migration of MP/APol complexes and of free APol particles. Save the curves in the program Gussi
(http://biophysics.swmed.edu/MBR/software.html) for an easy superposition of the different c(s) curves
(for different optics and samples) and for integration of the peak areas (Le Roy et al. 2015).
The c(s) analyses are generally robust, i.e. their results do not depend on the input parameters
(partial specific volume, solvent density and viscosity, the frictional ratio being fitted).
Comparing the c(s) curves at different concentrations allows to determine whether the samples
can be considered as containing non-interacting species (in which case the peaks are always at
the same s-values, which can be interpreted in terms of molar mass), or if there are association/
dissociation equilibria (in which case s increases when increasing the concentration), or if
non-ideality has to be considered (in which case s decreases at the higher concentrations).
• The signals from the integration of the peak areas are interpreted in terms of bound APol, in g
per g protein in the complexes, with the input of the extinction coefficients and refractive
index increments.
The precision of the results depends on that of the extinction coefficients.
If not taken into account, lipid binding can bias the results, leading to overestimating the mass
of APol bound. Indeed, combining absorbance and interference optics data provides the amount
of bound surfactants (APol + lipids), assuming lipids and APol, whose UV absorbance spectra
are different, to contribute similarly to the interference signal.
• The s-values from the integration of the peak areas are interpreted for non-interacting species
using the Svedberg equation, taking into consideration the amount of bound APol and using
either an estimate of R S (but beware that the latter is overestimated by SEC; see Gohon et al.
2008 and § 9.3.2) or assuming a shape. Typical values for the frictional ratio are 1.25 for a
globular compact shape, 1.5 for a moderately anisotropic shape, 1.8 for a significantly
anisotropic shape; glycosylated proteins are characterized by larger values, such as 1.5–1.8
for a moderately asymmetrical shape (Le Roy et al. 2014).
Alternatively, values of protein molar masses corresponding to different association states
(e.g. monomer, dimer) can be used, along with the protein/APol mass ratio within the complex,
to derive, from the values of the sedimentation coefficients, those of R S and/or of the frictional
ratio. Only the putative association states corresponding to experimental values of R S , or to
reasonable – see above – values of frictional ratio, will be considered as plausible.
Samples do not have to be homogeneous for such an analysis to be performed. If, from the c(s)
analysis, a sample is found to comprise several non-interacting species (at most three), such as APol
particles and two different MP/APol complexes, their properties can be determined separately, the
analysis providing, for each species, the values of its diffusion coefficient D t and sedimentation
coefficient s. D t is directly related to R S . Combining D t and s yields the buoyant molar mass,
M b ¼ s‧RT‧D
À1
t ¼
X
M i
À
1 À ρ
v i
Á
, with M i and
v i the molar mass and partial specific volumes
of the protein, APol, and, if present, lipids, and ρ
the solvent density. Thus, knowing the masses of
bound APol and lipid in g per g protein, the buoyant molar mass provides the mass of the protein (see
Gohon et al. 2008).
Protocol prepared by Christine Ebel on the basis of Gohon et al. (2008) and published reviews
and protocols (Ebel 2011; Le Roy et al. 2014, 2015).
9.5 Protocols
443
(http://biophysics.swmed.edu/MBR/software.html) for an easy superposition of the different c(s) curves
(for different optics and samples) and for integration of the peak areas (Le Roy et al. 2015).
The c(s) analyses are generally robust, i.e. their results do not depend on the input parameters
(partial specific volume, solvent density and viscosity, the frictional ratio being fitted).
Comparing the c(s) curves at different concentrations allows to determine whether the samples
can be considered as containing non-interacting species (in which case the peaks are always at
the same s-values, which can be interpreted in terms of molar mass), or if there are association/
dissociation equilibria (in which case s increases when increasing the concentration), or if
non-ideality has to be considered (in which case s decreases at the higher concentrations).
• The signals from the integration of the peak areas are interpreted in terms of bound APol, in g
per g protein in the complexes, with the input of the extinction coefficients and refractive
index increments.
The precision of the results depends on that of the extinction coefficients.
If not taken into account, lipid binding can bias the results, leading to overestimating the mass
of APol bound. Indeed, combining absorbance and interference optics data provides the amount
of bound surfactants (APol + lipids), assuming lipids and APol, whose UV absorbance spectra
are different, to contribute similarly to the interference signal.
• The s-values from the integration of the peak areas are interpreted for non-interacting species
using the Svedberg equation, taking into consideration the amount of bound APol and using
either an estimate of R S (but beware that the latter is overestimated by SEC; see Gohon et al.
2008 and § 9.3.2) or assuming a shape. Typical values for the frictional ratio are 1.25 for a
globular compact shape, 1.5 for a moderately anisotropic shape, 1.8 for a significantly
anisotropic shape; glycosylated proteins are characterized by larger values, such as 1.5–1.8
for a moderately asymmetrical shape (Le Roy et al. 2014).
Alternatively, values of protein molar masses corresponding to different association states
(e.g. monomer, dimer) can be used, along with the protein/APol mass ratio within the complex,
to derive, from the values of the sedimentation coefficients, those of R S and/or of the frictional
ratio. Only the putative association states corresponding to experimental values of R S , or to
reasonable – see above – values of frictional ratio, will be considered as plausible.
Samples do not have to be homogeneous for such an analysis to be performed. If, from the c(s)
analysis, a sample is found to comprise several non-interacting species (at most three), such as APol
particles and two different MP/APol complexes, their properties can be determined separately, the
analysis providing, for each species, the values of its diffusion coefficient D t and sedimentation
coefficient s. D t is directly related to R S . Combining D t and s yields the buoyant molar mass,
M b ¼ s‧RT‧D
À1
t ¼
X
M i
À
1 À ρ
v i
Á
, with M i and
v i the molar mass and partial specific volumes
of the protein, APol, and, if present, lipids, and ρ
the solvent density. Thus, knowing the masses of
bound APol and lipid in g per g protein, the buoyant molar mass provides the mass of the protein (see
Gohon et al. 2008).
Protocol prepared by Christine Ebel on the basis of Gohon et al. (2008) and published reviews
and protocols (Ebel 2011; Le Roy et al. 2014, 2015).
9.5 Protocols
443