THE JOURNAL OF BIOLOQICAL CHEMISTRY
`Vol.245, No.19, Issue of October 10, pp. 5161-51135, 1970
`Printed
`in U.S.A.
`
`of Protein-Sodium
`The Gross Conformation
`Dodecyl
`Sulfate
`Complexes*
`
`TANFORD~
`AND CHARLES
`A. REYNOLDS~
`JACQUELINE
`From
`the Departments of Biochemistry and Anatomy, Duke University Medical Center, Dwham, North Carolina
`27706
`
`(Received
`
`for publication,
`
`June 8, 1970)
`
`SUMMARY
`a wide
`with
`sulfate
`of sodium
`dodecyl
`interaction
`The
`ratio
`binding
`is characterized
`by a high
`of proteins
`variety
`5 x
`exceeds
`concentration
`of amphiphile
`the monomer
`when
`ratio on a gram
`to gram basis
`is identi-
`This binding
`10W4 M.
`The protein
`portion
`of the
`cal
`for all proteins
`investigated.
`complex
`contains
`a high degree
`of order,
`and hydrodynamic
`studies
`suggest
`that
`the
`complex
`is a rodlike
`particle,
`length
`of which
`varies uniquely
`with
`the molecular
`weight
`the
`protein
`moiety.
`These
`results
`explain
`the
`empirical
`observation
`that proteins
`dissolved
`in aqueous
`solutions
`con-
`taining
`high
`concentrations
`of sodium
`dodecyl
`sulfate
`have
`electrophoretic
`mobilities
`on polyacrylamide
`gels which
`are
`a unique
`function
`of their molecular
`weights.
`In addition,
`the data
`suggest
`a possible
`model
`for
`the
`conformation
`membrane
`proteins
`and
`their
`interactions
`with phospholipid.
`
`the
`of
`
`of
`
`hydro-
`containing
`of proteins with amphiphiles
`interaction
`The
`to a con-
`tails with 12 or more carbon atoms usually
`leads
`carbon
`Specifically,
`the
`formational
`change
`in
`the protein moiety
`(1).
`amphiphile,
`sodium
`dodecyl
`sulfate,
`induces alterations
`in pro-
`tein structure
`at monomer
`concentrations
`of 1 X 10m4 M. Re-
`cently, we have demonstrated
`that
`a wide variety
`of reduced
`proteins
`bind
`identical
`amounts
`of SDS on a gram
`to gram basis
`when
`the equilibrium
`monomer
`concentration
`of amphiphile
`is
`greater
`than 5 X
`lop4 M (2).
`A saturated
`complex with a stoi-
`chiometry
`of 0.4 g of SDS per g of protein
`is formed
`between 5
`and 8 X 10m4 M SDS monomer,
`and a second complex which
`is
`saturated
`at 1.4 g of SDS per g of protein
`is observed above 8 X
`This nonspecificity
`of SDS
`toward pro-
`low4 M SDS monomer.
`teins derived
`from a multitude
`of sources and having
`very differ-
`ent
`“native”
`states
`suggests a binding-induced
`conformational
`change
`in
`the polypeptide
`chain which
`leads
`to a uniform
`struc-
`
`Science Advancement
`
`from
`
`Award
`
`by Research Grant GB-14844
`* This work was supported
`the National
`Science Foundation.
`$ Partial
`support
`from Health
`1 SO4 FR 06148.01.
`$ Research Career Awardee, National
`United
`States Public Health
`Service.
`1 The abbreviations
`used are: SDS,
`GuHCl,
`guanidine
`hydrochloride;
`ORD,
`sion.
`
`Institutes
`
`of Health,
`
`sodium
`optical
`
`dodecyl
`rotatory
`
`sulfate;
`disper-
`
`the hydrodynamic
`state. We have studied
`in the complexed
`ture
`by means of in-
`shape and size of these protein-SDS
`complexes
`states of
`the
`trinsic
`viscosity
`and compared
`the conformational
`dispersion.
`protein
`in
`its altered
`form using optical
`rotatory
`Since all amphiphilic
`compounds
`have properties
`in common
`which are the result of their
`containing
`both a polar head group
`and a hydrophobic
`tail on the same molecule,
`it may be possible
`to extend
`the results of these studies
`to other amphiphile-protein
`systems.
`In particular,
`the SDS-protein
`complex may be a rea-
`sonable model
`system
`for biological membranes
`in which
`the am-
`phiphilic
`species
`is primarily
`phospholipid.
`The
`interaction
`of
`lipids with some membrane
`proteins
`has been shown
`to be pri-
`marily
`hydrophobic
`(3,4)
`just as the SDS-protein
`complex
`is the
`result of primarily
`hydrophobic
`interactions
`(2).
`However,
`the
`conformational
`state of
`the membrane
`“structural”
`protein
`is
`still a highly
`controversial
`subject,
`and
`the exact
`location
`of the
`protein within
`the biological
`membrane
`is not known
`(5).
`The
`studies
`reported
`here may be used as a basis for tentative
`sugges-
`tions on this subject.
`
`EXPERIMENTAL
`
`PROCEDURE
`
`in
`sources are shown
`their
`used and
`proteins
`:Mater&---The
`from Baker
`analyt-
`Table
`I.
`Phosphate
`buffers were prepared
`Sodium
`dodecyl
`ical grade NaZHP04.7H20
`and NaH2P04.H20.
`sulfate was a highly pure grade obtained
`from Mann.
`Guanidine
`hydrochloride
`from Heico, Delaware Water Gap, Pennsylvania
`was used without
`further
`purification.
`Preparation of Protein-SDS &‘omplexes-Protein
`was dissolved
`in 6 M GuHCl
`and 0.1% P-mercaptoethanol
`to obtain polypeptide
`chains
`in
`the
`random
`coil conformation
`(9).
`The GuHCl
`was
`then removed
`by dialysis against Hz0 containing
`a reducing
`agent.
`SDS, P-mercaptoethanol,
`and
`the appropriate
`buffer were dia-
`lyzed
`into
`the bag containing
`the protein
`solution.
`The amount
`of bound SDS and
`the.concentration
`of protein were determined
`as described
`previously
`(2).
`It has been demonstrated
`that
`the
`same
`final binding
`ratio
`is reached when
`the complex
`is obtained
`by
`the above procedure
`as when
`it
`is formed
`by treating
`the na-
`tive protein
`directly with SDS and P-mercaptoethanol.
`Protein-
`SDS complexes were formed by both procedures
`and compared
`in
`the experiments
`reported
`here.
`in Cannon-
`were determined
`MethodsIntrinsic
`viscosities
`immersed
`in a water bath
`ther-
`Manning
`semimicro
`viscometers
`flow
`times
`ranged
`from 250
`to
`mostated
`to
`rtO.005”.
`Solvent
`300 sec. Optical
`rotatory
`dispersion measurements
`were carried
`
`5161
`
`GNE 2003
`Page 1
`
`

`
`5162
`
`Cmform,ation
`
`of Protein-SDS
`
`Complexes
`
`Vol. 245, No.
`
`19
`
`Proteins
`
`I
`TABLE
`used
`in XDX complexes
`
`Protein
`
`c
`
`Cytochrome
`Lysozyme
`Hemoglobin
`
`&Lactoglobulin
`
`Fl-histone,
`
`FZal-histone
`
`Chymotrypsinogen
`Lactate
`dehydrogenase
`&albumin
`Alkaline
`
`phosphatase
`
`Rabbit
`
`y-G, heavy
`
`chain
`
`Bovine
`Myosin,
`
`albumin
`serum
`rabbit muscle
`
`Erythrocyte
`
`ghost proteins
`
`Sigma
`Pentex
`George-
`J. Steinhardt,
`Gift
`of Dr.
`town
`University,
`Washington,
`I). c.
`Gift of Dr. R. Townend,
`University,
`Waltham,
`setts
`Gift of Dr. K. McCarty,
`versity,
`Durham,
`North
`Sigma
`Worthington
`Sigma
`Wash-
`Gift of Dr. M. Schlesinger,
`ington University,
`St. Louis, Mis-
`souri
`by method
`Prepared
`(6)
`Sober
`Nut#ritional
`Prepared
`(7)
`Prepared
`Mitchell,
`
`Brandeis
`Massachu-
`
`Duke Uni-
`Carolina
`
`of Levy
`
`and
`
`Biochemicals
`by
`the method
`
`of Perry
`
`the method
`by
`and Hanahan
`
`of Dodge,
`(8)
`
`4.0
`
`0
`
`1.0
`
`3.0
`2.0
`c x IO3 (g/cc)
`Concen-
`complexes.
`of protein-SDS
`viscosity
`1. Reduced
`0,1.4
`g of SDS per g of
`equals dry weight
`of the protein.
`buffer,
`pH 7.2, SDS
`ionic
`strength
`= 0.026, phosphate
`10-a M.
`l
`,0.4 g of SDS per g of protein,
`ionic
`= 3.5 X
`= 0.52, phosphate
`buffer,
`pH 7.2, SDS monomer
`= 5 X
`BSA,
`bovine
`serum
`albumin.
`
`FIG.
`tration
`protein,
`monomer
`strength
`10-d M.
`
`60 recording
`out on a Cary model
`brated
`cells of O.l- and
`l&mm
`
`spectropolarimeter
`path
`lengths.
`
`using cali-
`
`2.4
`
`t
`
`1
`
`RESULTS
`in-
`Complexes-The
`of Protein-SDS
`Properties
`Hydrodynamic
`teraction
`of SDS with 17 different
`reduced proteins
`of molecular
`The
`weight 12,400
`to 200,000 has been reported
`elsewhere
`(2).
`degree of binding
`is dependent
`only on
`the SDS monomer
`con-
`centration
`and
`is independent
`of the number
`or size of the mi-
`10e4 M,
`celles present.
`At monomer
`concentrations
`above 5 X
`two saturation
`levels of binding
`are observed, one at 0.4 g of SDS
`per g of protein,
`and a second at 1.4 g of SDS per g of protein.
`The
`intrinsic
`viscosities,
`[q], of seven of these protein-SDS
`com-
`plexes have been measured
`at both
`saturation
`levels of binding.
`Fig. 1 shows representative
`data
`for
`three proteins
`from which
`[v] is obtained
`by extrapolation
`of the reduced
`viscosity,
`~,~/c,
`to
`c = 0. We have no
`immediate
`explanation
`for
`the negative
`slopes
`that were
`found with most of the complexes
`except
`the
`possibility
`of aggregation
`to a more
`symmetrical
`particle
`at
`in-
`creased concentrations.
`This hypothesis
`is currently
`being
`in-
`vestigated
`by means of low angle x-ray scattering,
`the results of
`which will be published
`elsewhere.
`is
`viscosity
`intrinsic
`the
`For perfect,
`unsolvated
`spheres
`However,
`pendent
`of the molecular
`weight or chain
`length.
`more extended
`homopolymers
`(10)
`
`inde-
`for
`
`[q] = KM2
`
`= K’nz
`
`(1)
`weight, and
`is the molecular
`where K, K’, and x are constants, M
`x, in Equa-
`The exponent,
`n is the number
`of residues per chain.
`Values of x
`tion 1 varies between 0.5 and 0.8 for random
`coils.
`above 1.0 indicate
`rodlike
`particles
`of roughly
`constant
`diameter
`The sim-
`with
`the
`length
`proportional
`to the molecular
`weight.
`plest model
`is a prolate
`ellipsoid,
`with constant minor
`axis, b, and
`
`01
`
`I
`2.0
`
`I
`
`I
`2.5
`
`I
`log n
`
`I
`3.0
`
`I
`
`I
`3.5
`
`log n for protein-SDS
`against
`[q] plotted
`2. Log
`FIG.
`n = residues
`per polypeptide
`chain.
`0 and
`l asinFig.
`
`complexes.
`1.
`
`x is a slowly
`this model,
`For
`to $1.
`axis, a, proportional
`major
`increasing
`function
`of the ratio a: b, from 1.4 for short,
`thick ellip-
`soids to about 1.8 for long
`thin ellipsoids.
`Similar
`values of x are
`observed
`for other models used to represent
`rodlike
`particles
`(11).
`For
`The value of x decreases
`for rods
`that are not entirely
`rigid.
`example,
`DNA
`of relatively
`low molecular
`weight
`(up
`to 2
`x
`106) has an x value of 1.32
`(12), although
`the
`ratio a: b is very
`large.
`is
`complexes
`of protein-SDS
`viscosity
`intrinsic
`log of the
`The
`for
`The slope
`function
`of log n (Fig. 2).
`linear
`an approximately
`both
`sizes of complex
`is 1.2, and
`it can be concluded,
`therefore,
`that
`all protein
`polypeptide
`chains
`assume
`a similar,
`rodlike
`The
`function
`re-
`shape when complexed
`with
`this amphiphile.
`lating hydrodynamic
`properties
`to chain
`length must be the same
`for all
`these complexes.
`Studies of the protein-SDS
`complex
`by
`electron microscopy
`are underway,
`and preliminary
`results
`con.
`firm
`the rodlike model.
`
`GNE 2003
`Page 2
`
`

`
`Issue of October
`
`10, 1970
`
`J. A. Reynolds and C. Tanjord
`
`5163
`
`idea of the possible dimensions,
`some
`To obtain
`in terms of a prolate
`ellipsoidal
`preted
`the results
`axis, b.
`assuming
`a constant
`value
`for the minor
`The
`intrinsic
`viscosity
`is related
`to the hydrodynamic
`of a particle
`through
`the
`following
`equation
`(10) :
`
`inter-
`we have
`model, without
`
`volume
`
`Protein
`
`Hydrodynamic
`
`properties
`
`II
`TABLE
`complexes
`of protein-SDS
`g of SDS per g of protein”
`L
`
`at binding
`
`level
`
`[VI = V@Z + &VI0 + 6&O)
`
`(2)
`of
`specific volume
`factor, & = partial
`shape
`where v is the Simha
`the protein,
`61 = g of Hz0 per g of protein,
`VJ~O = specific volume
`of HnO, 82 = g of SDS per g of protein,
`and uzo = specific volume
`of SDS.
`The
`following
`values
`of
`these parameters
`were used
`together with
`the experimentally
`determined
`intrinsic
`viscosity
`to calculate
`v.
`
`82 = 0.725
`& = 0.9 at 1.4 g of SDS per g of protein
`per g of protein2
`
`and 0.2 at 0.4 g of SDS
`
`or 0.4 g of SDS per g of protein
`
`c.. .
`
`Cytochrome
`Lysozyme.
`Hemoglobin.
`P-Lactoglobulin..
`Myosin
`light.
`Fl-histone..
`Chymotrypsinogen..
`Lactate
`dehydrogen-
`ase................
`Ovalbumin.
`phospha-
`Alkaline
`tase...............
`7-G heavy.
`Bovine
`serum
`min.
`
`albu-
`
`PQ&T
`
`chain
`moleculal
`weight
`
`11,700
`14,400
`15,950
`18,400
`20,000
`23,900
`25,700
`
`35,000
`43,000
`
`43,000
`49,500
`
`[?I
`
`Shape
`factor,
`Y
`
`Ellipsoid
`axes
`
`b/
`
`I
`A
`
`CC/P”
`(9.7)
`9.0
`8.5
`(10.7)
`(14.0)
`(14.5)
`15.8
`
`(28.4)
`33.5
`
`33.5
`(37.5)
`
`3.38
`3.12
`2.95
`3.73
`4.88
`5.05
`5.54
`
`9.90
`11.7
`
`11.7
`13.1
`
`17.2
`18.9
`20.6
`17.8
`17.6
`18.4
`18.2
`
`17.2
`17.6
`
`17.6
`17.9
`
`Stokes
`radius.
`
`”
`
`a
`
`A
`26.2
`44.8
`27.3
`45.4
`27.7
`42.2
`31.5
`65.8
`35.4
`73.7
`38.0
`80.8
`87.4 40.1
`
`134.3
`156.9
`
`156.9
`174.7
`
`54.0
`61.3
`
`61.3
`66.5
`
`69,000
`
`54.2
`
`18.9
`
`II,0 = 1
`82 = 1.4 g of SDS per g of protein
`v2O = 0.886
`(13)
`the axial
`to obtain
`it is now possible
`ellipsoid,
`Assuming
`a prolate
`v and a : b. The
`ratio using
`the Simha
`relationship
`(10) between
`absolute
`values
`for these
`radii
`can
`then be determined
`from
`the
`known particle
`volume,
`i.e.
`
`M/N@2
`
`+ C?IVI~ + 82~2”) = $rab2 =
`
`(&ra/b)b3
`
`(3)
`
`number.
`is Avogadro’s
`weight and N
`is the molecular
`where J4
`values of v, a, and b are given
`II and
`in Tables
`The calculated
`for
`the complexes
`containing
`0.4 and 1.4 g of SDS per g of
`III
`axis, b, is seen
`to be approximately
`con-
`protein.
`The minor
`stant, about
`14 A and 18 A, respectively,
`for
`the
`two classes of
`complex.
`When
`the major
`axis, a, is plotted
`against molecular
`weight, an essentially
`linear
`relation
`is obtained.
`The
`rod
`length
`(2~)
`is about 0.61 A per amino
`acid residue
`for the 0.4 g of SDS
`per g of protein
`complex,
`and about 0.74 A for
`the 1.4 g of SDS
`per g of protein
`complex.
`
`in a
`chosen
`is frequently
`proteins
`of globular
`2 The hydration
`However,
`fashion
`as 0.2 g of Hz0 per g of protein.
`rather arbitrary
`of SDS micelles
`at ionic strengths
`0.03 to 0.50 has
`the hydration
`directly
`by
`intrinsic
`viscosity
`(J. A. R.eynolds,
`been measured
`unpublished
`results).
`
`Ionic
`
`strength
`
`IIs0
`
`per
`
`SDS
`
`0.03
`0.10
`0.30
`0.50
`
`g/g
`0.9
`0.5
`0.4
`0
`
`18.4 230.5
`
`83.5
`
`a Based on an assumed hydration
`of 0.9 g of Hz0 per g of pro-
`would
`alter b (it would
`be
`tein.
`Changes
`in assumed
`hydration
`about 4 A less
`for zero hydration,
`for example)
`but have no signifi-
`cant effect
`on a.
`were
`6 Values
`in parentheses
`chromatography
`data
`,described
`viscosity
`values are
`in
`terms
`protein.
`
`gel
`from
`indirectly
`determined
`All
`(14).
`in the
`following
`paper
`of cubic
`centimeter
`per g of dry
`
`III
`TABLIG
`complexes
`of protein-SDS
`properties
`of 0.4 g of SDS per g of protein”
`
`at binding
`
`level
`
`Hydrodynamic
`
`Protein
`
`Lysozyme.
`Chymotrypsinogen..
`Ovalbumin.
`Bovine
`serum
`min.
`.
`Myosin..
`
`albu-
`
`14,400
`25,700
`43,000
`
`69,000
`205,0006
`
`4.2
`7.2
`13.7
`
`30.0
`94.8
`
`Shape
`factor,
`Y
`
`3.24
`5.61
`IO.6
`
`23.4
`74.0
`
`Ellipsoid
`
`axes
`
`Stokes
`
`A
`
`35.7
`14.3
`66.7
`13.9
`13.8 114.0
`
`A
`21.2
`30.8
`45.2
`
`13.4 194.3
`15.1 453.0
`
`69.0
`145.5
`
`of 0.2 g of Hz0 per g of protein.
`hydration
`a Based on an assumed
`Changes
`in assumed
`hydration
`would
`alter b but have no signifi-
`cant effect on a.
`assuming
`weight,
`B Viscosity
`average molecular
`220,000 each and two
`chains
`= 20,000 each.
`
`two
`
`chains
`
`=
`
`in Fig. 2
`complexes
`of protein-SDS
`viscosities
`intrinsic
`the
`Since
`at two different
`ionic strengths
`(0.026 and 0.52))
`were determined
`to assume different
`values
`of 61 for
`the
`two sizes of
`it is reasonable
`complex.
`Since
`the hydration
`of the complex
`is not necessarily
`the same as that of either an SDS micelle at a given
`ionic strength
`or a pure protein,
`compromise
`values of 61 were used such
`that 6r is
`somewhat
`larger at an ionic strength
`of 0.026 (1.4 g of SDS per g of
`protein)
`than at 0.52 (0.4 g of SDS per g of protein).
`These values
`primarily
`affect
`the minor axis, b. The length
`of the particle
`(2~)
`is affected
`to a negligible
`extent
`(e.g. for ovalbumin,
`b = 17.6 and
`a = 156.9 when 61 = 0.9 and 82 = 1.4; b = 19.2 and a = 153.4 when
`& = 1.34 and 82 = 1.4).
`
`viscosity
`intrinsic
`The
`treated with an “equivalent
`
`of nonspherical
`sphere” model.
`
`particles
`
`can also be
`
`[q] = 2.5 N/M(&-Rs3)
`the same hydrodynamic
`is the radius of a sphere with
`where R,
`as
`the actual molecule
`under
`investigation.3
`(The
`properties
`3 The value of R, applicable
`to viscosity
`measurements
`would
`the same as the value applicable
`to frictional
`coeffi-
`not be exactly
`cients.
`An equivalent
`sphere model
`is not adequate
`for
`the de-
`tailed
`treatment
`of hydrodynamic
`properties.
`
`(4)
`
`GNE 2003
`Page 3
`
`

`
`5164
`
`Cmformation
`
`of Protein-SDS
`
`Complexes
`
`Vol.
`
`245, No.
`
`19
`
`1.4 g of SDS per g of
`complexes;
`FIG. 3. ORD of protein-SDS
`buffer, pH 7.2. Spe-
`protein;
`ionic strength
`= 0.026; phosphate
`index.
`1, erythrocyte
`cific rotation
`is uncorrected
`for refractive
`4, F2al-histone;
`6, p-
`ghost proteins;
`2, myosin;
`9, Fl-histone;
`Iactoglobulin;
`6, lysozyme;
`7, chymotrypsinogen.
`
`-3000
`
`-4000
`
`-5000
`
`-6000
`
`for several protein-SDS
`to obtain R, values
`technique
`mental
`complexes
`for which viscosity
`data were not directly
`determined.
`R,
`from gel chromatography
`has been converted
`to
`intrinsic
`vis-
`cosities using Equation
`4, and v, a, and b calculated
`as before.
`These
`results are also given
`in Table
`II and are seen to be entirely
`consistent
`with
`the results obtained
`by direct
`intrinsic
`viscosity
`measurements.
`(Experimental
`and
`theoretical
`details of the gel
`chromatography
`results
`are given
`in
`the
`following
`paper
`(14))
`It
`is of interest
`to compare
`the rod
`lengths
`obtained
`for SDS-
`protein
`complexes with
`the
`lengths
`of native
`tropomyosin
`and
`paramyosin.
`These proteins
`consist of two o( helices,
`side by
`side, and
`twisted
`slightly
`about one another.
`Molecular
`weights
`and
`intrinsic
`viscosities
`have been measured
`by Lowey, Kucera,
`and Holtzer
`(15), Holtzer,
`Clark, and Lowey
`(16), and Olander,
`Emerson,
`and Holtzer
`(17).
`If we treat
`their data
`in terms of a
`prolate
`ellipsoid model,
`assuming
`hydration
`of 0.2 g per g of pro-
`tein
`(although
`again hydration
`has little
`effect on the result) we
`obtain
`b = 11.4 A and a = 219 A for paramyosin
`and b = 11.3 A
`and a = 634 A
`for
`tropomyosin.
`These dimensions
`are very
`close to
`those calculated
`in
`the original
`papers by quite different
`The
`lengths
`correspond
`to 0.66 A per residue and
`procedures.4
`0.72 A per residue
`for tropomyosin
`and paramyosin,
`respectively.
`This
`comparison
`shows
`that a model
`consisting
`of a helical poly-
`peptide
`chain,
`folded back upon
`itself near
`its middle
`to give a
`double
`helical
`rod, with
`the SDS
`forming
`a shell about
`the
`rod,
`Of course,
`would be consistent
`with
`the hydrodynamic
`data.
`there
`is no rea.son to believe
`that
`this
`is the actual structure
`of the
`complexes.
`In
`fact,
`the ORD
`data cited below suggest
`that
`the
`helix content
`of the complexes
`is less than 100%.
`Regardless
`of
`what
`the actual
`structure may be,
`the over-all
`particle
`length
`is
`established
`as being about
`half
`the
`length
`of a fully extended
`01
`helix.
`spectra of pro-
`3 shows ORD
`Optical Rotatory Dispersion-Fig.
`tein-SDS
`complexes
`in
`the wave length
`range 2250
`to 2600 A at
`binding
`levels of 1.4 g of SDS per g of protein.
`The protein
`is
`clearly not
`in a random
`coil conformation,
`since
`in that
`case the
`spectra would be a smooth
`curve of much
`lower magnitude,
`such
`as is observed when proteins
`are dissolved
`in 6 M GuHCl
`(9).
`The qualitative
`similarity
`between all
`the complexes
`is striking,
`especially when one keeps
`in mind
`that
`the “native”
`ORD
`spec-
`tra of these proteins
`are quite different,
`both
`from each other and
`from
`that of the corresponding
`SDS complex.
`The effect of the amount
`of bound SDS on the ORD of myosin
`is shown
`in Fig. 4.
`The over-all
`shape of the spectra
`is the same,
`but
`the magnitude
`of levorotation
`decreases at the higher
`level of
`SDS binding.
`This
`same phenomenon
`has been observed
`for
`lysozyme,
`P-lactoglobulin,
`and chymotrypsinogen
`in SDS.
`All of the spectra have
`troughs near 2330 A and,
`in general,
`semble spectra
`for polypeptide
`chains
`in the helical
`conformation.
`However,
`the magnitude
`of rotation
`is considerably
`less than
`the
`reported
`values
`for polypeptides
`or proteins
`that have been exam-
`ined under
`conditions
`where
`they are known
`to be 100% helical.
`This could be accounted
`for in part by the
`fact
`that
`the helices
`in
`protein-SDS
`complexes might
`exist within
`a hydrophobic
`shell
`formed by the SDS; a medium
`of low dielectric
`constant
`is known
`to depress
`the magnitude
`of rotation
`somewhat
`(18), but
`the dif-
`ference
`in magnitude
`between different
`proteins
`could not be ac-
`counted
`for in this way.
`
`re-
`
`treated
`4 Both proteins were
`lengths
`The corresponding
`rod
`respectively,
`for
`tropomyosin
`
`rods of 10 A radius.
`as cylindrical
`(2a) are given as 490 and 1330 A,
`and paramyosin.
`
`2250
`
`2350
`
`I
`
`I
`
`I
`2550
`
`I
`o 2450
`X (A)
`un-
`rotation
`Specific
`complexes.
`FIG. 4. ORD of myosin-SDS
`1, 1.4 g of SDS per g of protein,
`corrected
`for refractive
`index.
`buffer, pH 7.2; 2, 0.4 g of SDS
`ionic strength
`= 0.026, phosphate
`per g of protein,
`ionic strength
`= 0.52, phosphate
`buffer, pH 7.2.
`
`The use
`radius.)
`the Stokes
`called
`is sometimes
`R,,
`parameter,
`4 makes
`it possible
`to compare
`the results of viscosity
`of Equation
`and gel chromatography.
`All existing
`theories of gel chromatog-
`raphy are expressed
`in terms of R,, and we have used
`this experi-
`
`GNE 2003
`Page 4
`
`

`
`Issue of October 10, 1970
`
`J. A. Reynolds and C. Tanford
`
`5165
`
`Acknowledgment-We
`to Dr.
`to express our gratitude
`wish
`Wayne W. Fish
`for many
`helpful
`discussions
`and suggestions
`during
`the course of
`this work.
`
`in
`
`Biophys.
`
`Acta, 160,
`
`DISCUSSION
`of Biological MembranesSome
`to Ultrastructure
`Application
`defined proteins,
`such as cytochrome
`c and ATPase
`functionally
`from Streptococcus
`jaecalis,
`are bound
`tightly
`to biological
`mem-
`branes but are apparently
`external
`to the bimolecular
`lipid
`leaflet
`(19,20).
`The
`interaction
`between
`these proteins
`and
`the lipid
`is
`thought
`to be coulombic
`in the case of cytochrome
`c and
`through
`a ternary Mg+f
`complex
`in
`the case of
`the bacterial
`ATPase.
`However,
`there are polypeptide
`chains which are extremely
`diffi-
`cult
`to dissociate
`from membrane
`lipid
`and are probably
`inti-
`mately
`associated with
`the
`lipid
`through
`primarily
`hydrophobic
`forces (3, 5).
`It
`is this
`latter group of proteins
`for which
`the pro-
`tein-SDS
`complex may be a reasonable model.
`If
`the interaction
`of amphiphilic
`lipid with some specific membrane
`proteins
`paral-
`lels that of SDS, one can postulate
`an extended,
`ordered polypep-
`tide chain associated with
`the hydrophobic
`regions of the bimo-
`lecular
`leaflet.
`The
`charged or polar groups
`(or both)
`on
`this
`chain would be expected
`to interact with
`the
`lipid phosphate
`head
`groups and be exposed
`to
`the aqueous media which bathes
`the
`membrane
`in Go.
`It should be noted particularly
`that
`the use
`of the SDS-protein
`complex
`as a model
`system
`for the membrane
`structure
`precludes
`the possibility
`of the protein
`being
`in either a
`random
`coil or globular
`form when
`it is associated with
`lipid.
`of
`SDS Gel ElectrophoresisThe
`recently
`developed
`method
`determining
`the molecular
`weight of polypeptide
`chains
`in SDS
`on polyacrylamide
`gels
`(21, 22) has been empirically
`successful,
`but has
`thus
`far been without
`theoretical
`basis.
`The electro-
`in
`phoretic mobility
`of polypeptide
`chains
`these gels can be a
`unique
`function
`of molecular
`weight only when
`the
`following
`cri-
`teria are met.
`(a) The charge per unit mass must be approxi-
`mately
`constant.
`(b) The hydrodynamic
`properties must be a
`function
`of molecular
`length only.
`The high
`level of binding
`of
`SDS
`to proteins
`and
`the constant
`binding
`ratio on a gram
`to gram
`basis under
`the experimental
`conditions
`used
`in polyacrylamide
`gel electrophoresis
`assure a constant
`charge per unit mass.
`addition,
`the data
`in
`this paper
`show
`that
`the hydrodynamic
`properties
`of protein-SDS
`complexes
`are a unique
`function
`of the
`polypeptide
`chain
`length.
`
`In
`
`1.
`
`2.
`
`3.
`
`4.
`5.
`
`6.
`
`7.
`
`8.
`
`9.
`10.
`
`11.
`12.
`13.
`
`14.
`
`15.
`
`16.
`
`17.
`
`18.
`
`19.
`
`20.
`21.
`
`22.
`
`REFERENCES
`equilibria
`J. A., Multiple
`IYND REYNOLDS,
`J.,
`STEINHARDT,
`Academic
`Press, New York,
`1970..
`-
`proteins,
`J. A.. AND TANFORD.
`C.. Proc. Nat. Acad. Sci.
`REYNOLDS.
`’
`U. S. A:, 66, 1602 (1970).
`’
`JI, T. H., AND BENSON, A. A., Biochim.
`686 (1968).
`(1969).
`8,431O
`BRAUN, P. E., AND RADIN, N. S., Biochemistry,
`STOECKENIUS, W., AND ENGLEMAN, D. M., J. Cell Biol., 42,613
`(1969).
`LEVY, H. B., END SOBER, H. A., Proc. Sot. Exp. Biol. Med., 103,
`250 (1960).
`(Editors),
`in S. P. COLOWICK AND N. 0. KAPLAN
`PERRY, S. V.,
`in enzymology,
`II, Academic Press, New York,
`Vol.
`Methods
`1955, p. 582.
`C., AND HANAHAN,
`DODGE,
`J. T., MITCHELL,
`Biochem.
`Biophys.,
`100, 119 (1963).
`TANFORD, C., Advan. Protein
`Chem., 23, 121 (1968).
`John Wiley
`TANFORD, C., Physical
`chemistry
`of macromolecules,
`and Sons, New York,
`1961.
`49, 5486 (1968).
`ULLMAN, R., J. Chem. Phys.,
`J., BND DOTY, P., J. Mol. Biol., 12,549
`EIGNER,
`ANACKER, E. W., RUSH, R. M., AND JOHNSON,
`Chem., 68, 81 (1964).
`FISH, W. W., REYNOLDS,
`Chem., 246, 5166 (1970).
`LOWEY, S., KUCERA,
`J., AND HOLTZER, A., J. Mol. Biol., 7,234
`(1963).
`HOLTZER, A., CLARK, R.,
`(1965).
`J., EMERSON, M. F., AND HOLTZER, A., J. Amer.
`OLANDER,
`Chem. Sot., 89, 3058 (1967).
`SCHECHTER, E., AND BLOUT, E. R., Proc. Nat. Acad.
`U. S. A., 51, 794 (1964).
`CHAPMAN, D., AND WALLACH, D. F. H., Biological membranes,
`Academic
`Press, New York,
`1968.
`ABRAMS, A., AND BARON, C., Biochemistry,
`(1968).
`7,501
`SHAPIRO, A. L., VIRUEL.~, E., AND MAIZEL,
`J. V., JR., Biochem.
`Res. Commun.,
`28, 815 (1967).
`Biophys.
`WEBER, K., AND OSBORN, M., J. Biol. Chem., 244,4406
`
`D.
`
`J., Arch.
`
`(1965).
`J. S., J. Phys.
`
`J. A., AND TANFORD,
`
`C., J. Biol.
`
`.~ND LOWEY, S., Biochemistry,
`
`4,240l
`
`Sci.
`
`(1969).
`
`GNE 2003
`Page 5
`
`

`
`The Gross Conformation of Protein-Sodium Dodecyl Sulfate Complexes
`Jacqueline A. Reynolds and Charles Tanford
`1970, 245:5161-5165.
`
`J. Biol. Chem.(cid:160)(cid:160)
`
`Access the most updated version of this article at
`
`
`
` http://www.jbc.org/content/245/19/5161
`
`Alerts:
`When this article is cited
`•(cid:160)
`
` When a correction for this article is posted
`•(cid:160)
`(cid:160)
`Click here
`
`
`
` to choose from all of JBC's e-mail alerts
`
`This article cites 0 references, 0 of which can be accessed free at
`
` http://www.jbc.org/content/245/19/5161.full.html#ref-list-1
`(cid:160)
`
`GNE 2003
`Page 6
`
`(cid:160)