 3     • 10-10  cm2/sec
in metals:                          carbon in steel      D = 0,14 • 10-12  cm2/sec

(-5, -10 and -12 are exponents)

For a survey on diffusion coefficients see Lax and Synowietz 1964 and Weast 1969.

  • Random walk

Molecules in liquids or gases are not at rest. They are very rapidly and perpe-
tually moving, colliding with each other and changing speed and directions due to these collisions. These movements are due to thermal energy and are also called thermal noise (Berg 1993, Feynman et al. 1970). On the average the kinetic energy of a molecule at absolute temperatur T is kT/2, where k is  Boltzmann’s constant. On the other side, the kinetic energy of a molecule with mass m and

moving with velocity v is given by

which leads to equation

resulting in

In reality the velocity is anything but constant in time but fluctuates very rapidly around this mean value of kT/m. Brackets < > denote the average over time and

the correct expression therefore is

This kinetic energy is deep-rooted into every particle of any solution and is the cause of various phenomena in it.
Random migration and continuous agitation of very small bodies (bacteria, tiny dust particles or colloids) in water for instance is caused by a permanent bom-
bardment and unbalanced impacts of surrounding water molecules where the inequality of the momentary collisions on one side to the other lead to a “jiggling around”.
These movements have been observed soon after the discovery of the light mi-
croscope, especially by its inventor Leeuwenhoek (1632-1723) and some others.

Robert Brown (1773-1858, figure 2) indeed was the first, who discussed this phenomenon in detail (1828, 1829) during his microscopical investigations on pollen grains of some flowering plants. Due to his contributions the irregular motion of small particles in gases and liquids is called the “Brownian (molecular) motion”.



















Fig.2:
Robert Brown in the 3rd         and          6th decennium of the 18th century

This motion never ceases and can be observed even in liquid inclusions in quartz thousands of years old. This experience together with the fact that non-viable mineral objects like dust particles also show these phenomena led Brown – in contradiction to other scientists in his time – to the conclusion that these move-
ments are not due to any type of animation.

The Brownian motion is completely irregular in a haphazard fashion and in all directions. Observations of these movements are mostly performed in the two dimensional plane. Perrin (1923) was one of the first scientists, who made extensive investigations on this phenomenon and fig.3 gives an expample of his data published in Atoms (Die Atome ).
















































Fig.3:
Planar Brownian motion of microscopic particles in two dimensions

In his capital year 1905 Einstein published a mathematical theory for the explana-
tion of the Brownian motion and deduced a way to use this motion for the deter-
mination of Avogadro’s number.
In 1908, the French physicist Jean-Baptiste Perrin verified these ideas experimen-
tally. He did a lot of work on measuring this quantity and they both provided the evidence for the molecular structure of matter, which is well-established today.
When Brown did his observations in 1827 it was not so!

For his work Perrin was honoured with the Nobel Prize in 1926.

Upon Brownian motion a particle moves back and forth but the distance x from the starting point gradually increases. This spreading out in only one dimension

leads to the relationship

where <x^2>  is the mean-square displacement of particles in time t. The quantity D characterizes the migration velocity of the particles of a given kind in a given medium at a given temperature and is our well-known diffusion coefficient. The mean-square of the diffusion path of a particle is proportional to the diffusion time and the diffusion coefficient.

The average displacement of a particle is then
                                                     given by

The time t to migrate along the path x is

For urea in water we had

This molecule will walk through a
bacterium (width x ~ 0,0001 cm) in

or roughly half of a millisecond.

It diffuses a distance of 1 cm (width of a test tube) in t = 36-thousand seconds, lasting 10 hours.
Eq.7 is valid for each direction in the three dimensional space. Diffusive displacements in more than one dimension are connected with the relationship
                                   <x^2> = 4Dt in planar and
                                   <x^2> = 6Dt in cubic dimensions.

  • diffusion coefficient of DAPI

We will try to derive the diffusion coefficient of DAPI by using the coefficient of rhodamine B or rhodamine 6G (furthermore refered to as rhodamine) as one of the reference molecules in fluorescence spectroscopy (Thaer and Sernetz 1973, Rost 1991, Lakowicz 2006). A publication of Rigler et al. 1993 on fluorescence corre-
lation spectroscopy has often been cited by refering to the diffusion coefficient of rhodamine to be
                                          2,8*10^-6 cm^2/sec.

But if one searches in detail through that paper it will become clear that Rigler solely did use this value without any reference. All publications until now were based on data which have been generated nearly a century ago (Herzog and Polotzky 1914).
For details see section “tracing rhodamine”.
The diffusion coefficient increases with temperature and fluidity of the surrounding medium and is proportional to the square root of the mass of the diffusing molecule (Herzog and Polotzky 1914, Wawilow 1925, Wilke and Chang 1955, Chapman and Cowling 1970, Cussler 1997). Assuming the value of 2,8*10^-6 cm^2/sec for rhodamine with a molecular weight of 479 we calculate - as a first correction - a value for DAPI with molecular weight of 368 to be
                                        479^1/2 / 368^1/2 = 1,14  larger.
The second correction is based on the fact that the diffusion is dependent on the shape (ellipticity) of a particle. Elongated particles are moving faster than sphe-
rical ones (Perrin 1934, 1936). The respective formulas are given by a modified copy from Berg (1993, page 57).


For randomly moving molecules with ellipsoidal shape we have to multiply by
                                                  ln(2a/b)
with a as the major and b the minor axies of the ellipsoid. Concerning DAPI the quotient a/b = 3 and ln(2a/b) = 1,8 which finally leads to the coefficient of DAPI of being

                  D(DAPI)  =  2,04 ● D(R6G)  = 5,7 ● 10 ^-6 cm^2/sec .

  • DAPI migrates rapidly through its environment

Using <X^2> = 6Dt for random diffusive motion in space the following graphs in fig. 4 demonstrate the behaviour of DAPI in aqueous solutions to travel along paths in the nano- and micrometer scale.





















Fig.4:
The parabolic relationship between travel time and the respective distance
covered by DAPI in aqueous solutions

Free binding sites in native DNA – maximally 30 µm apart from the next DAPI molecule - can be populated within 0,27 µsec, whereas the penetration of a cell nucleus with a diameter of 5 µm and housing 6 pg DNA is done within 4 sec.

During a typical fcm pulse of 2-5 µsec in length there is time enough for DAPI molecules to associate with binding sites which have been released in the meanwhile.

                    This guarantees the high resolution and sensitivity
                                 of DNA flow cytometric measurements.


To estimate the order of the leaking time of DAPI from the sample core into the surrounding sheath flow we have to apply <X^2> = 4Dt. Assuming 10 µm as the radius of the sample core in a flow cytometer it will take about 43 seconds for the core to be completely evacuated from DAPI molecules. This is long enough not to influence the cytometric accuracy upon leakage of dye out of the sample stream during the time from the formation of hydrodynamic focussing to the point of excitation and fluorescence measurement.

Furthermore, upon application of the multi laser excitation mode (fig.5) with pulse delay of about 10 to 30 µsec between sequential foci no significant degradation of the fluorescence quality is to be expected.




















Fig.5:
Flow cytometry using solitary dual (or multi) laser excitation of cellular particles
generates fluorescence pulses which are separated in time and have to be correlated using sample and hold electronics together with time-of-flight circuits.

In an affort to investigate photo-bleaching and photon saturation upon high power excitation of dyes in flow Ger van den Engh and Colleen Farmer (1992) used a dual laser cytometer with laser lines at identical wavelengths. The first laser was tunable and served to excite the dye at different power levels, whereas the second beam had a constant low power output to monitor the remaining fluorescence after being bleached by the first one.
Their investigations have been focussed on the behaviour of Hoechst 33258 and propidium iodide.
As far as DAPI and HO-33258 have similar binding characteristics to ds-DNA we apply the data thus obtained to our explanations of the binding dynamics between DAPI and DNA.
No declaration was given concerning the time delay between the two laser beams. Therefore, a value is assumed around 20 µsec (fig.5).
Tremendous photo-bleaching could be demonstrated with their experiment. This can only be explained by the fact that the DNA-dye complex - despite the dye being destroyed upon excitation -  was stable during the intersection time of 20 µsec by the two laser beams. Otherwise released binding sites could be repopu-
lated by DAPI within 0,27 µsec and bleached molecules could thus be replaced by unexhausted ones present in the surrounding dye medium.
This leads to the conclusion that the DNA-DAPI complex does dissociate with a rather large constant rate. A dissociation constant of 8,5/sec was observed by Tanious et al. 2008 for DAPI - polyA-polyT complexes and this value corresponds well with our argumentation.
In conclusion, DAPI complexes with cellular DNA are stable during the few micro-
seconds of interrogation by a laser beam for fluorescence excitation in flow. Binding sites temporarily released from a DAPI molecule will be repopulated very fast within less than one third of a microsecond.


                                                    In conclusion:

DAPI is one of the favourites for quantitative DNA flow cytometry due to:

1.
   selectivity of DAPI fluorescence (and bisbenzimidazole derivatives) to
      AT-rich DNA,

2.  binding stability, and

3.  fast diffusive replenishment of released binding sites.