The Amide II Band

The amide II band of primary amides comes mainly from the scissoring motion ofNH2. The band is located at 1650-1620cm in solids, and at 1620-1585cm in dilute solutions (Randall et al., 1949 Clarke et al, 1949 Richards and Thompson, 1947 Jones and Cleverley, 1956). The band is weaker in most cases than the amide 1 band. Tertiary amides have no band in the 1620 cm region since they have no NH2 group. 

The amide II band in secondary amides appears to be due to a mixed vibration involving the N—H in-plane bending and the C—N stretching vibration (Fraser and Price, 1952, 1953 Miyazawa et al., 1956, 1958). (For further discussion on this point see Chapter 10.) The amide I band also appears to be produced by a mixed vibration, the major contribution coming from C=0 stretching. 

Primary and secondary amides show a second strong band in the 1600—1500 cm region which is absent from the spectra of tertiary amides and also from those of small ring cyclic lactams under normal conditions. The origins of this band are different in the two cases. Nevertheless, the band is referred to generally as the amide II band. 

---

The polyamides and polyureas exhibited broad, intense N-H stretches around 3300 cm- , a very strong carbonyl stretching vibration was present at 1630 cm- . The amide II band was evident near 1540 cm- . jn addition, sp C-H stretches occurred around 3100 cm- an(j asymmetric and symmetric sp3 c-H stretches at 2950 and 2860 cm- , respectively. The polyurethane showed the carbonyl absorption near 1700 cm-1 and C-0 stretches in the vicinity of... 

Primary amides and secondary amides, and a few lactams, display a band or bands in the region of 1650-1515 cm-1 caused primarily by NH2 or NH bending the amide II band. This absorption involves coupling be-... 

II band in the neighborhood of 1540 cm. 1 disappears and is replaced by a band at about 1450 cm. 1. The extent of diminution of the amide II band is, at least in principle, a measure of the extent of proton exchange. Little effect is observed in ordinary water since, of course, proton exchange does not occur. Table II summarizes some of the characteristic frequencies observed with proteins (83). 

The observation by Maddy and Malcolm (53) that the amide I band of bovine erythrocyte ghosts in D20 is not shifted is remarkable because it implies that all of the membrane protein is either deeply buried in an environment of hydrophobic lipids or exists in a tightly folded a-helical conformation. We have examined extensively the infrared spectra of bovine erythrocyte ghosts, both as dry films and as intact ghosts in D20 and H20 (73). The results for dry films essentially agree with those of other workers and show no evidence of f3 structure. Little change occurs in water. In D20, however, we consistently obtained a shift in the amide I band and a considerable decrease in absorption of the amide II band. 

The spectra of dry films of intact ghosts prepared by lysis in 20 millios-molal phosphate buffer (21) and of ghost protein prepared by cold butyl alcohol extraction (51) are shown in Figure 6. In both cases the amide I band occurs at 1651 cm. 1 and shows no shoulder near 1630 cm. 1, characteristic of fl structure. The amide II band is also unaffected by removal of lipid and occurs at 1540 cm."1. As expected, extraction of lipid results in removal of the band at 1737 cm. 1 assigned to lipid ester carbonyl stretch and a decrease of the band at about 1455 cm."1 arising from methylene and methyl bending. 

The amide II band is not present in the spectra of lactams. As in the case of cyclic ketones, however, the carbonyl (amide I) band shifts to higher frequency as the size of the ring decreases. 

These kinetics studies required development of reproducible criteria of subtraction of foe H-O-H bending band of water, which completely overlaps foe Amide I (1650 cm 1) and Amide II (1550 cm"1) bands (98). In addition, correction of foe kinetic spectra of adsorbed protein layers for foe presence of "bulk" unadsorbed protein was described (99). Examination of kinetic spectra from an experiment involving a mixture of fibrinogen and albumin showed that a stable protein layer was formed on foe IRE surface, based on foe intensity of the Amide II band. Subsequent replacement of adsorbed albumin by fibrinogen followed, as monitored by foe intensity ratio of bands near 1300 cm"1 (albumin) and 1250 cm"1 (fibrinogen) (93). In addition to foe total amount of protein present at an interface, foe possible perturbation of foe secondary structure of foe protein upon adsorption is of interest. Deconvolution of foe broad Amide I,II, and m bands can provide information about foe relative amounts of a helices and f) sheet contents of aqueous protein solutions. Perturbation of foe secondary structures of several well characterized proteins were correlated with foe changes in foe deconvoluted spectra. Combining information from foe Amide I and m (1250 cm"1) bands is necessary for evaluation of protein secondary structure in solution (100). 

Once adsorbed on a substrate, albumin was firmly bound, as indicated by the minimal change in the intensity of the Amide II band after the saline rinse was initiated. At pH 4.0 and 7.4, very little protein desorbed from the... 

The filtered preparation of gum arabic contained 13 pg/mg protein (dry wt). When a 0.01% (w v) aqueous solution, at pH 6.5, was exposed to a Ge IRE, no polysaccharide was observed to adsorb from a flowing solution at the aqueous/solid interface, as the characteristic C-0 stretching bands of gum arabic were not visible in the water-subtracted spectra. The protein (1.3%) associated with gum arabic demonstrated a high affinity for the Ge surface, adsorbing from a flowing solution at a concentration of 1.3 ppm, while polysaccharide, present at a concentration of 100 ppm, did not adsorb on the IRE. Distinct Amide I and Amide II bands of this adsorbed protein are visible in the water subtracted spectra. At the end of 4 hr, the 1549 cm 1 band intensity was 0.9 mAU. Rinsing with Milli-Q water (pH 6.5) did not affect the Amide II band intensity, indicating that the protein was firmly adsorbed to the Ge surface. 

At a gum arabic concentration of 0.1% (w v), protein and a small amount of polysaccharide were detected at the Ge surface as shown in Figure 5. The Amide II band intensity increased slowly and steadily throughout the initial 4-hr period, as shown in Figure 6. The Amide II band intensity peaked at approximately 11 mAU however, protein adsorption did not stop before the water rinse was initiated. Polysaccharide adsorbed rapidly onto the IRE... 

At a gum arabic concentration of 1.0% (w v), polysaccharide was detected at the Ge surface within the first minute of exposure to the flowing solution as shown in the first spectrum (T 0 min) in Figure 7. The intensities of both the protein and polysaccharide bands then rose rapidly in a 15-min period of time as shown in Figure 6. The 1070 cm 1 polysaccharide band plateaued at 14.7 mAU after 30 min whereas the Amide II band stabilized at 12.5 mAU after approximately 2 hr. The intensity of the 1070 cm 1 polysaccharide band dropped rapidly when Milli-Q water was pumped through the flow cell. A 90% decrease in the 1070 cm 1 band intensity occurred over the 4-hr rinse period. The final intensity of this band was not significantly different from the intensity observed when gum arabic was adsorbed onto germanium at a concentration that was 10 times less. The protein was more firmly bound to the IRE surface as indicated by the Amide II band intensity which dropped less than 10% during the rinse period. Only 15% less protein remained firmly attached to the Ge IRE when it was adsorbed from a gum arabic solution concentration of 0.1% as compared to 1%. Experiments to study adsorption of proteins and polysaccharides on copper and nickel are not yet complete, but appear to show similar trends. 

At the low protein concentration (0.01% w v) used in our studies, the signal from protein can be assumed to be entirely due to adsorbed material. Very little or no change in the Amide II band intensities occurred when protein in the bulk phase was replaced by saline or water during the rinse period. Protein remaining after the saline rinse can be considered as firmly adsorbed material. 

Figure 4. Adsorption of 190 ppm lysozyme onto hydrophobic and hydrophilic Ge IREs as determined by the amide II band absorbance after subtraction of the solvent spectrum.

While limited changes in the features of the amide I band are observed after BSA adsorption on the two silicas the amide II band, present in the spectrum of the native BSA in solution (curve a) is absent in the spectrum of the protein adsorbed on A50 (curve b), whilst it is partially maintained in that of BSA on Qzm (curve c). This behaviour indicates that the interaction with the surface of the amorphous silicas resulted in an opening of the hydrophobic pockets, allowing the N-H groups therein contained to be converted by contact with D20 in N-D, producing an IR absorption at lower frequency (ca. 1450 cm, not shown). On Qzm the conformational changes are less pronounced. 

Fig. 8. Simulation of IRRAS spectra of a /1-sheet lying flat at the air/buffer interface. The calculation was performed for p-polarized light and different incident angles (1) 32°, (2) 40°, (3) 48°, (4) 56°, (5) 62° for the amide I bands at 1627 and 1690 cm-1 and the amide II band at 1535 cm-1. Taken from Ref. [69] permission pending from Wiley Interscience Journals.

Fig. 12.6. Optical (a) and FTIR (b) imaging data for a slowly dehydrated larva. Mapped are intensities of the characteristic 992-cm-1 peak, which were normalized by being divided by that of the amide II band...

Suspension of soluble proteins in 2H20 shifts the amide II band (principally N-H bending) to frequencies near 1450cm- (amide II ). 

---