Conformational comparison of human pituitary growth hormone and human chorionic somatomammotropin (human placental lactogen) by second-order absorption spectroscopy

ARCHIVES

OF BIOCHEMISTRY

Vol. 233, No. 1, August

AND

15,

BIOPHYSICS

pp. 219-227, 1984

Conformational Comparison of Human Pituitary Growth Hormone and Human Chorionic Somatomammotropin (Human Placental Lactogen) by Second-Order Absorption Spectroscopy’ THOMAS Laboratmy

of Molecular

A. BEWLEY

Endocrinology,

Received February

University

AND

CHOH HA0

of California,

LIZ

San Francisco,

13, 1984, and in revised form April

California

94143

25, 1984

Second-order absorption spectra strongly suggest the presence of a hydrogen bond between the single Trp of human pituitary growth hormone (hGH) and a carboxylate ion. This hydrogen-bonded complex is buried within the hydrophobic interior of the hGH molecule. Although the homologous Trp in human chorionic somatomammotropin [human placental lactogen, HCS(hPL)] is also buried within the hydrophobic interior of the molecule, there is no evidence that it is hydrogen bonded in the native protein. However, during the early stages of thermolysin digestion of HCS(hPL), both difference and second-order absorption spectra do indicate the transient presence of a similar hydrogen-bonded Trp-carboxylate complex. The molar extinction coefficients of hGH and HCS(hPL) have been refined.

While human pituitary growth hormone (hGH)3 and the placental hormone HCS(hPL) are both potent lactogens (l-3), they differ markedly in growth promoting potency (3-6). The primary structures (7,8) of these two proteins have been shown (9, 10) to be very closely related, with 28 replacements out of 191 residues, only seven of which can be considered nonconservative. Many laboratories have reported both similarities and differences in these proteins’ secondary and tertiary structures (11, 13-16).

i From the Laboratory of Molecular Endocrinology, University of California, San Francisco, Calif. 94143. This work was supported in part by Grant AM-18677 from the National Institutes of Health, and was presented in part at the “Human Growth Hormone Symposium,” Baltimore, Maryland, Nov. 1983. * To whom correspondence should be addressed. 3 Abbreviations used: hGH, human pituitary growth hormone; HCS(hPL), human chorionic somatomammotropin (human placental lactogen); EM, molar extinction coefficient; A, absorptivity; Abs., absorbance; E/S, enzyme to substrate ratio. 219

It seems quite probable that it is the similarities in primary, secondary, and tertiary structure that are responsible for the nearly equivalent lactogenic activities of these molecules. It is equally probable that one or more of the few differences in primary structure, acting through subtle alterations in secondary and/or tertiary structures, give rise to the quantitative disparity in growth-promoting potencies. Much might be learned about how hGH functions to promote growth by understanding why HCS(hPL) does not. It has been shown that, in both proteins, the single Trp side chain is buried within the hydrophobic interior of the molecule (11,13-16). However, one of the major differences noted in previous conformational studies involves the local microenvironment of this residue. This difference, whose precise nature has never been elucidated, gives rise to inequalities in uv absorption (11, 12) and significant differences in CD spectra (11, 13-16). The differences in uv absorption are particularly noteworthy in view of the fact that all major chromo0003-9861/84 $3.00 Copyright All rights

0 1984 by Academic Press. Inc. of reproduction in any form reserved.

BEWLEY

220

phoric residues, including the single Trp, all 8 Tyr, both disulfide bonds, and 11 of the 13 Phe residues in hGH occur in identical sequence positions in HCS(hPL), and are flanked by identical or highly homologous peptide segments. In this report, the uv absorption spectra of hGH and HCS(hPL) are further compared. In addition, the molar extinction coefficients (EM) of both hormones have been reevaluated in 0.1 M Tris-HCl (pH 8.2). MATERIALS

AND

AND

LI

.---

METHODS

The hGH and HCS(hPL) were prepared as described by Li et CLL(26) and Neri et al (27) respectively. Thermolysin, lot no. ‘73326, was obtained from Calbiochem. All uv spectra were taken at 25°C on a Perkin-Elmer, Model 552 double-beam, recording spectrophotometer. The spectrophotometer was also equipped with the second-derivative accessory supplied by the manufacturer. Second-order spectra were scanned at 60 nm/min. The wavelength positions of all second-order minima were corrected as described previously (19). Molar extinction coefficients were determined from zero-order spectra of native and thermolysin-digested proteins (19). Protein concentrations were determined using these extinction coefficients. Two different preparations of each hormone were used, with three to five digestions of each preparation. Additional thermolysin digests were performed within the spectrophotometer at 25’ + O.l”C in 0.1 M Tris-HCl buffer (pH 8.2). These reactions were followed both by second-order spectra and by the difference spectra1 technique (20). RESULTS

AND

HCS IhPLI

WAVELENGTH

(nml

FIG. 1. Zero-order absorption spectra of (A) hGH and (B) HCS(hPL), including the native protein (p), a 24-h thermolysin digest at pH 8.2 (- - -), and the digest after titration to pH 1.5 (----). The heavier dashed line near the bottom of each set of spectra is the light-scattering correction determined by the log-log procedure described previously (19). The zero position on the ordinate scale has been placed at the height of the light-scattering correction under the wavelength of maximum absorption.

DISCUSSION

Zer@Order Spectra and EM The zero-order uv absorption spectra of native hGH and HCS(hPL) at pH 8.2 are presented in Fig. 1. Both spectra display a very weak shoulder between 310 and 300 nm, a weak shoulder near 290 nm, a distinct shoulder near 285 nm, and an absorption maximum at 276.6 + 0.1 nm for hGH and 276.4 f 0.1 nm for HCS(hPL). In addition, there is a series of closely spaced, weak shoulders between 270 and 245 nm. Although there is considerable apparent similarity in these two spectra, the EM for hGH was found to be 20,350 f 100 (M-’ cm-l) at the absorption maximum, while that for HCS(hPL) was found to be only

19,200 + 180 (Map cm-‘). These EM values correspond to specific absorptivities of A&l’% 1.ocm= 0.920 + 0.004 for hGH, and Ao.1” 1.0em= 0.863 f 0.008 for HCS(hPL). Both are within 4% of previously published values (12, 21, 22). This difference in the EM of the two proteins is significant, and clearly implies that the conformations of the molecules have produced unequal hyperchromic effects on one or more of the major chromophoric groups (19,23). Since hyperchromic effects are generally accompanied by red-shifting of the absorption bands toward higher wavelengths (19,22, 23), we might expect to see some absorption features of hGH at higher wavelengths than in HCS(hPL). Indeed, the absorption

COMPARISON

OF GROWTH

maximum of hGH is slightly “red” of that of HCS(hPl), but we cannot readily assign this to any particular chromophore type, nor does it indicate what sort of structural difference might be involved. A more-concrete feeling for the degree of hyperchromicity and red-shifting which the conformation of a globular protein can produce may be obtained by comparing the native spectra with spectra taken after exhaustive digestion with a proteolytic enzyme such as thermolysin (19, 20, 22). As shown in Fig. 1, both digested proteins (pH 8.2) display a considerable loss in extinction (hypochromicity), and a general blue-shifting of the entire spectrum toward lower wavelengths. The relatively large alterations in the hGH spectrum between 310 and 290 nm are most noteworthy. It is also of interest that, in order to completely remove all conformational effects on the absorption spectrum, it is necessary to titrate both digests to pH 1.5. This is particularly true of the hGH digest, which still retains small conformational effects at pH 8.2. Secmd-Order

HORMONE

221

CONFORMATIONS

A.

d2Abr. 230

220

270

280

290

303

310

320

-a-

Spectra

Although the hypo- and hyperchromic shifts can be readily evaluated from zeroorder spectra in terms of changes in EM, the exact degree of red- or blue-shifting of the many weak shoulders is not at all clear. This problem can be largely solved by employing second-order absorption spectroscopy in which the second derivative of absorption (dZAbs/dX’) is plotted against wavelength. In this mode, the shoulders appearing in Fig. 1 will all become negative maxima. These second-order peaks indicate the wavelength positions of the absorption band centers in the various shoulders to about fO.l nm. Figure 2 presents the second-order absorption spectra of native hGH and HCS(hPL). The weak shoulders in the zeroorder spectra between 270 and 245 nm appear as six sharp, negative peaks in the second-order spectra. The corrected wavelength positions of these maxima in both hGH and HCS(hPL) are listed in Table I. By comparison with second-order spectra

FIG. 2. Second-order absorption spectra of native hGH at pH 8.2 (p), native HCS(hPL) at pH 8.2 (- - -), and the 24-h thermolysin digest of either protein after titration to pH 1.5 (* * *). The pH 1.5 digests of hGH and HCS(hPL) provide essentially equivalent spectra. In (A) the entire spectra of the native proteins are shown along with a partial spectrum of the digests. (B) Displays an expanded view of the two native molecules in the region of Trp absorption. The wavelength positions of the Trp ‘L. absorption bands have been calculated from the known (25) spacing of the vibronic modes after assigning the positions of the [O-O] bands. These calculated positions compare very favorably with the positions of actual second-order peaks (Table I).

of model compounds (Table I), all six of these absorption bands may be assigned to the Phe-‘Lb [O-O]transition and five major, higher vibrational modes (17, 18, 23). It appears that the average Phe residue in HCS(hPL) may absorb at a slightly higher

BEWLEY

AND

TABLE SECOND-ORDER

SPECTRAL

Native 304.2 (s) 301.0 298.2 290.9 283.2 276.2 -273 (vws) 268.3 264.5 261.5 (w) 258.6 252.9 247.8

Trp Trp Trp Trp Tyr Tyr Trp Phe Phe Phe Phe Phe Phe

IL.* ‘L. ‘L, ‘Lb* ‘Lb* ‘Lb ‘Lb ‘Lb ‘Lb IL,, ‘Lb ‘Lb ‘Lb

I

BANDS OF NATIVE

hGH

LI

hGH AND HCS(hPLpb

AND DIGESTED

HCS(hPL) Digest

Native

287.8* 281.5* 274.5 -272 (ws) 267.1* 263.2 260.3 257.3 251.3 246.4

298.1* 295 (ws) 290.0* 283.4* 276.0 -273 (vws) 268.6* 264.7 261.4 (w) 258.9 253.2 247.9

-

Model compounds

Trp Trp Trp Tyr Tyr Trp Phe Phe Phe Phe Phe Phe

IL.* IL.* ‘Lb* ‘Lb ‘Lb ‘Lb ‘Lb* ‘Lb ‘Lb ‘Lb ‘Lb ‘Lb

Digest

Trp”

Tyr”

Phe”

-

-

-

-

287.7* 281.1

281.9* 274.8 266.7 -

267.3* 263.5 260.0 (w) 257.4 251.7 246.7

288.0 281.6 274.9 -272 (ws) 267.1 263.5 260.8 (w) 257.5 246.7

271.6

-

-

-

’ Corrected as described elsewhere (19). Native at pH 8.2; digests and models at pH 1.5. *The * denotes the [O-O] band of each chromophoric electron. ’ N-Acetyl-L-Trp-amide, N-acetyl-L-Tyr-amide, and N-acetyl-L-Phe-ethyl ester. w, weak peak; s, shoulder; ws, weak shoulder; vws, very weak shoulder.

wavelength than in HGH. This would indicate slightly more hydrophobic (nonpolar) environment for the average HCS(hPL) Phe residue (23). Following digestion and titration to pH 1.5, all six Phe absorption bands, in both proteins, blue-shift to wavelength positions equivalent to the model compound (Table I). This proves that the digestion and titration have successfully removed all conformational effects from the Phe absorption (normalization). The data in Table I further indicate that the Tyr and Trp absorptions are also essentially normalized in the titrated digests. As discussed elsewhere (19), this point is crucial to the validity of the EM determinations reported above. By further reference to model compound spectra (17, 18, 23), the absorption bands arising from the Tyr-‘Lb [O-O] transition and one of its higher vibrational modes may also be identified in the second-order spectra of the native hormones. The Tyr bands in hGH appear at essentially the same wavelengths as those in HCS(hPL) (Table II), suggesting a similar nonpolarity for the average environment of these chromophores.

In this region of the spectrum, two different electrons in the indole ring of Trp may simultaneously produce both ‘L, and ‘Lb [O-O] transitions along with their higher vibronic modes (24,25). The ‘Lb [O-O] transition is easily identified (17-19, 24, 25) as the relatively intense second-order peak at 290.0 nm in HCS(hPL), and 290.9 nm in hGH (Table I). This nearly l.O-nm difference is highly significant and readily visible in the second-order spectra (Fig. 2B). In TABLE AVERAGE

Protein

II

RED-SHIFTING OF CHROMOPHORES NATIVE hGH AND HCS(hPL)

Chromophore transition

IN

Red shift (nm)

hGH

Trp Trp Tyr Phe

‘L, ‘Lb ‘Lb ‘Lb

12.2 2.9 1.3 1.0

HCS(hPL)

Trp Trp Tyr Phe

‘L, ‘Lb ‘Lb iLb

6.1 2.0 1.5 1.3

COMPARISON

OF GROWTH

HORMONE

CONFORMATIONS

223

the absence of additional information, it [O-O] position in Hz0 is taken to be 292.0 would seem that the microenvironment of nm (25). These average red-shift values are listed in Table II for both hormones. The the hGH Trp residue is more hydrophobic than that of HCS(hPL). However, as ex- shifts shown for Tyr, Phe, and Trp in are typical of many folded, plained below, this is probably not the case. HCS(hPL) globular proteins (22, 24, 25), and are due In the absorption spectra of most Trp to these residues being buried (on the avmodel compounds dissolved in HzO, the ‘L, erage) within the nonpolar interiors of [O-O] transition is not visible due to overthese molecules. The same is true of the lapping with the more powerful ‘Lb band. Phe and Tyr red-shifts in hGH. In contrast, Addition of nonpolar solvents, or reagents the enormous, 12.2-nm red shift in the Trpwhich form hydrogen bonds with the indole ‘L, [O-O] absorption of hGH is not typical. ring >NH group, usually produce unequal It cannot be explained as being due to an red-shifting of the ‘L, and ‘Lb bands unusually hydrophobic environment (25). (24,25). The ‘L, bands red-shift much more Trp-‘L, red-shifts of this strongly, pulling them out from under the Fortunately, more-intense ‘Lb bands. The presence of magnitude, although rare, are not unabsorption and CD bands above -293 nm known, As reported by Strickland et al. (25), in the spectra of many native, unconjuTrp-‘L, red-shifts of this magnitude (1213 nm) can occur when the indole ring >NH gated, globular proteins is due to this phegroup is simultaneously surrounded by a nomenon (23, 24). Similarly, it is the lack of conformation which makes it impossible and is hydrogen nonpolar environment, to identify the Trp-‘L, [O-O] transition in bonded to a negatively charged carboxylate the spectra of either the model compound ion. Hydrogen bonding to other groups or the titrated digests (Table I). such as protonated carboxyls, amide carIn Fig. 1, the very broad, weak shoulders bonyl, and amino groups, or imidazole side appearing above 300 nm in the zero-order chains, even in nonpolar solvents, will prospectra of native hGH and HCS(hPL) can duce red shifts of only 8-9 nm (25). Hydrogen bonding of the indole ring >NH arise only from weak ‘L, absorption bands (24, 25). It is impossible to give a detailed group to a carboxylate ion also very condescription of this absorption from the veniently explains the extra red shift in the hGH Trp-‘L,, absorption, beyond that zero-order spectra. However, the secondorder spectra (Fig. 2 and Table I) depict it produced by the nonpolar environment very clearly. In HCS(hPL), the Trp-‘L, seen in HCS(hPL) (25). Thus, a Trp-carboxylate hydrogen[O-O] band lies at 298.1 nm, with one higher bonded complex, buried in a hydrophobic vibronic mode being visible as a shoulder near 295 nm. The second higher vibronic environment, may be the specific structural mode, which should appear at 291.1 nm detail which gives rise to the long known (calculated as indicated elsewhere (24) differences in the uv absorption and CD properties of the Trp residues in hGH and from the position of the [O-O] band), is still hidden by the powerful Trp-‘Lb [O-O] ab- HCS(hPL). sorption at 290 nm. In native hGH, the Trp-‘L, bands are so strongly red-shifted Diflerence Spectra of that the [O-O] transition and both higher Thermolysin Digests vibronic modes are visible well above the Trp-‘Lb absorption. Figure 3A presents a series of increasThe amount of red shift existing in both ingly intense difference absorption spectra Trp transitions, as well as in the Tyr and generated during thermolysin digestion of Phe transitions, can be obtained by simply hGH. The negative peaks and shoulders subtracting the wavelength positions of the arise from blue-shifts in all major chrovarious [O-O] bands of appropriate model mophores (the negative Phe blue-shift compounds dissolved in H,O from the peaks are superimposed on the positive wavelengths of the corresponding [O-O] limb of the more-intense Trp and Tyr bluebands in the native proteins. The Trp-‘L, shifts below 273.5 nm). These blue-shifts

BEWLEY

224

AND

LI

are equivalent to those previously noted from second-order spectra, and again indicate exposure of previously buried Trp, Tyr, and Phe residues (22,28). The shoulder above 300 nm and the peak at 292.5 nm (AEM = -3900 M-’ cm-‘) can be assigned to the single Trp residue (22,28). The peak at 285.6 nm (AEM = -4900 M-’ cm-‘), and the shoulder near 279 nm are mostly due to Tyr, with some contribution from Trp (22,28). The unusual red shift in the native Trp-‘L, absorption, caused by the proposed

o

nAbs.

I

!50

260

270

I

I

I

280

290

300

WAVELENGTH

250

260

270

280

290

xx)

310

AAbs

320

0. +

310

(nm)

FIG. 4. Continuation, in time, of the thermolysin digestion of HCS(hPL) shown in Fig. 3B. The times of the various spectra (dashed lines) are, in order of increasing intensity below 295 nm, 285, 405, 540, and 1440 min. The solid line (-) shows the final difference spectrum taken after titration from pH 8.2 to pH 1.5. Protein concentration, E/S, and ordinate scale are as in Fig. 3B.

0

250

260

270

280

290

WAVELENGTH

300

310

320

(nm)

FIG. 3. Difference absorption spectra taken during thermolysin digestion of hGH (A), and the early portion of the HCS(hPL) digest (B). In (A), the times of the various spectra (dashed lines) are, in order of increasing intensity, 10, 45, 135, 480, and 1440 min. The solid line (-) shows the final spectrum taken after titration of the digest from pH 8.2 to 1.5. In (A) the protein concentration was 2.62 X lo-’ M; the E/ S = l/50 (w/w). In (B) the times of the various spectra (dashed lines) are, in order of increasing intensity, 20,45,80, and 130 min. The solid line (pH 8.2), taken after 205 min, is the last spectra to show the isosbestic point at 270 nm, and an increase in the red-shift (positive) peak near 304 nm. The protein concentration was 2.77 x 1O-5 M; E/S = l/50 (w/w).

Trp-carboxylate hydrogen bond, is clearly reflected in the width and relative prominence of the blue-shift shoulder found above 300 nm in the fully denatured form (compared with fully denatured HCS(hPL) in Fig. 4). The appearance of a sharp isosbestic point at 273.5 nm is also noteworthy. At this unique wavelength, the absorption of hGH is independent of the conformational state of the molecule. The EM at this wavelength (EM = 19,600 M-l cm-‘) can be used4to accurately determine the concentration of hGH solutions in which the protein may be only partially folded. Similar isosbestic points (although

EM EM

4 This isosbestic is valid only for purely aqueous solutions (pH 1.5 to 8.2). Addition of denaturants or nonpolar components, or pH values above 8.2, have not yet been tested, but can be expected to alter both the wavelength and the amplitude of the

COMPARISON

OF GROWTH

at different wavelengths) have been reported for five species of pituitary prolactin (20, 29, 30). The AE, values for the blueshift peaks in hGH are also significantly greater than those reported for prolactins (20,29,30). Most of this difference is again due to the Trp-carboxylate hydrogen bond in hGH, which does not occur in the native prolactins studied to date. Figure 3B presents difference spectra generated during the early stages of thermolysin digestion of HCS(hPL). Although normal blue-shifting of Tyr and Phe absorptions is evident, the Trp absorption unexpectedly shows the increasing emergence of a moderately intense red-shift (positive) peak near 304 nm. This can arise only from red-shifts in the Trp-‘L, bands.5 The Trp-‘Lb [O-O]band near 290 nm appears as an abnormally narrow and fairly weak blue-shift shoulder. Continued digestion of HCS(hPL) eventually shows a gradual loss of the red-shift band, with the difference spectra ultimately attaining a more normal, all blue-shift appearance (Fig. 4). The final difference spectrum (pH 1.5 digest) displays a Trp-‘L, blue-shift shoulder just above 300 nm, a Trp-‘Lb blue-shift peak at 291.0 nm (AEEn = 3300 M-’ cm-‘), and blueshifts due mostly to Tyr at 285.2 nm (AE, = 3900 M-’ cm-‘) and near 278 nm. Again, the Phe blue-shifts are superimposed on the positive Trp-Tyr limb below 270 nm. The AE, values for the two peaks near 291 and 285 nm are considerably less than for hGH, with the Trp-‘L, shoulder above 300 nm being a much less prominent feature of the overall final difference spectrum. This is a direct reflection of the greater EMvalue for native hGH, as shown in Fig. 1. During the early stages of the HCS(hPL) digest (Fig. 3B), a distinct isosbestic point appears near 270 nm (EM = 16,000M-l cm-‘). However, during later stages (Fig. 4) where the Trp-‘L, red5 The possibility that this band is caused by partial ionization of a Tyr, buried in the native protein in a fully protonated state, is eliminated by the observation that, in a digestion carried out at pH 7.40, a pH at which an emerging Tyr will remain essentially unionized (23), the entire difference spectral pattern is identical to that at pH 8.2.

HORMONE

CONFORMATIONS

225

shift peak is disappearing, the isosbestic point shifts to nearer 274 nm (EM = 18,150) as seen for hGH. The HCS(hPL) difference spectra strongly suggest that partial thermolysin digestion produces a conformational change resulting in red-shifting of the Trp-‘L, absorption to the region 301-304 nm, equivalent (or nearly so) to that of native hGH. It is tempting to suggest that the first few proteolytic nicks loosen up the local microenvironment around the HCS(hPL) Trp enough to allow formation of a Trp-carboxylate hydrogen bond analogous to the one proposed to exist in native hGH. More extensive digestion eventually produces sufficient general conformational collapse to disrupt that feature, along with the continued overall exposure of all chromophoric groups. Second-Order Spectra during Thermolysin Digestion Figure 5 presents partial second-order spectra taken during thermolysin digestion of highly concentrated samples of hGH and HCS(hPL). High concentrations (~2 mg/ml) were used to optimize the signalto-noise ratio in the spectral region of the Trp-‘L, absorption. Therefore, the kinetics do not correspond to those of Figs. 3 and 4, which were done at lower concentrations. In hGH (Fig. 5A) there is a rapid and extensive blue-shifting of the Trp-‘L, bands between 304 and 298 nm. These spectra are fully consistent with the difference spectra of Fig. 3A, and arise from extensive blueshifting of the hGH Trp-‘L, absorption to wavelengths near and below 292 nm (25) due to breaking of the Trp-carboxylate hydrogen bond, and exposure of the Trp residue to the external solvent. We have not yet been able to separate these two different blue-shift processes kinetically. For HCS(hPL), the situation is significantly different. The major Trp-‘L, band near 298 nm appears to lose intensity, but is transiently replaced by a new secondorder minimum near 301 nm (see spectrum at 30 min in Fig. 5B). The transient appearance of this second-order band is almost certainly related to the Trp-IL, red-

BEWLEY

0.04

OD2

0

-0.02 dZ Abr. -s0.24

AND

LI

a mixture is technically very difficult. However, we feel that the second-order spectra in Fig. 5B, when combined with the difference spectra of Fig. 3B, do strongly suggest the presence of a transient Trp-carboxylate hydrogen-bonded complex buried within the interior of the partially digested HCS(hPL) molecule. Previous partial denaturations of HCS(hPL), carried out by acid titration, and also followed by difference spectra, do not present evidence of a similar red-shift in Trp absorption (13,14). At present, the source of the carboxylate group is completely unknown, as is the role of this structural detail in the biochemical functioning of the hGH and HCS(hPL) molecules. REFERENCES

0.12

0

-0.12

1. LYONS, W. R., LI, C. H., AND JOHNSON, R. E. (1960) Acta EndocrinoL SuppL 51,1145. 2. CHADWICK, A., FOLLEY, S. J., AND GEMZELL, C. A. (1961) Lancet 2, 241-243. 3. JOSIMOVICH, J. B., AND MACLAREN, J. S. (1962) Endocrinology 71, 209-220. 4. FLORINI, J. R., TONELLI, G., BREUER, C. B., COPPOLA, J., RINGLER, J., AND BELL, P. H. (1966) End@

crinology 5. KAPLAN, -0.24

290

300 WAVELENGTH

310 hlrn)

FIG. 5. Second-order absorption spectra of (A) native hGH (p); and a thermolysin digest (pH 8.2) at 15 (- -). 45 (- - -), and 220 ( * * e). Protein concentration was 8 X 10e5 M; E/S = l/50 (w/w). In (B) HCS(hPL) is shown as native protein (-), and a thermolysin digest at 30 (- -), 60 (- - -), and 220 min (0. . ). Protein concentration was 1 X lo-’ M; E/S = l/50 (w/w).

shift difference spectra of Fig. 3B, and most probably represents the intense, first vibronic mode of the Trp-‘L, absorption of a buried Trp-carboxylate hydrogen bond as proposed for hGH. Unfortunately, during its early stages, the digest is necessarily a very heterogeneous mixture of undigested, partially digested, and more fully digested forms of the protein. Resolution of the Trp-lL, second-order bands in such

79, 692-708.

S. L., AND GRUMBACH,

M. M. (1964)

J.

Clin EndocrinoL Metab. 24, 80-100. 6. LI, C. H. (1970) Ann Sclavo Inst. Siena 12, 651662. 7. LI, C. H. (1972) Proc. Amer. Phil. Sot. 116, 365382. 8. LI, C. H., DIXON, J. S., AND CHUNG, D. (1973) Arch. B&hem. Biophys. 155,95-110. 9. BEWLEY, T. A., DIXON, J. S., AND LI, C. H. (1972) Ink J. Pept. Protein Res. 4, 281-287. 10. NIALL,H. D., HOGAN, M. L., SAUER, R., ROSENBLUM, I. Y., AND GREENWOOD, F. C. (1971) Proc. NatL

Acad Sti

USA 68, 866-869.

11. BEWLEY, T. A. (1977) in Hormonal Proteins and Peptides (Li, C. H., ed.), Vol. 4, pp. 62-95, Academic Press, New York. 12. SHOME, B., AND FRIESEN, H. G. (1971) Endocrinology 89.631-641. 13. BEWLEY, T. A., KAWAUCHI, J., AND LI, C. H. (1972) Biochemistry 11,4179-4187. 14. ALOJ, S., AND EDELHOCH, J. (1971) J. BioL Chem 246, 5047-5052. 15. ALOJ, S., AND EDELHOCH, H. (1972) J. BioL Chem. 247, 1146-1152. 16. RUSSELL, J., SHERWOOD, J. M., KOWALSKI, K., AND SCHNEIDER, A. B. (1981) J. BioL Chem 256,296300. 17. BALESTRIERI, C., COLONNA, G., GIOVANE, A., IRACE,

COMPARISON

18.

19. 20. 21. 22. 23.

OF GROWTH

G., AND SERVILLO, L. (1978) Eur. J. B&hem. so, 433-440. BALESTRIERI, C., COLONNA, G., GIOVANE, A., IRACE, G., AND SERVILLO, L. (1980) Anal Biochem. 106, 49-54. BEWLEY, T. A. (1982) Anal Biochem. 123. 55-65. BEWLEY, T. A., COLOSI, P., AND TALAMANTES, F. (1982) Biochemistry 21, 4238-4243. BEWLEY, T. A., AND LI, C. H. (1967) Biochim Bicphys. Acta 140,201-207. BEWLEY, T. A., AND LI, C. H. (1971) Arch. Biochem Biophys. 144, 589-595. WETLAUFER, D. B. (1962) Adv. Protein. Chem. 17, 303-390.

HORMONE

CONFORMATIONS

227

24. STRICKLAND, E. H. (1974) Rev. B&hem 2, 113175. 25. STRICKLAND, E. H., BILLUPS, C., AND KAY, E. (1972) Biochemistrg 11,3657-3662. 26. LI, C. H., Lru, W. K., AND DIXON, J. S. (1962) Arch. B&hem. Biqphys. Suppl 51,327-332. 27. NERI, P., TARLI, P., AND COCOLA, F. (1972) Ital. J. B&hem. 19, 111-11’7. 28. HERSKOVITS, T. T., AND SORENSEN, M. (1968) Bie chemistry 7, 2533-2545. 29. BEWLEY, T. A., COLOSI, P., AND TALAMANTES, F. (1983) Arch. Biochem. Biqphys. 225, 482-489. 30. BEWLEY, T. A., AND LI, C. H. (1983) Arch B&hem Biophys. 227, 618-625.