- Email: [email protected]
Amphiphilic growth hormone releasing factor (GRF) analogs: Peptide design and biological activity in vivo
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Vol. 139, No. 2, 1986
Pages 763-770
September 16, 1986
AMPHIPHILIC GROWTH HORMONE RELEASING FACTOR (GRF) ANALOGS: PEPTIDE DESIGN AND BIOLOGICAL ACTIVITY IN VIVO
Jacob S. Tou, Larry A. Kaempfe, Billy D. Vineyard, Frances C. Buonomo, Mary Anne Della-Fera and Clifton A. Baile
Animal Sciences Division, Monsanto Company, St. Louis, Missouri 63198
Received July 18, 1986
SUMMARY. The first twenty-nine amino acids of human Growth Hormone Releasing Factor (hGRF) possess a distinct amphiphilic character. This is seen as twisted hydrophobic and hydrophilic bands in the helical net projection. Four amidated analogs were designed by optimizing amphiphilic and helical potentials of the native sequence. These designed analogs, with up to eight-amino acid changes, were tested in sheep via intravenous injection. The growth hormone-stimulating activities of the analogs were significantly higher when compared to bovine Growth Hormone Releasing Factor (bGRF44-NIIo). This suggests that the amphiphilic conformation of GRF(I-29) i~ important to the receptor. ®i98~AcademicPr....Inc.
Human Growth Hormone Releasing Factor (hGRF), a peptide hormone, was independently isolated and sequenced by Guillemin (la) and by Vale (Ib) in 1982 as 44 and 40 residues, respectively.
Since it has potential
applications in human and veterinary medicines (2), much effort has been directed to the designing of potent and efficient analogs as well as to fundamental physiological studies (3). For example, potent analogs were reported using incorporation of unnatural amino acid(s) (4a,b), Cterminus deletions (4c,d), etc..
In our laboratory, we were involved in
the identification of potent GRF analogs via the modification of its amphiphilic secondary structure. The amphiphilic approach to peptide design was first reported by Kaiser et.al, in the areas of melittin (5) and calcitonin (6).
These
peptide hormones acting at the membrane have characteristic amphiphilic secondary structure (7).
One face of the molecule has preferentially
763
0006-291X/86 $I.50 Copyrght © 1986 ~ Aca~mic Press, Inc. A# t-~h~ of r~roduct~n in any jorm reserve~
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Vol. 139, No. 2, 1986
10
20
29
YADAIFTNSYRKVLGQLSARKLLQDIMSR
Figure 1 . The Garnier Secondary Structure Prediction of hGRF(I-29) region. ~ potential a-helix, ~ potential ~-turn.
hydrophobic residues, whereas the hydrophilic domain is on the opposite side.
For such peptides one can possibly enhance the biological
activities beyond that of the native peptides by optimizing the amphiphilic character.
The GRF(I-29) amidated sequence was used as the basis
of our study not only because it retains most of the in vitro growth hormone (GH) stimulating activity (4c), but also because of its unique secondary structure.
Analysis of the secondary structure calculation
(8a) implies that GRF(1-29) contains a B-turn in residues 8-12 followed by a pronounced a-helical potential in the area of 13-29 (Figure I).
A
net projection (9) of the region 13-29 indicates distinct twisted amphiphilic bands along the helical cylinder as shown in Figure 2.
Further-
more, assuming that the whole 1-29 segment forms an a-helical conformation, which may be induced by the membrane amphiphilic environment, the amphiphilicity as mentioned above is highly conserved as depicted in the net projection (Figure 3).
Indeed, according to Kaiser et.al.
(7), the
Tyr
lie
/
/
/
// jAJa
/
®
® Figure 2.
Net projection of hGRF(13-29) in the a-hellcal conformation.
Figure 3.
Net projection of hGRF(I-29) in the a-helical conformation.
764
Vol. 139, No. 2, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
amphiphilic potential of hGRF is strongly evidenced by the observations in the binding of single bilayer phospholipid vesicles and the formation of a monomolecule layer at the air-water interface.
Here we report:
I. The design and synthesis of potent GRF analogs by optimizing its structural amphiphilicity. 2. The GH stimulating activity observed in sheep via intravenous (IV) injection.
MATERIALS AND METHODS Peptide analogs were prepared by the solid phase peptidetsynthesis technlque(10). The N -amlno function was protected with the butyloxycarbonyl (BOC) group. Side chain functional groups were protected as follows: benzyl for Ser,Thr and Asp; p-toluenesulfonyl for Arg; 2,6dichlorobenzyl for Tyr; 2-chlorobenzyloxycarbonyl for Lys; Xanthyl for Asn and Gln. All except BocNLe were purchased from either Peptides International or Peninsula Laboratories. BocNLe and p-methylbenzyhydrylamine resin (0.41 meq N/g) were obtained from United States Biochemical Corporation. After HF cleavage, the crude peptide was purified on a low-pressure column (4 X 25 cm) packed with Vydac CI8silica (300A pore size, 15-20 p particle size). The solvents used for elution were: A, 5% acetonitrile-water with 0.05% trifluoroacetic acid (TFA); B, 50% acetonitrile-water with 0.05% TFA. The crude peptide was eluted at ca 2 ml/min with a power gradient (FMI pump). The purity of the resulting peptide was determined by analytical HPLC and amino acid analysis. [NLe-27]hGRF(I-29)-NH 2 was a gift from Dr. J. Rivier of Salk Research Institute. Circular Dichroism (CD) spectra were obtained on the JASCO 500 CD spectrometer and water was used as solvent (I mg peptide per 5 ml solvent). Peptides were tested in sheep weighing between 27 and 45 kg. The treatments were randomly assigned according to a 4 X 4 Latin square design, to four animals. Peptide was administered intravenously and blood samples were drawn via jugular catheter at -15, 0, 5, 15, 30, 45, 60, 90 and 120 minutes. Blood was placed on ice in EDTA tubes and plasma was separated. Plasma was stored at -20°C until assayed for ovine OH concentration by homologous RIA using reagents supplied by the NIADDK. The injection dosage for all the treatments was 0.065 nmole/kg body weight, bGRF44-NH 2 (one residue different from hGRF in the first 29 residues, prepared by Dr. Kam Fok, Monsanto Company) was used as the positive control. In two studies, 0.9% sterile saline was used as the basal adjustment. The activity represented as percent (%) of positive control is calculated as I00 times the mean (+5 to 120 min) GH concentration snbstracting the basal (-15 to 0 minute) GH concentration divided by that of positive control. The confidence limits (p <0.05) were determined by ANOVA and Duncan's multiple range tests. Treatment standard errors (Tse) are reported for each experiment.
RESULTS AND DISCUSSIONS [NLe-27]hGRF(I-29)-NH 2
~ was reported to be biologically equivalent
to hGRF40-NH 2 in the rat pituitary cell assay (4d).
765
A similar result was
Vol. 139, No. 2, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
also observed in our in vivo study using sheep ] .
To test the amphi-
philic modification hypothesis, four peptides with varying degree of substitution were designed, synthesized and tested in sheep.
All these
peptide analogs are amidated [NLe-27]-based 29-amino acid fragments. In the analysis of the helical net projection of residues 13 to 29, Gly-15 and Ser-18 do not fit well to the hydrophilic and hydrophobic domains.
Glutamine, a better hydrophilic amino acid, was chosen to
replace Gly at the 15 position, Ser-18 was replaced by either Ala (peptide 2) or Leu (peptide ~) to enhance the corresponding amphiphilicity Mean circulating GH levels were increased significantly (P<0.05) 4.05and 5.85-fold above baseline GH levels in sheep for two hours following IV injection of analogs 2 and 3 respectively.
Compared to the increase
in GH concentrations observed following bGRF44-N}{ 2 administration, response to analogs 2 and ~ were 48% and 52% greater (P<0.05).
the
The GH
profiles are shown in Figures 4a,b. On the other hand, according to the secondary structure calculation, the replacement of Set-9 by a good s-helix former (such as Ala) would favor an extended helical conformation 2.
Indeed, the incorporation
of Ala-9 would also enhance the hydrophobicity of the twisted hydrophobic band.
Along with the substitutions at positions 9, 15, 18 and 27, other
residues were also replaced, such as 2, 7, 13 and 28 of peptide ~, in order to optimize the amphiphilicity and helicity characters.
Mean
circulating GH levels were increased significantly (P<0.05) 3.05- and 2.84-fold above baseline GH levels in sheep for two hours following IV injection of GRF-analogs 4 and 5 respectively.
]
The mean elevation
o [NIe-27]hGRF29-NH o is about 90% as potent as hGRF40-NHo(but not significantly different). Meanwhile bGRF44-NH~ and hGRF40-NH7 did not differ significantly in their effectivenesN in causing OHsecretion (at 0.00g nmole/kg body weight dosage).
21t is interesting to note that when the native hGRF(I-29) sequence is displayed as a ~ helix conformation, if it exists, the hydrophobic and hydrophilic domains are well segregated on the opposite sides of the helical cylinder as best illustrated by the axial projection of an "Edmundson wheel" (Fig. 5).
766
Vol. 139, No. 2, 1 9 8 6
95"
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
8o
a
85.
70
75-
60
65.
b
50
~55 40 • Z 45 30 35 20
25
10
15
-
5, - ; -20
J
~
--
-
2'0 "
~o
-
',
-
7"
60
;
" ~ i ~" , , , 6:o loo ~2o
160-
0
"----~ _ II~l
=
. =
-
I 210 ' 410 I 60
' 80
- - ~
' 100'
~.
120 I 140'
. 160
180
110' d
C
lO0,
14o-
90 12o-
80 70
100 -
60
"oo] "o.,o ~80-
5O
40 30
10 . ~
-20
20
Fisnre
40 60 TIME (mktutes) 4.
Circulating
80
100
120
GH c o n c e n t r a t i o n s
010
-
I
;
following
I
210
I
dO' dO I TIME (minutes)
a ~-......-........_._~
810
GP~ a n a l o g
administration in sheep. Plasma GH levels were significantly (p<0.05) increased within five minutes after IV administration of bGRF44-NH_, a n a l o g s 2 ( a ) , 3 ( b ) , 4 ( c ) and 5 ( d ) i n s h e e p . The mean two hourZGH response was significantly (p
Figure 5.
N a t i v e hGRF(1-29)
i n an Edmundson w h e e l w i t h a H c o n f o r m a t i o n .
767
'
1;0
'
li0
Vol. 139, No. 2, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table I Fractional u-helix contents calculated from Circular Dichroism in water Peptides
% Helix Content
bGRF44-NII2
8
1
9
2
25 28
4
36 33
produced by these two peptides was 65% and 85% greater (P
Therefore, the
result of the substitutions should be peptides with better amphiphilicity and higher ~-helicity.
The enhanced helicities of the designed
analogs are indicated by the Circular Dichroism (CD) analysis (Table I). The relative GH stimulating potencies of the designed analogs vs. control peptides are summarized in Table II.
The analogs reported here,
with up to 8 amino acid modifications (see Figure 6), exhibit high
Table 2 Relative potencies of GRF(I-29) analogs to stimulate GH secretion by IV injection in sheep Peptides
Relative Potency
bGRF44-NI{2
100 148
3
152
4
165
5
185
768
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Vol. 139, No. 2, 1986
hGRF(I-29)
Y A D A I F T N S Y R K V L G Q L S A R K L L Q D I M
S R
Y A D A I F T N S Y R K V L G Q L S A R K L L Q D I Nle S R 2
Y A D A I F T N S Y R K V L_Q Q L A_ A R K L L Q D I Nle S R
3
Y A D A I F T N S Y R K V L Q Q L LA
4
Y S D A I F S N A Y R K I L Q Q L L A R K L L Q D I Nle Q R
R K L L Q D I Nle S R
Y A D A I F S N A Y R K I L Q Q L L A R K L L Q D I Nle Q R Figure
6.
Sequence
comparison
of n a t i v e
hGRF(I-29)
with
analogs
1 £o 5.
biological activities which suggest the importance of the amphiphilic conformation of GRF(I-29) to the receptor.
The results of this study
show that one could possibly, by further optimizing the amphiphilic character and/or increasing the stability towards proteolytic degradation, design GRF analogs with even higher levels of activity.
ACKNOWLEDGMENTS We thank Dr. J. Freeman for his Circular Dichroism analysis and Dr. C. Schasteen for his valuable discussions of secondary structures. We are also indebted to D. Hartzell, C. Atwell, M. Jennings and B. Walkenhorst for their excellent technical assistance.
REFERENCES 1.
a.
b. 2. a. b. 3. a.
4.
a.
b. C.
d.
5.
6.
Guillemin, R., Brazeau, P., Bohlem, P., Esch, F., Ling, N., and Wehrenberg, W. (1982) Science 218, 585-587. Rivier, J., Spiess, J., Thorner, M., and Vale, W. (1982) Nature 300, 276-278. Scanes, C., Carsie, R., Lauterio, T., Huybrechts, L., Rivier, J., and Vale, W. (1984) Life Science 34 No.12, 1127-1134. Coude, F., Diaz, J., Morre, M., Roskam, W., and Roneaucci, R. (1984) Trends in Biotechnology 2, No.4, 83-88. Guillemin, R., Zeytin, F., Ling, N. Bohlen, P., Esch, F., Brazeau, P. Bloch, B., and Wehrenberg, W. (1984) Proceedings of the Society for Experimental Biology and Medicine 175, 407-413. Coy, D., Murphy, W., Sueiras-Diaz, J., Coy, E., and Lance, V. (1985) J. Med. Chem. 28, 181-185. Lance, V., Murphy, W., Sueiras-Diaz, J., and Coy, D. (1984) Biochem. Biophys. Res. Commun. 119, 265-272. Ling, N., Baird, A., Wehrenberg, W., Uenu, N., Munegumi, T., and Brazeau, P. (1984) Biochem. Biophys. Res. Commun. 123, 854-861. Vale, W., Vanghan, J., Sawchenko, P., Seifer, H., Bilezikjian, L. Perrin, M., Swanson, L., Thurney, M., Spiess, J., and Rivier, J. Proc. of the Human Growth Hormone Symposium, Baltimore, Maryland, Nov. 20-22 (1983) (in Press). DeGrado, W., Kezdy, F., and Kaiser, E. (1981) J. Amer. Chem. Soc. 103, 679-681. Mue, G. Miller, R., and Kaiser, E. (1983) J. Amer. Chem. Soc. 105, 4100-4102. 769
Vol. 139, No. 2, 1986
7. 8,
a,
b. 9.
a.
b. 10.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Kaiser, E., and Kezdy, F. (]984) Science 223, 249-255. Garnier, J. Osguthorpe, D. J., and Robson, B. (1978) J. Mol. Biol. 120, 97-120. Fasman, G.D. (1980) Annals New York Academy of Sciences, 348, 147-159. Taylor, J. W., Miller, R. J., and Kaiser, E. T. (1983) J. Biol. Chem. 258, 4464-4471. The net projection is drawn as an open form of the helical cylinder. Steward, J. and Young, J. (1984) Solid Phase Peptide Synthesis, 2nd Edition, Pierce Chem. Co., Rockford, Illinois.
770
Vol. 139, No. 2, 1986
Pages 763-770
September 16, 1986
AMPHIPHILIC GROWTH HORMONE RELEASING FACTOR (GRF) ANALOGS: PEPTIDE DESIGN AND BIOLOGICAL ACTIVITY IN VIVO
Jacob S. Tou, Larry A. Kaempfe, Billy D. Vineyard, Frances C. Buonomo, Mary Anne Della-Fera and Clifton A. Baile
Animal Sciences Division, Monsanto Company, St. Louis, Missouri 63198
Received July 18, 1986
SUMMARY. The first twenty-nine amino acids of human Growth Hormone Releasing Factor (hGRF) possess a distinct amphiphilic character. This is seen as twisted hydrophobic and hydrophilic bands in the helical net projection. Four amidated analogs were designed by optimizing amphiphilic and helical potentials of the native sequence. These designed analogs, with up to eight-amino acid changes, were tested in sheep via intravenous injection. The growth hormone-stimulating activities of the analogs were significantly higher when compared to bovine Growth Hormone Releasing Factor (bGRF44-NIIo). This suggests that the amphiphilic conformation of GRF(I-29) i~ important to the receptor. ®i98~AcademicPr....Inc.
Human Growth Hormone Releasing Factor (hGRF), a peptide hormone, was independently isolated and sequenced by Guillemin (la) and by Vale (Ib) in 1982 as 44 and 40 residues, respectively.
Since it has potential
applications in human and veterinary medicines (2), much effort has been directed to the designing of potent and efficient analogs as well as to fundamental physiological studies (3). For example, potent analogs were reported using incorporation of unnatural amino acid(s) (4a,b), Cterminus deletions (4c,d), etc..
In our laboratory, we were involved in
the identification of potent GRF analogs via the modification of its amphiphilic secondary structure. The amphiphilic approach to peptide design was first reported by Kaiser et.al, in the areas of melittin (5) and calcitonin (6).
These
peptide hormones acting at the membrane have characteristic amphiphilic secondary structure (7).
One face of the molecule has preferentially
763
0006-291X/86 $I.50 Copyrght © 1986 ~ Aca~mic Press, Inc. A# t-~h~ of r~roduct~n in any jorm reserve~
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Vol. 139, No. 2, 1986
10
20
29
YADAIFTNSYRKVLGQLSARKLLQDIMSR
Figure 1 . The Garnier Secondary Structure Prediction of hGRF(I-29) region. ~ potential a-helix, ~ potential ~-turn.
hydrophobic residues, whereas the hydrophilic domain is on the opposite side.
For such peptides one can possibly enhance the biological
activities beyond that of the native peptides by optimizing the amphiphilic character.
The GRF(I-29) amidated sequence was used as the basis
of our study not only because it retains most of the in vitro growth hormone (GH) stimulating activity (4c), but also because of its unique secondary structure.
Analysis of the secondary structure calculation
(8a) implies that GRF(1-29) contains a B-turn in residues 8-12 followed by a pronounced a-helical potential in the area of 13-29 (Figure I).
A
net projection (9) of the region 13-29 indicates distinct twisted amphiphilic bands along the helical cylinder as shown in Figure 2.
Further-
more, assuming that the whole 1-29 segment forms an a-helical conformation, which may be induced by the membrane amphiphilic environment, the amphiphilicity as mentioned above is highly conserved as depicted in the net projection (Figure 3).
Indeed, according to Kaiser et.al.
(7), the
Tyr
lie
/
/
/
// jAJa
/
®
® Figure 2.
Net projection of hGRF(13-29) in the a-hellcal conformation.
Figure 3.
Net projection of hGRF(I-29) in the a-helical conformation.
764
Vol. 139, No. 2, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
amphiphilic potential of hGRF is strongly evidenced by the observations in the binding of single bilayer phospholipid vesicles and the formation of a monomolecule layer at the air-water interface.
Here we report:
I. The design and synthesis of potent GRF analogs by optimizing its structural amphiphilicity. 2. The GH stimulating activity observed in sheep via intravenous (IV) injection.
MATERIALS AND METHODS Peptide analogs were prepared by the solid phase peptidetsynthesis technlque(10). The N -amlno function was protected with the butyloxycarbonyl (BOC) group. Side chain functional groups were protected as follows: benzyl for Ser,Thr and Asp; p-toluenesulfonyl for Arg; 2,6dichlorobenzyl for Tyr; 2-chlorobenzyloxycarbonyl for Lys; Xanthyl for Asn and Gln. All except BocNLe were purchased from either Peptides International or Peninsula Laboratories. BocNLe and p-methylbenzyhydrylamine resin (0.41 meq N/g) were obtained from United States Biochemical Corporation. After HF cleavage, the crude peptide was purified on a low-pressure column (4 X 25 cm) packed with Vydac CI8silica (300A pore size, 15-20 p particle size). The solvents used for elution were: A, 5% acetonitrile-water with 0.05% trifluoroacetic acid (TFA); B, 50% acetonitrile-water with 0.05% TFA. The crude peptide was eluted at ca 2 ml/min with a power gradient (FMI pump). The purity of the resulting peptide was determined by analytical HPLC and amino acid analysis. [NLe-27]hGRF(I-29)-NH 2 was a gift from Dr. J. Rivier of Salk Research Institute. Circular Dichroism (CD) spectra were obtained on the JASCO 500 CD spectrometer and water was used as solvent (I mg peptide per 5 ml solvent). Peptides were tested in sheep weighing between 27 and 45 kg. The treatments were randomly assigned according to a 4 X 4 Latin square design, to four animals. Peptide was administered intravenously and blood samples were drawn via jugular catheter at -15, 0, 5, 15, 30, 45, 60, 90 and 120 minutes. Blood was placed on ice in EDTA tubes and plasma was separated. Plasma was stored at -20°C until assayed for ovine OH concentration by homologous RIA using reagents supplied by the NIADDK. The injection dosage for all the treatments was 0.065 nmole/kg body weight, bGRF44-NH 2 (one residue different from hGRF in the first 29 residues, prepared by Dr. Kam Fok, Monsanto Company) was used as the positive control. In two studies, 0.9% sterile saline was used as the basal adjustment. The activity represented as percent (%) of positive control is calculated as I00 times the mean (+5 to 120 min) GH concentration snbstracting the basal (-15 to 0 minute) GH concentration divided by that of positive control. The confidence limits (p <0.05) were determined by ANOVA and Duncan's multiple range tests. Treatment standard errors (Tse) are reported for each experiment.
RESULTS AND DISCUSSIONS [NLe-27]hGRF(I-29)-NH 2
~ was reported to be biologically equivalent
to hGRF40-NH 2 in the rat pituitary cell assay (4d).
765
A similar result was
Vol. 139, No. 2, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
also observed in our in vivo study using sheep ] .
To test the amphi-
philic modification hypothesis, four peptides with varying degree of substitution were designed, synthesized and tested in sheep.
All these
peptide analogs are amidated [NLe-27]-based 29-amino acid fragments. In the analysis of the helical net projection of residues 13 to 29, Gly-15 and Ser-18 do not fit well to the hydrophilic and hydrophobic domains.
Glutamine, a better hydrophilic amino acid, was chosen to
replace Gly at the 15 position, Ser-18 was replaced by either Ala (peptide 2) or Leu (peptide ~) to enhance the corresponding amphiphilicity Mean circulating GH levels were increased significantly (P<0.05) 4.05and 5.85-fold above baseline GH levels in sheep for two hours following IV injection of analogs 2 and 3 respectively.
Compared to the increase
in GH concentrations observed following bGRF44-N}{ 2 administration, response to analogs 2 and ~ were 48% and 52% greater (P<0.05).
the
The GH
profiles are shown in Figures 4a,b. On the other hand, according to the secondary structure calculation, the replacement of Set-9 by a good s-helix former (such as Ala) would favor an extended helical conformation 2.
Indeed, the incorporation
of Ala-9 would also enhance the hydrophobicity of the twisted hydrophobic band.
Along with the substitutions at positions 9, 15, 18 and 27, other
residues were also replaced, such as 2, 7, 13 and 28 of peptide ~, in order to optimize the amphiphilicity and helicity characters.
Mean
circulating GH levels were increased significantly (P<0.05) 3.05- and 2.84-fold above baseline GH levels in sheep for two hours following IV injection of GRF-analogs 4 and 5 respectively.
]
The mean elevation
o [NIe-27]hGRF29-NH o is about 90% as potent as hGRF40-NHo(but not significantly different). Meanwhile bGRF44-NH~ and hGRF40-NH7 did not differ significantly in their effectivenesN in causing OHsecretion (at 0.00g nmole/kg body weight dosage).
21t is interesting to note that when the native hGRF(I-29) sequence is displayed as a ~ helix conformation, if it exists, the hydrophobic and hydrophilic domains are well segregated on the opposite sides of the helical cylinder as best illustrated by the axial projection of an "Edmundson wheel" (Fig. 5).
766
Vol. 139, No. 2, 1 9 8 6
95"
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
8o
a
85.
70
75-
60
65.
b
50
~55 40 • Z 45 30 35 20
25
10
15
-
5, - ; -20
J
~
--
-
2'0 "
~o
-
',
-
7"
60
;
" ~ i ~" , , , 6:o loo ~2o
160-
0
"----~ _ II~l
=
. =
-
I 210 ' 410 I 60
' 80
- - ~
' 100'
~.
120 I 140'
. 160
180
110' d
C
lO0,
14o-
90 12o-
80 70
100 -
60
"oo] "o.,o ~80-
5O
40 30
10 . ~
-20
20
Fisnre
40 60 TIME (mktutes) 4.
Circulating
80
100
120
GH c o n c e n t r a t i o n s
010
-
I
;
following
I
210
I
dO' dO I TIME (minutes)
a ~-......-........_._~
810
GP~ a n a l o g
administration in sheep. Plasma GH levels were significantly (p<0.05) increased within five minutes after IV administration of bGRF44-NH_, a n a l o g s 2 ( a ) , 3 ( b ) , 4 ( c ) and 5 ( d ) i n s h e e p . The mean two hourZGH response was significantly (p
Figure 5.
N a t i v e hGRF(1-29)
i n an Edmundson w h e e l w i t h a H c o n f o r m a t i o n .
767
'
1;0
'
li0
Vol. 139, No. 2, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table I Fractional u-helix contents calculated from Circular Dichroism in water Peptides
% Helix Content
bGRF44-NII2
8
1
9
2
25 28
4
36 33
produced by these two peptides was 65% and 85% greater (P
Therefore, the
result of the substitutions should be peptides with better amphiphilicity and higher ~-helicity.
The enhanced helicities of the designed
analogs are indicated by the Circular Dichroism (CD) analysis (Table I). The relative GH stimulating potencies of the designed analogs vs. control peptides are summarized in Table II.
The analogs reported here,
with up to 8 amino acid modifications (see Figure 6), exhibit high
Table 2 Relative potencies of GRF(I-29) analogs to stimulate GH secretion by IV injection in sheep Peptides
Relative Potency
bGRF44-NI{2
100 148
3
152
4
165
5
185
768
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Vol. 139, No. 2, 1986
hGRF(I-29)
Y A D A I F T N S Y R K V L G Q L S A R K L L Q D I M
S R
Y A D A I F T N S Y R K V L G Q L S A R K L L Q D I Nle S R 2
Y A D A I F T N S Y R K V L_Q Q L A_ A R K L L Q D I Nle S R
3
Y A D A I F T N S Y R K V L Q Q L LA
4
Y S D A I F S N A Y R K I L Q Q L L A R K L L Q D I Nle Q R
R K L L Q D I Nle S R
Y A D A I F S N A Y R K I L Q Q L L A R K L L Q D I Nle Q R Figure
6.
Sequence
comparison
of n a t i v e
hGRF(I-29)
with
analogs
1 £o 5.
biological activities which suggest the importance of the amphiphilic conformation of GRF(I-29) to the receptor.
The results of this study
show that one could possibly, by further optimizing the amphiphilic character and/or increasing the stability towards proteolytic degradation, design GRF analogs with even higher levels of activity.
ACKNOWLEDGMENTS We thank Dr. J. Freeman for his Circular Dichroism analysis and Dr. C. Schasteen for his valuable discussions of secondary structures. We are also indebted to D. Hartzell, C. Atwell, M. Jennings and B. Walkenhorst for their excellent technical assistance.
REFERENCES 1.
a.
b. 2. a. b. 3. a.
4.
a.
b. C.
d.
5.
6.
Guillemin, R., Brazeau, P., Bohlem, P., Esch, F., Ling, N., and Wehrenberg, W. (1982) Science 218, 585-587. Rivier, J., Spiess, J., Thorner, M., and Vale, W. (1982) Nature 300, 276-278. Scanes, C., Carsie, R., Lauterio, T., Huybrechts, L., Rivier, J., and Vale, W. (1984) Life Science 34 No.12, 1127-1134. Coude, F., Diaz, J., Morre, M., Roskam, W., and Roneaucci, R. (1984) Trends in Biotechnology 2, No.4, 83-88. Guillemin, R., Zeytin, F., Ling, N. Bohlen, P., Esch, F., Brazeau, P. Bloch, B., and Wehrenberg, W. (1984) Proceedings of the Society for Experimental Biology and Medicine 175, 407-413. Coy, D., Murphy, W., Sueiras-Diaz, J., Coy, E., and Lance, V. (1985) J. Med. Chem. 28, 181-185. Lance, V., Murphy, W., Sueiras-Diaz, J., and Coy, D. (1984) Biochem. Biophys. Res. Commun. 119, 265-272. Ling, N., Baird, A., Wehrenberg, W., Uenu, N., Munegumi, T., and Brazeau, P. (1984) Biochem. Biophys. Res. Commun. 123, 854-861. Vale, W., Vanghan, J., Sawchenko, P., Seifer, H., Bilezikjian, L. Perrin, M., Swanson, L., Thurney, M., Spiess, J., and Rivier, J. Proc. of the Human Growth Hormone Symposium, Baltimore, Maryland, Nov. 20-22 (1983) (in Press). DeGrado, W., Kezdy, F., and Kaiser, E. (1981) J. Amer. Chem. Soc. 103, 679-681. Mue, G. Miller, R., and Kaiser, E. (1983) J. Amer. Chem. Soc. 105, 4100-4102. 769
Vol. 139, No. 2, 1986
7. 8,
a,
b. 9.
a.
b. 10.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Kaiser, E., and Kezdy, F. (]984) Science 223, 249-255. Garnier, J. Osguthorpe, D. J., and Robson, B. (1978) J. Mol. Biol. 120, 97-120. Fasman, G.D. (1980) Annals New York Academy of Sciences, 348, 147-159. Taylor, J. W., Miller, R. J., and Kaiser, E. T. (1983) J. Biol. Chem. 258, 4464-4471. The net projection is drawn as an open form of the helical cylinder. Steward, J. and Young, J. (1984) Solid Phase Peptide Synthesis, 2nd Edition, Pierce Chem. Co., Rockford, Illinois.
770