Biochimica et Biophysica Ac~a, 1121 (1992) 183-188 © 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4838/92/$05.00
183
BBAPRO 341q8
Aggregation of chymotrypsinogen: portrait by infrared spectroscopy Ashraf A. Ismail, Henry H. Mantsch and Patrick T.T. Wong Steacie Institute for Molecular Sciences, National Rt:~t'arclz Council of Canada, Ottawa (Canada) (Received 4 October 1991)
Key words: High pressure: Temperature; Protein: Aggregation; Secondary structure; Infrared spectro.~copy; Denaturation
Changes in the secondary structure and aggregation of chymotrypsinogen were investigated by infrared difference spectroscopy in conjunction with temperature and pressure tuning IR spectroscopy; both the amide 1' band and side chain bands were studied. A prominent component of the amidc !' band in the difference spectrum obtained upon cooling a chymotrypsinogcn solution, or increasing the hydrostatic pressure, was observed in :he region between 1627 and 1622 cm-~. Under denaturing conditions a white gel was formed, which is attributed to irreversible self-association or aggregation. This process was accompanied by the appearance of two new amide I" bands in the infrared spectrum of the protein: a very strong band at 1618 cm- 1 and a weak band at 1685 cm- l. These bands are assigned to peptide segments with anti-parallel aligned fl-strands.
Introduction Only a few experimental methods are available which can provide information on the structural changes in proteins that take place under a variety of physicochemical conditions favouring the formation of precipitates or gel assemblies, in this work, we employ infrared spectroscopic methods in order to investigate protein-protein interaction a n d / o r protein aggregation. Indeed, IR spectroscopy is one of the most powerful and simple methods for elucidating structure at the molecular level. To date, the majority of infrared studies of, proteins have focused on the elucidation of the secondary structure~ employing the structure-sensitive amide I band in the infrared spectrum [1,2]. However, examination of the side chain v~rations of prot~;ins can provide additional information on interactions within the protein [3]. This work entails further exploitation of such a region of the infrared spectrum, in addition to the structure-sensitive amide I band, to elucidate, at the molecular level, the events leading to protein reorganization and aggregation.
t Issued as NRCC publication No. 3281 I. Correspondence: A.A. lsmail. Steacie Institute for Molecular Sciences, National Research Council of Canada. Ottawa, Ontario, Canada KIA OR6.
Experimental procedures Chymotrypsinogen A (recrystallized 6 times) was obtained from Sigma (St. Louis, MO). DzO was from MSD Isotopes (Montreal, Canada). Chymotrypsinogen solutions in phosphate-buffered D , O of various ionic strengths were made up in concentrations ranging from 0.3 to 10% w / v . The pH values recorded for these solutions are uncorrected p D values. The solutions were left to stand for a minimum of 24 h at room temperature prior to the IR spectroscopic measurements. F o r the pressure experiments, a diamond anvil cell was employed. A drop of the chymotrypsinogen solution (along with a very small amount of quartz powder) was placed in a 0.37-mm hole of a 0.23 mmthick stainless steel gasket resting on a diamond window. Pressures at the sample were determined from the 695 c m - t infrared absorption of a-quartz, as described earlier [4]. Infrared spectra were recorded on a Digilab FTS-60 Fourier-transform spectrometer equipped with a Iiquid-nitrogen-cooled mercury-cadmium-telluride detector. A total of 512 scans at 4 c m - i resolution were co-added for each spectrum. For the temperature studies, the chymotrypsinogen solution was placed in a CaF., cell with a 50 g m Teflon spacer. The temperature of the sample was regulated by mounting the IR cell in a temperature-controlled ceil holder. A deuterated triglycine sulfate detector was employed. A total
184 of 256 scans at 4 cm-~ resolution were co-added. Resolution enhancement of the infrared bands was carried out as described previously [5,6]. Results ,
Resolution enhancement through band narrowing of the amide 1' band in the infrared spectrum of chymotrypsinogen A reveals seven peaks at 1695, 1687, 1681, 1674, 1665, 1652 and 1637 cm- ~ and a shoulder at 1628 cm -l (Fig. 1). The 1687 and 1637 c m - ! bands were previously assigned to an anti-parallel /~-sheet structure while the shoulder at 1628 cm-~ was assigned to an extended anti-parallel /3-sheet [7]. The 1652 eraband was assigned to an a-helical component while bands at 1665, 1674 and 1681 crn -~ were assigned to turns and bends [7]. As a result of the excellent signalto-noise ratio (approaching 100000" 1) it was possible to perform deconvolution with a high resolution enhancement factor (k = 2.8). An additional band not previously resolved can be discerned at 1645 cm -~. The presence of this band was further affirmed by a new resolution enhancement method employing maximum entropy [8]. The 1645 c m - ' band is assigned to random or unordered structure in accordance with Susi et al. [7].
75
~,~r"
~.
-
'
tt~tO ~tD
I
to
f.D ¢D e~
I
u-) i ~-"
I
!
I
i
1700
1650
1600
WGvenumber,
cm
I
-1
Fig. 1. Amide I' band of chymot~psinogen A before (bottom), and after Fourier self deconvolution by use of a Loren~ian line shape with a band~dth of 13 cm-z and a band narrowing factor of 2.8 (top).
_=~V ~
37.7
_i..,
41.9
1700 1675 16~50 16~25 1600 Wovenumber,
I
h~)
29.0 33.3
cm
'
-1
Fig. 2. Stacked plot of the deconvolved amide 1' band contour of chymotryosinogen A (10% w / v in 0.01 M phosphate buffer in D 2 0 at pH 8.5), recorded as a function of decreasing the temperature from 42 to 7.5°C. The cbymotrypsinogen A solution was allowed to equilibrate for 15 rain at each temperature prior to recording of the next infrared spectrum.
1'--
^
" r
f~ ~) ¢D
~--(D I
12.4 16.6 2o8
in the present study, IR spectra were recorded, in 5°C increments, as the temperature was decreased from 42 to 7°C. Fig. 2 shows a stacked plot of the deconvolved spectra. It is apparent from a visual examination of the spectra that the overall amide I' band profile does not undergo a perceptible change in this temperature range. The difference spectrum obtained from the infrared spectra recorded at the two extremes of the temperature range studied reveals a band at 1627 cm -~, with an intensity on the order of 8% of that of the amide 1' peak maximum in the unsubtracted spectrum recorded at 7°C. Autoscaling the difference spectrum reveals three bands at 1649, 1627 and 1564 cm-~. The appearance of these bands was found to be re,,ersible upon warming the sample. The infrared difference spectra obtained by subtracting the spectrum recorded at 42°C from spectra recorded at lower temperatures are shown in Fig. 3. A plot of the integrated intensity of the 1627 cm-~ band, obtained from these difference spectra was found to be linear over the whole temperature range studied. The above temperature study was repeated for a series of solutions containing varying concentrations of chymotrypsinogen A, covering the range 0.3-10%. In all cases, the intensity of the 1627 cm-~ band in the difference spectra was found to increase with decreasing temperature. Furthermore, at each temperature investigated, the integrated intensity of the 1627 eraband decreased linearly with decreasing concentration of chymotrypsinogen; however, the decrease in inten-
185 sity of the 1627 cm-~ band was proportional to the decrease in the amide I' peak maximum (at 1637 cm - I ) with decreasing concentration. This indicates that the 1627 c m - i band is representative of an equilibrium state that is sensitive to variations in temperature, but not to changes in concentration. In all the above studies, the growth of the 1627 cm -I band was accompanied by the appearance and growth of a band at ~ 1564 c m - i (see Fig. 3). This band originates from the antisymmetric C O O - stretching vibration of the side chain carboxylate groups of glutamate [3,9]. The appearance of this band implies that the structural changes taking place upon cooling are likely to result from solvation of Glu residues involved in ionic interactions. At lower temperatures the band appears to consist of two components, reflecting two different hydration environments around this group. Prolonged cooling (5 days at 4°C) results in the appearance of a second band at 1618 c m ~ Fig. 4 shows the difference spectrum (solid line) obtained by subtraction of two spectra of a chymotrypsinogen solution at 4°C recorded 5 days apart. Two peaks can be readily discerned, at 1622 and 1618 c m - i . The differ-
r--
oq tD
C'400 ¢',,I ,,(,0 (,,ID
4,,1. /
O. 005L,) E O _1:3
t
t
L
r
o
,i
O3
183
BBAPRO 341q8
Aggregation of chymotrypsinogen: portrait by infrared spectroscopy Ashraf A. Ismail, Henry H. Mantsch and Patrick T.T. Wong Steacie Institute for Molecular Sciences, National Rt:~t'arclz Council of Canada, Ottawa (Canada) (Received 4 October 1991)
Key words: High pressure: Temperature; Protein: Aggregation; Secondary structure; Infrared spectro.~copy; Denaturation
Changes in the secondary structure and aggregation of chymotrypsinogen were investigated by infrared difference spectroscopy in conjunction with temperature and pressure tuning IR spectroscopy; both the amide 1' band and side chain bands were studied. A prominent component of the amidc !' band in the difference spectrum obtained upon cooling a chymotrypsinogcn solution, or increasing the hydrostatic pressure, was observed in :he region between 1627 and 1622 cm-~. Under denaturing conditions a white gel was formed, which is attributed to irreversible self-association or aggregation. This process was accompanied by the appearance of two new amide I" bands in the infrared spectrum of the protein: a very strong band at 1618 cm- 1 and a weak band at 1685 cm- l. These bands are assigned to peptide segments with anti-parallel aligned fl-strands.
Introduction Only a few experimental methods are available which can provide information on the structural changes in proteins that take place under a variety of physicochemical conditions favouring the formation of precipitates or gel assemblies, in this work, we employ infrared spectroscopic methods in order to investigate protein-protein interaction a n d / o r protein aggregation. Indeed, IR spectroscopy is one of the most powerful and simple methods for elucidating structure at the molecular level. To date, the majority of infrared studies of, proteins have focused on the elucidation of the secondary structure~ employing the structure-sensitive amide I band in the infrared spectrum [1,2]. However, examination of the side chain v~rations of prot~;ins can provide additional information on interactions within the protein [3]. This work entails further exploitation of such a region of the infrared spectrum, in addition to the structure-sensitive amide I band, to elucidate, at the molecular level, the events leading to protein reorganization and aggregation.
t Issued as NRCC publication No. 3281 I. Correspondence: A.A. lsmail. Steacie Institute for Molecular Sciences, National Research Council of Canada. Ottawa, Ontario, Canada KIA OR6.
Experimental procedures Chymotrypsinogen A (recrystallized 6 times) was obtained from Sigma (St. Louis, MO). DzO was from MSD Isotopes (Montreal, Canada). Chymotrypsinogen solutions in phosphate-buffered D , O of various ionic strengths were made up in concentrations ranging from 0.3 to 10% w / v . The pH values recorded for these solutions are uncorrected p D values. The solutions were left to stand for a minimum of 24 h at room temperature prior to the IR spectroscopic measurements. F o r the pressure experiments, a diamond anvil cell was employed. A drop of the chymotrypsinogen solution (along with a very small amount of quartz powder) was placed in a 0.37-mm hole of a 0.23 mmthick stainless steel gasket resting on a diamond window. Pressures at the sample were determined from the 695 c m - t infrared absorption of a-quartz, as described earlier [4]. Infrared spectra were recorded on a Digilab FTS-60 Fourier-transform spectrometer equipped with a Iiquid-nitrogen-cooled mercury-cadmium-telluride detector. A total of 512 scans at 4 c m - i resolution were co-added for each spectrum. For the temperature studies, the chymotrypsinogen solution was placed in a CaF., cell with a 50 g m Teflon spacer. The temperature of the sample was regulated by mounting the IR cell in a temperature-controlled ceil holder. A deuterated triglycine sulfate detector was employed. A total
184 of 256 scans at 4 cm-~ resolution were co-added. Resolution enhancement of the infrared bands was carried out as described previously [5,6]. Results ,
Resolution enhancement through band narrowing of the amide 1' band in the infrared spectrum of chymotrypsinogen A reveals seven peaks at 1695, 1687, 1681, 1674, 1665, 1652 and 1637 cm- ~ and a shoulder at 1628 cm -l (Fig. 1). The 1687 and 1637 c m - ! bands were previously assigned to an anti-parallel /~-sheet structure while the shoulder at 1628 cm-~ was assigned to an extended anti-parallel /3-sheet [7]. The 1652 eraband was assigned to an a-helical component while bands at 1665, 1674 and 1681 crn -~ were assigned to turns and bends [7]. As a result of the excellent signalto-noise ratio (approaching 100000" 1) it was possible to perform deconvolution with a high resolution enhancement factor (k = 2.8). An additional band not previously resolved can be discerned at 1645 cm -~. The presence of this band was further affirmed by a new resolution enhancement method employing maximum entropy [8]. The 1645 c m - ' band is assigned to random or unordered structure in accordance with Susi et al. [7].
75
~,~r"
~.
-
'
tt~tO ~tD
I
to
f.D ¢D e~
I
u-) i ~-"
I
!
I
i
1700
1650
1600
WGvenumber,
cm
I
-1
Fig. 1. Amide I' band of chymot~psinogen A before (bottom), and after Fourier self deconvolution by use of a Loren~ian line shape with a band~dth of 13 cm-z and a band narrowing factor of 2.8 (top).
_=~V ~
37.7
_i..,
41.9
1700 1675 16~50 16~25 1600 Wovenumber,
I
h~)
29.0 33.3
cm
'
-1
Fig. 2. Stacked plot of the deconvolved amide 1' band contour of chymotryosinogen A (10% w / v in 0.01 M phosphate buffer in D 2 0 at pH 8.5), recorded as a function of decreasing the temperature from 42 to 7.5°C. The cbymotrypsinogen A solution was allowed to equilibrate for 15 rain at each temperature prior to recording of the next infrared spectrum.
1'--
^
" r
f~ ~) ¢D
~--(D I
12.4 16.6 2o8
in the present study, IR spectra were recorded, in 5°C increments, as the temperature was decreased from 42 to 7°C. Fig. 2 shows a stacked plot of the deconvolved spectra. It is apparent from a visual examination of the spectra that the overall amide I' band profile does not undergo a perceptible change in this temperature range. The difference spectrum obtained from the infrared spectra recorded at the two extremes of the temperature range studied reveals a band at 1627 cm -~, with an intensity on the order of 8% of that of the amide 1' peak maximum in the unsubtracted spectrum recorded at 7°C. Autoscaling the difference spectrum reveals three bands at 1649, 1627 and 1564 cm-~. The appearance of these bands was found to be re,,ersible upon warming the sample. The infrared difference spectra obtained by subtracting the spectrum recorded at 42°C from spectra recorded at lower temperatures are shown in Fig. 3. A plot of the integrated intensity of the 1627 cm-~ band, obtained from these difference spectra was found to be linear over the whole temperature range studied. The above temperature study was repeated for a series of solutions containing varying concentrations of chymotrypsinogen A, covering the range 0.3-10%. In all cases, the intensity of the 1627 cm-~ band in the difference spectra was found to increase with decreasing temperature. Furthermore, at each temperature investigated, the integrated intensity of the 1627 eraband decreased linearly with decreasing concentration of chymotrypsinogen; however, the decrease in inten-
185 sity of the 1627 cm-~ band was proportional to the decrease in the amide I' peak maximum (at 1637 cm - I ) with decreasing concentration. This indicates that the 1627 c m - i band is representative of an equilibrium state that is sensitive to variations in temperature, but not to changes in concentration. In all the above studies, the growth of the 1627 cm -I band was accompanied by the appearance and growth of a band at ~ 1564 c m - i (see Fig. 3). This band originates from the antisymmetric C O O - stretching vibration of the side chain carboxylate groups of glutamate [3,9]. The appearance of this band implies that the structural changes taking place upon cooling are likely to result from solvation of Glu residues involved in ionic interactions. At lower temperatures the band appears to consist of two components, reflecting two different hydration environments around this group. Prolonged cooling (5 days at 4°C) results in the appearance of a second band at 1618 c m ~ Fig. 4 shows the difference spectrum (solid line) obtained by subtraction of two spectra of a chymotrypsinogen solution at 4°C recorded 5 days apart. Two peaks can be readily discerned, at 1622 and 1618 c m - i . The differ-
r--
oq tD
C'400 ¢',,I ,,(,0 (,,ID
4,,1. /
O. 005L,) E O _1:3
t
t
L
r
o
,i
O3
