Changes in the ratio between serum and “specific” levels of human chorionic gonadotropin in different trimesters of pregnancy B.
DATTATREYAMURTY,
A.
R.
SHETH,
L.
R.
JOSHI,
SHANTA Bombay,
M.Sc. PH.D. M.Sc.
S.
RAO,
PH.D.
India
Sixty-one sera from different trimesters of pregnancy were analyzed by two ‘25I-NIH-HCG assay systems, employing anti-intact HCG serum and anti P-HCG serum, respectively. The ratios of the levels measured in the two assay systems changed with the duration of pregnancy. The ratios during trimesters 1, 2, and 3 were 2.94, I .99, and 2.37, respectively. The cross-reactivity of proteohormones other than HCG was tested in both the assay systems. The two assay systems could be comparable in their high degree of specificity. However, the relative afinities of intact HCG and F-HCG in the two assay systems were observed to be different. It was suggested that the significant differences in the ratios of the levels measured by the two assay systems might have been influenced by the occurrence of P-HCG in serum and that the levels of the subunit must have changed at different stages of pregnancy.
Materials
THE EXISTENCE of two subunits in human chorionic gonadotropin (HCG) was first reported by Bahll and Bell and associates.2 Since then, several worker9 6 have shown that glycoprotein hormones consist of two nonidentical, non-covalently bound subunits. It is now well known that the (Y subunit of HCG is common to other proteohormones while ,L?-HCG is hormone specific. Specific antisera can be prepared against the /3 subunit and such antisera prove to be valuable in carrying out specific radioimmunoassays.7 In the present study, maternal sera were analyzed by two radioimmunoassay systems to find out if there are changes in the ratio between HCG and specific HCG levels in different trimesters of pregnancy. From the Institute (ICMR). Received Revised Accepted Reprint Research Bombay
for Research
for publication June
December
in Reproduction
methods
Results
18, 1973.
Homologous intact HCG radioimmunoassay. Fig.
19, 1974.
1 shows the dose-response lines of HCG, and (Y-HCG. cy-HCG did nificantly up to a dose of 250 ng. finities of HCG and P-WCC were
June 28, 1974. requests: Dr. A. R. Sheth, in Reproduction (ICMR), 400 012, India.
and
Radioimmunoassays were performed by means of the classical double-antibody method. HCG and specific HCG levels in the maternal sera from different stages of pregnancy were determined with the following assay systems, respectively: lZ51-HCG + anti-HCG serum system and ?I-HCG + anti-pHCG system. Assay procedure. The samples were initially incubated with antiserum at 4O C. for 48 hours. Following this period of time, a second antibody (sheep anti-rabbit gamma globulin) was added and incubation was continued for a minimum of 48 hours at 4O C. before separation of bound and free hormone by centrifugation. The reference preparation used for each radioimmunoassay was IR-HCG supplied by WHO.
Institute Par&,
for
300
intact HCG, pnot inhibit sigThe relative afquite different.
Volume Number
121 3
Changes
in serum
and
“specific”
NIH-HCG*+Anti
&
HCG
levels
in pregnancy
301
HCG SERUM
60-
1. 2 P E
4020-
10 L “l------
--llCm--;.117-I’-&
----
SERUK SAMPI E FWM 2 Ll Q
0,“““““““““““““““‘1”
0.1
0.3
0.5
0.7
09
1.1 -
612Sng)
(*25ng)
Mng)
(lng)
NIH-hCG*+
I I.0 Ina
(Sng)
(2ng)
Fig. 1. Inhibition curves of intact of the serum sample from normal
I , 0.6 0.7 0.5 ng
1.3 1.5 +? 1.9 LOG DOSE HCG, (u-HCG and woman. Percentage
AS TO
f I.3
2.3
2.5
2.7 60ng)
(long)
(20ng)
/3-HCG. of TPC
x - - - - X, The
=
(Bx/Bo)
+ob,,,
\1(Oh.
2.9
3.1
3.3
(1OOng)
inhibition x 100.
3.5
( 2bOng) curve
hCG
1 1.7
I CO lOlIp
2w LOsGw DOSE
Fig. 2. Inhibition system. Percentage
2.1
NORMA,
curves of hTSH and HCG of TPC = (Bx/Bo) x 100.
P-HCG required 16 times more mass for a 50 per cent inhibition comparable to the homologous antigen, intact HCG. The inhibition curves of hTSH (human thyroid-stimulating hormone) and HCG were compared and the results indicated that hTSH required approximately 23 times more mass for a 50 per cent inhibition as compared to intact HCG (Fig. 2). Table I shows the slopes of the regression line generated for intact HCG, /3-HCG, and hTSH. hFSH (human follicle-stimulating hormone) did not inhibit labeled HCG significantly even up to a dose of 800 m1.U. (Fig. 3). The dose-response lines of hLH and HCG plotted on logarithmic logistic ruling can be seen in Fig. 4. The immunoreactivity
1 2.5 tong
I 2.7 SOng
I 1 34 250 ng
1 30 lOOfIg
-ng
in a homologous
Table
intact
HCG
radioimmunoassay
I. Slopes of regression Antigen
Slope*
HCG 8-hCG hTSH *From
probit
lines
-1.34 -0.67 -0.44 regression
lines.
patterns of hLH (human luteinizing hormone) and HCG were different. The slopes (calculated according to the method described by Hunterg) for hLH and HCG were, respectively, 1.8 and 5.3 per cent. This discrimination became more obvious as the concentration of antigen (hLH) increased, since the direction of the dose-response line changed sharply with a much lower slope value, -0.65 (slope
302
Dattatreyamurty
February 1, 1975 Am. J. Obstet. Gynecol.
et al.
Nltl
hCG*
+
AS
hCG
hFSH
P 0 a
I
1 0.0
I I.0
I 2
I 4
III 88lo
I 26
ANTIQEN
Fig. 3. Inhibition system. Percentage
curves of hFSH and of TPC = (Bx/Bo)
I 80
I 100
I 200
’
I
I
III
2
4
6
6
intact
HCG
12
I6
20
radioimmunoassay
hCG*+
I
I
I
40
60
100
II IO
I 800
- rn.1.u.
HCG in a homologous x 100.
NIH
IO
I 4w
AS hCG
I
200
I
300
m.1.u.
Fig. 4. Inhibition system. Percentage
curves of hLH and of TPC = (Bx/Bo)
HCG
value for HCG, -5.64). Here the slope represents the fall in the per cent bound per log dose of antigen.
NIH-HCG
in
a homologous
intact
HCG
radioimmunoassay
x 100.
+ anti p-HCG serum assay system.
The dose-response lines of intact HCG, &HCG, and (Y-HCG are illustrated in Fig. 5. c~-HCG failed to effectively inhibit 1251-HCG, since the dose-response line of (w-HCG started at a much higher dose level of antigen, 100 ng. Although the dose-response lines of &HCG and intact HCG occurred in the same portion of the dose-response graph, their relative affinities were observed to be different. Intact HCG required two times more mass for a 50 per cent inhibition than /3-HCG. From the regression
lines, the specificity index was calculated at different y values (90 to 10 per cent of total precipitable counts) and the mean value was observed to be 2.03 with a standard error kO.11. The slopes of the regression lines for P-HCG and intact HCG were -1.92 and -1.78, respectively. These observations suggested that ,&HCG displayed two times higher affinity than HCG throughout the effective range of the dose-response curve. The cross-reactivity of other proteohormones-hLH, hFSH, and hTSH-was also tested in this assay system. Our results, indicating antigenic dissimilarity between these hormones and HCG, suggest that this assay system can be used for measur-
Volume Number
121 3
Changes
in serum
and
“specltic”
NM- HCG*+ Antip-HCG
(425ng)
(-25ng)
(*brig)
(Ing)
LOG
DOSE
Fig.
5. Same
levels
in pregnancy
303
SERUM
(long)
c5ng)
(2ng)
HCG
as given
in Fig.
(2Ong)
(song)
doonp)
(250ng)
1.
p -HCG*+ Anti fi-HCG
SERUM
oC-HCG ./*-.--*-*---.A*
(.125ng)
f*25ng)
(-5ng)
(lng)
LOG
DOSE
(2ng)
Fig. 6. Inhibition curves of intact HCG, immunoassay system. Percentage of TPC
(Sng)
(long)
(Y-HCG and /3-HCG = (Bx/Bo) x 100.
ing specific HCG levels in sera and thus support the recommendation of the National Institute of Arthritis and Metabolic Diseases. Comparative evaluation of two assay systems. In “NIH-HCG + anti ,B-HCG serum” and homologus ,0-HCG systems (Figs. 5 and 6) /3-HCG displayed higher affinity than intact HCG, but the degree of cross-reactivity of intact HCG in the two assay systems was different. For a 50 per cent inhibition comparable to P-HCG, intact HCG required 4.3 times more mass, a value which is approximately two times greater than that observed in the assay system “NIH-HCG + anti ,&HCG serum.” Serum hormone levels determined by the two different assay systems. The results indicated in
(20ng)
(50n9)
in a homologous
m0ng)
/3-HCG
c25ong4
radio-
Fig. 7 show the mean levels in sera of different stages of pregnancy, determined by a homologous intact HCG system and “NIH-HCG + anti /3-HCG serum” system. The ratio between HCG and specific HCG levels in sera changed with the duration of pregnancy. During the first trimester the ratio was the highest (2.94 t 0.16) and it decreased in the second trimester ( 1.99 t 0.08). There was again an increase in the third trimester (2.37 + 0.09) but not to the extent observed in the first trimester. Statistical analysis of the results indicated that these differences are highly significant (Table II). To ensure that no false-positive HCG values were encountered by either assay system, a few serum samples from normally cycling women were also analyzed. The results (Figs. 1 and 4) indicated that
304
Dattatreyamurty
February 1, 1975 Am. J. Obstet. Gynecol.
et al.
12t-
0
NM- HCG*+ Anti HCG Serum
IO -
b
NIH-HCG*+AntipHCG
Ii
Serum
9-
1 rt
TRIMESTER
Fig. 7. Serum HCG mean values.
Table II. The ratio between levels in different
Trimester of Pregnancy
First Second Third
trimesters
No. of observations
and
2nd “specific”
HCG
TRWESTER levels
HCG and specific HCG of pregnancy* Ratio between levels measured by homologous intact HCG system/NIHHCG + anti-P-HCG serum system (f S.E.)
18 21 22
2.94 2 0.16 1.99 -L 0.08 2.37 -c 0.09
*Significant levels: Between trimesters 1 and 2, P = < 0.001. Between trimesters 1 and 3, P = 0.001 < P < 0.01. Between trimesters 2 and 3, P = 0.001 < P < 0.01.
no significant inhibition was given up to the volume of 200 ~1.
by the samples
Comment Earlier studies to determine HCG levels in serum by a wide array of assay systems have revealed that HCG levels reach a peak at 60 to 80 days of gestation, after which there is a sharp decline to low values. There was a controversy, however, regarding the occurrence of the second peak in the last trimester of pregnancy. In most studies a small but rather consistent secondary rise in the HCG level was observed in the last trimester.g-lZ On the other hand, Albert and Berkson,13 Loraine,14 and Mishell and associatesl’ did not observe the occurrence of a second peak. In the present study, sera from different stages of pregnancy were analyzed by a homologous intact HCG radioimmunoassay. When
in
3rd
TRIMESTER
normal pregnancy. Vertical
. MONTHS OF GESTATION columns
represent
the mean levels during the different trimesters of pregnancy were compared, the HCG level during the first trimester was observed to be the highest. The value for the last trimester was higher than that of midpregnancy, but not to the extent of the first trimester. A comparative evaluation of the results obtained in “homologous intact HCG” and “NH-HCG + anti /3-HCG serum” systems indicated that the ratio of the levels measured in the two assay systems changed with the duration of pregnancy. During the first trimester the ratio was the highest (2.94 f 0.16) and it decreased in the second trimester (1.99 + 0.08). There was again an increase in the third trimester (2.37 t 0.09) but not to the extent observed in the first trimester. It is unlikely that the significant differences in the ratios are influenced by proteohormones other than HCG. This is because the cross-reactivity of hFSH in homologous intact HCG system was not significant and hTSH required much larger amounts to inhibit labeled HCG. The affinity and slopes of hLH were considerably different from those of HCG, indicating incomplete cross-reactivity between the two hormones. This discrimination between hLH and HCG probably reflected the antigenic dissimilarity between the two hormones. Furthermore, in our study, hLH had a sharp break in its dose-response line. This observation, according to Vaitukaitis and ROSS,’ indicates “interaction with a subset of lower af-
Vo!ume Number
121 3
Changes
finity antibodies which were readily saturated with relatively small amounts of antigen.” FranchimontlG has characterized an antiserum raised to intact HCG and demonstrated incomplete cross-reactivity between hLH and HCG. Further, in our study, screening of serum samples from normal women revealed that a sample volume up to 200 ~1 did not significantly inhibit labeled HCG. These findings do manifest a high degree of specificity of the homologous intact HCG system, and thus it is comparable to the assay system “NIH-HCG + anti ,0-HCG serum,” but the two assay systems could be distinguished on the basis of the immunoreactivity patterns of intact HCG and P-HCG. In the homologous intact HCG system, intact HCG and /?-HCG gave dose-response lines with different slopes and affinities. The intact hormone displayed a higher affinity while ,0-HCG showed a weak cross-reaction, since this subunit required 16 times more mass for a 50 per cent inhibition comparable to intact HCG. On the other hand, in the assay system “NIH-HCG i- anti-/3-HCG serum,” /3-HCG exhibited a small but higher affinity than intact HCG. In other words, ‘““I-NIH-HCG system using anti-,&HCG serum, when employed for measuring the HCG specific levels in scra, would preferentially measure /3-HCG
1045,
and
1971.
Swaminathan, N., and Bahl, 0. P.: Biochem. Biophys. Res. Commun. 40: 422, 1970. 6. Saxena, B. B., and Rathnam, P.: J. Biol. Chem. 246: 3549, 1971. 7. Vaitukaitis, J. L., and Ross, G. T.: In Saxena, B. B., Beling, C. G., and Gandy, H. M., editors: Gonadotropins, International Symposium on Gonadotropins. New York, 1971. 8. Hunter, W. M.: In Weir, D. M., editor: Handbook of Experimental Immunology, ed. 2, Oxford, 1973, Blackwell Scientific Publications, p. 17.
“specific”
HCG
levels
in pregnancy
305
if present in serum and also intact HCG due to a cross-reaction between the antibody and the hormone specific moiety of intact HCG. Further, there is no risk of interference by a-HCG in both the assay systems used in the present study, since this subunit requires very large amounts of antigen to inhibit labeled HCG. Thus, in view of these findings, the significant differences in the ratios of levels reported in the present study could be explained by the occurrence of /3-HCG in serum and the fact that the levels of the subunit must have changed at different stages of pregnancy. Vaitukaitis” has recently presented evidence to indicate that serum, during pregnancy, may contain 8-HCG in addition to intact HCG. The hLH, hTSH, HCG, P-HCG, and antiserum to P-HCG used for carrying out radioimmunoassays were made available by the National Institute of Arthritis and Metabolic Diseases, National Institutes of Health. The a-HCG was the generous gift of Dr. Canfield, College of Physicians and Surgeons, Columbia University, New York, New York. Antiserum to intact HCG was obtained from rabbits hyperimmunized with a preparation of HCG supplied by Aycrst Laboratories, New York, New York; the antiserum was suitably purified and used for the HCG radioimmunoassay.
9. Behrman,
REFERENCES
1. Bahl, 0. P.: In Butt, W. R., Crooke, A. C., and Ryle, M., editors: Gonadotropins. A workshop conference, Birmingham, England, 1969. 2. Bell, J. J., Canfield, R. E., and Sciarra, J. J.: Endocrinology 84: 298, 1969. 3. Pierce, J. G.: Endocrinology 89: 1331, 1971. 4. Morgan, F. J., and Canfield, R. E.: Endocrinology 88:
in serum
236,
S. J., and Niemann,
P.: Fertil.
Steril. 6:
1955.
10.
Brody, S., and Carlstrom, G.: In Wolstenholme, G. E. W., and Cameron, M. P., editors: Immunoassay of hormones, Ciba Found. Colloq. Endocrinol.
11.
Faiman, C., Ryan, R. J., and Zwirek, Endocrinol. 28: 1323, 1968.
12.
Varma, K., Larraga, L., and Salenkow, H. A.: Obstet. Gynecol. 37: 10, 1971. Albert, A., and Berkson, J.: J. Clin. Endocrinol. 11: 805, 19.51. Loraine, J. A.: In Wolstenholme, G. E. W., and Millar, E. C. P., editors: Hormones in Blood, Ciba Found. Colloq. Endocrinol. 11: 19, 1957. Mishell, D. R., Jr., Wide, L., and Gemzell, C. A.: J. Clin. Endocrinol. 23: 125, 1963. Franchimont. P.: Eur. J. Clin. Invest. 1: 65, 1970. Vaitukaitis, J. L.: In Program of Fifty-Fifth Annual Meeting, Endocrine Society, June, 1973, Chicago, 111.
14:
5.
13. 14.
15. 16. 17.
329,
1962.
S. J.: J. Clin.
DATTATREYAMURTY,
A.
R.
SHETH,
L.
R.
JOSHI,
SHANTA Bombay,
M.Sc. PH.D. M.Sc.
S.
RAO,
PH.D.
India
Sixty-one sera from different trimesters of pregnancy were analyzed by two ‘25I-NIH-HCG assay systems, employing anti-intact HCG serum and anti P-HCG serum, respectively. The ratios of the levels measured in the two assay systems changed with the duration of pregnancy. The ratios during trimesters 1, 2, and 3 were 2.94, I .99, and 2.37, respectively. The cross-reactivity of proteohormones other than HCG was tested in both the assay systems. The two assay systems could be comparable in their high degree of specificity. However, the relative afinities of intact HCG and F-HCG in the two assay systems were observed to be different. It was suggested that the significant differences in the ratios of the levels measured by the two assay systems might have been influenced by the occurrence of P-HCG in serum and that the levels of the subunit must have changed at different stages of pregnancy.
Materials
THE EXISTENCE of two subunits in human chorionic gonadotropin (HCG) was first reported by Bahll and Bell and associates.2 Since then, several worker9 6 have shown that glycoprotein hormones consist of two nonidentical, non-covalently bound subunits. It is now well known that the (Y subunit of HCG is common to other proteohormones while ,L?-HCG is hormone specific. Specific antisera can be prepared against the /3 subunit and such antisera prove to be valuable in carrying out specific radioimmunoassays.7 In the present study, maternal sera were analyzed by two radioimmunoassay systems to find out if there are changes in the ratio between HCG and specific HCG levels in different trimesters of pregnancy. From the Institute (ICMR). Received Revised Accepted Reprint Research Bombay
for Research
for publication June
December
in Reproduction
methods
Results
18, 1973.
Homologous intact HCG radioimmunoassay. Fig.
19, 1974.
1 shows the dose-response lines of HCG, and (Y-HCG. cy-HCG did nificantly up to a dose of 250 ng. finities of HCG and P-WCC were
June 28, 1974. requests: Dr. A. R. Sheth, in Reproduction (ICMR), 400 012, India.
and
Radioimmunoassays were performed by means of the classical double-antibody method. HCG and specific HCG levels in the maternal sera from different stages of pregnancy were determined with the following assay systems, respectively: lZ51-HCG + anti-HCG serum system and ?I-HCG + anti-pHCG system. Assay procedure. The samples were initially incubated with antiserum at 4O C. for 48 hours. Following this period of time, a second antibody (sheep anti-rabbit gamma globulin) was added and incubation was continued for a minimum of 48 hours at 4O C. before separation of bound and free hormone by centrifugation. The reference preparation used for each radioimmunoassay was IR-HCG supplied by WHO.
Institute Par&,
for
300
intact HCG, pnot inhibit sigThe relative afquite different.
Volume Number
121 3
Changes
in serum
and
“specific”
NIH-HCG*+Anti
&
HCG
levels
in pregnancy
301
HCG SERUM
60-
1. 2 P E
4020-
10 L “l------
--llCm--;.117-I’-&
----
SERUK SAMPI E FWM 2 Ll Q
0,“““““““““““““““‘1”
0.1
0.3
0.5
0.7
09
1.1 -
612Sng)
(*25ng)
Mng)
(lng)
NIH-hCG*+
I I.0 Ina
(Sng)
(2ng)
Fig. 1. Inhibition curves of intact of the serum sample from normal
I , 0.6 0.7 0.5 ng
1.3 1.5 +? 1.9 LOG DOSE HCG, (u-HCG and woman. Percentage
AS TO
f I.3
2.3
2.5
2.7 60ng)
(long)
(20ng)
/3-HCG. of TPC
x - - - - X, The
=
(Bx/Bo)
+ob,,,
\1(Oh.
2.9
3.1
3.3
(1OOng)
inhibition x 100.
3.5
( 2bOng) curve
hCG
1 1.7
I CO lOlIp
2w LOsGw DOSE
Fig. 2. Inhibition system. Percentage
2.1
NORMA,
curves of hTSH and HCG of TPC = (Bx/Bo) x 100.
P-HCG required 16 times more mass for a 50 per cent inhibition comparable to the homologous antigen, intact HCG. The inhibition curves of hTSH (human thyroid-stimulating hormone) and HCG were compared and the results indicated that hTSH required approximately 23 times more mass for a 50 per cent inhibition as compared to intact HCG (Fig. 2). Table I shows the slopes of the regression line generated for intact HCG, /3-HCG, and hTSH. hFSH (human follicle-stimulating hormone) did not inhibit labeled HCG significantly even up to a dose of 800 m1.U. (Fig. 3). The dose-response lines of hLH and HCG plotted on logarithmic logistic ruling can be seen in Fig. 4. The immunoreactivity
1 2.5 tong
I 2.7 SOng
I 1 34 250 ng
1 30 lOOfIg
-ng
in a homologous
Table
intact
HCG
radioimmunoassay
I. Slopes of regression Antigen
Slope*
HCG 8-hCG hTSH *From
probit
lines
-1.34 -0.67 -0.44 regression
lines.
patterns of hLH (human luteinizing hormone) and HCG were different. The slopes (calculated according to the method described by Hunterg) for hLH and HCG were, respectively, 1.8 and 5.3 per cent. This discrimination became more obvious as the concentration of antigen (hLH) increased, since the direction of the dose-response line changed sharply with a much lower slope value, -0.65 (slope
302
Dattatreyamurty
February 1, 1975 Am. J. Obstet. Gynecol.
et al.
Nltl
hCG*
+
AS
hCG
hFSH
P 0 a
I
1 0.0
I I.0
I 2
I 4
III 88lo
I 26
ANTIQEN
Fig. 3. Inhibition system. Percentage
curves of hFSH and of TPC = (Bx/Bo)
I 80
I 100
I 200
’
I
I
III
2
4
6
6
intact
HCG
12
I6
20
radioimmunoassay
hCG*+
I
I
I
40
60
100
II IO
I 800
- rn.1.u.
HCG in a homologous x 100.
NIH
IO
I 4w
AS hCG
I
200
I
300
m.1.u.
Fig. 4. Inhibition system. Percentage
curves of hLH and of TPC = (Bx/Bo)
HCG
value for HCG, -5.64). Here the slope represents the fall in the per cent bound per log dose of antigen.
NIH-HCG
in
a homologous
intact
HCG
radioimmunoassay
x 100.
+ anti p-HCG serum assay system.
The dose-response lines of intact HCG, &HCG, and (Y-HCG are illustrated in Fig. 5. c~-HCG failed to effectively inhibit 1251-HCG, since the dose-response line of (w-HCG started at a much higher dose level of antigen, 100 ng. Although the dose-response lines of &HCG and intact HCG occurred in the same portion of the dose-response graph, their relative affinities were observed to be different. Intact HCG required two times more mass for a 50 per cent inhibition than /3-HCG. From the regression
lines, the specificity index was calculated at different y values (90 to 10 per cent of total precipitable counts) and the mean value was observed to be 2.03 with a standard error kO.11. The slopes of the regression lines for P-HCG and intact HCG were -1.92 and -1.78, respectively. These observations suggested that ,&HCG displayed two times higher affinity than HCG throughout the effective range of the dose-response curve. The cross-reactivity of other proteohormones-hLH, hFSH, and hTSH-was also tested in this assay system. Our results, indicating antigenic dissimilarity between these hormones and HCG, suggest that this assay system can be used for measur-
Volume Number
121 3
Changes
in serum
and
“specltic”
NM- HCG*+ Antip-HCG
(425ng)
(-25ng)
(*brig)
(Ing)
LOG
DOSE
Fig.
5. Same
levels
in pregnancy
303
SERUM
(long)
c5ng)
(2ng)
HCG
as given
in Fig.
(2Ong)
(song)
doonp)
(250ng)
1.
p -HCG*+ Anti fi-HCG
SERUM
oC-HCG ./*-.--*-*---.A*
(.125ng)
f*25ng)
(-5ng)
(lng)
LOG
DOSE
(2ng)
Fig. 6. Inhibition curves of intact HCG, immunoassay system. Percentage of TPC
(Sng)
(long)
(Y-HCG and /3-HCG = (Bx/Bo) x 100.
ing specific HCG levels in sera and thus support the recommendation of the National Institute of Arthritis and Metabolic Diseases. Comparative evaluation of two assay systems. In “NIH-HCG + anti ,B-HCG serum” and homologus ,0-HCG systems (Figs. 5 and 6) /3-HCG displayed higher affinity than intact HCG, but the degree of cross-reactivity of intact HCG in the two assay systems was different. For a 50 per cent inhibition comparable to P-HCG, intact HCG required 4.3 times more mass, a value which is approximately two times greater than that observed in the assay system “NIH-HCG + anti ,&HCG serum.” Serum hormone levels determined by the two different assay systems. The results indicated in
(20ng)
(50n9)
in a homologous
m0ng)
/3-HCG
c25ong4
radio-
Fig. 7 show the mean levels in sera of different stages of pregnancy, determined by a homologous intact HCG system and “NIH-HCG + anti /3-HCG serum” system. The ratio between HCG and specific HCG levels in sera changed with the duration of pregnancy. During the first trimester the ratio was the highest (2.94 t 0.16) and it decreased in the second trimester ( 1.99 t 0.08). There was again an increase in the third trimester (2.37 + 0.09) but not to the extent observed in the first trimester. Statistical analysis of the results indicated that these differences are highly significant (Table II). To ensure that no false-positive HCG values were encountered by either assay system, a few serum samples from normally cycling women were also analyzed. The results (Figs. 1 and 4) indicated that
304
Dattatreyamurty
February 1, 1975 Am. J. Obstet. Gynecol.
et al.
12t-
0
NM- HCG*+ Anti HCG Serum
IO -
b
NIH-HCG*+AntipHCG
Ii
Serum
9-
1 rt
TRIMESTER
Fig. 7. Serum HCG mean values.
Table II. The ratio between levels in different
Trimester of Pregnancy
First Second Third
trimesters
No. of observations
and
2nd “specific”
HCG
TRWESTER levels
HCG and specific HCG of pregnancy* Ratio between levels measured by homologous intact HCG system/NIHHCG + anti-P-HCG serum system (f S.E.)
18 21 22
2.94 2 0.16 1.99 -L 0.08 2.37 -c 0.09
*Significant levels: Between trimesters 1 and 2, P = < 0.001. Between trimesters 1 and 3, P = 0.001 < P < 0.01. Between trimesters 2 and 3, P = 0.001 < P < 0.01.
no significant inhibition was given up to the volume of 200 ~1.
by the samples
Comment Earlier studies to determine HCG levels in serum by a wide array of assay systems have revealed that HCG levels reach a peak at 60 to 80 days of gestation, after which there is a sharp decline to low values. There was a controversy, however, regarding the occurrence of the second peak in the last trimester of pregnancy. In most studies a small but rather consistent secondary rise in the HCG level was observed in the last trimester.g-lZ On the other hand, Albert and Berkson,13 Loraine,14 and Mishell and associatesl’ did not observe the occurrence of a second peak. In the present study, sera from different stages of pregnancy were analyzed by a homologous intact HCG radioimmunoassay. When
in
3rd
TRIMESTER
normal pregnancy. Vertical
. MONTHS OF GESTATION columns
represent
the mean levels during the different trimesters of pregnancy were compared, the HCG level during the first trimester was observed to be the highest. The value for the last trimester was higher than that of midpregnancy, but not to the extent of the first trimester. A comparative evaluation of the results obtained in “homologous intact HCG” and “NH-HCG + anti /3-HCG serum” systems indicated that the ratio of the levels measured in the two assay systems changed with the duration of pregnancy. During the first trimester the ratio was the highest (2.94 f 0.16) and it decreased in the second trimester (1.99 + 0.08). There was again an increase in the third trimester (2.37 t 0.09) but not to the extent observed in the first trimester. It is unlikely that the significant differences in the ratios are influenced by proteohormones other than HCG. This is because the cross-reactivity of hFSH in homologous intact HCG system was not significant and hTSH required much larger amounts to inhibit labeled HCG. The affinity and slopes of hLH were considerably different from those of HCG, indicating incomplete cross-reactivity between the two hormones. This discrimination between hLH and HCG probably reflected the antigenic dissimilarity between the two hormones. Furthermore, in our study, hLH had a sharp break in its dose-response line. This observation, according to Vaitukaitis and ROSS,’ indicates “interaction with a subset of lower af-
Vo!ume Number
121 3
Changes
finity antibodies which were readily saturated with relatively small amounts of antigen.” FranchimontlG has characterized an antiserum raised to intact HCG and demonstrated incomplete cross-reactivity between hLH and HCG. Further, in our study, screening of serum samples from normal women revealed that a sample volume up to 200 ~1 did not significantly inhibit labeled HCG. These findings do manifest a high degree of specificity of the homologous intact HCG system, and thus it is comparable to the assay system “NIH-HCG + anti ,0-HCG serum,” but the two assay systems could be distinguished on the basis of the immunoreactivity patterns of intact HCG and P-HCG. In the homologous intact HCG system, intact HCG and /?-HCG gave dose-response lines with different slopes and affinities. The intact hormone displayed a higher affinity while ,0-HCG showed a weak cross-reaction, since this subunit required 16 times more mass for a 50 per cent inhibition comparable to intact HCG. On the other hand, in the assay system “NIH-HCG i- anti-/3-HCG serum,” /3-HCG exhibited a small but higher affinity than intact HCG. In other words, ‘““I-NIH-HCG system using anti-,&HCG serum, when employed for measuring the HCG specific levels in scra, would preferentially measure /3-HCG
1045,
and
1971.
Swaminathan, N., and Bahl, 0. P.: Biochem. Biophys. Res. Commun. 40: 422, 1970. 6. Saxena, B. B., and Rathnam, P.: J. Biol. Chem. 246: 3549, 1971. 7. Vaitukaitis, J. L., and Ross, G. T.: In Saxena, B. B., Beling, C. G., and Gandy, H. M., editors: Gonadotropins, International Symposium on Gonadotropins. New York, 1971. 8. Hunter, W. M.: In Weir, D. M., editor: Handbook of Experimental Immunology, ed. 2, Oxford, 1973, Blackwell Scientific Publications, p. 17.
“specific”
HCG
levels
in pregnancy
305
if present in serum and also intact HCG due to a cross-reaction between the antibody and the hormone specific moiety of intact HCG. Further, there is no risk of interference by a-HCG in both the assay systems used in the present study, since this subunit requires very large amounts of antigen to inhibit labeled HCG. Thus, in view of these findings, the significant differences in the ratios of levels reported in the present study could be explained by the occurrence of /3-HCG in serum and the fact that the levels of the subunit must have changed at different stages of pregnancy. Vaitukaitis” has recently presented evidence to indicate that serum, during pregnancy, may contain 8-HCG in addition to intact HCG. The hLH, hTSH, HCG, P-HCG, and antiserum to P-HCG used for carrying out radioimmunoassays were made available by the National Institute of Arthritis and Metabolic Diseases, National Institutes of Health. The a-HCG was the generous gift of Dr. Canfield, College of Physicians and Surgeons, Columbia University, New York, New York. Antiserum to intact HCG was obtained from rabbits hyperimmunized with a preparation of HCG supplied by Aycrst Laboratories, New York, New York; the antiserum was suitably purified and used for the HCG radioimmunoassay.
9. Behrman,
REFERENCES
1. Bahl, 0. P.: In Butt, W. R., Crooke, A. C., and Ryle, M., editors: Gonadotropins. A workshop conference, Birmingham, England, 1969. 2. Bell, J. J., Canfield, R. E., and Sciarra, J. J.: Endocrinology 84: 298, 1969. 3. Pierce, J. G.: Endocrinology 89: 1331, 1971. 4. Morgan, F. J., and Canfield, R. E.: Endocrinology 88:
in serum
236,
S. J., and Niemann,
P.: Fertil.
Steril. 6:
1955.
10.
Brody, S., and Carlstrom, G.: In Wolstenholme, G. E. W., and Cameron, M. P., editors: Immunoassay of hormones, Ciba Found. Colloq. Endocrinol.
11.
Faiman, C., Ryan, R. J., and Zwirek, Endocrinol. 28: 1323, 1968.
12.
Varma, K., Larraga, L., and Salenkow, H. A.: Obstet. Gynecol. 37: 10, 1971. Albert, A., and Berkson, J.: J. Clin. Endocrinol. 11: 805, 19.51. Loraine, J. A.: In Wolstenholme, G. E. W., and Millar, E. C. P., editors: Hormones in Blood, Ciba Found. Colloq. Endocrinol. 11: 19, 1957. Mishell, D. R., Jr., Wide, L., and Gemzell, C. A.: J. Clin. Endocrinol. 23: 125, 1963. Franchimont. P.: Eur. J. Clin. Invest. 1: 65, 1970. Vaitukaitis, J. L.: In Program of Fifty-Fifth Annual Meeting, Endocrine Society, June, 1973, Chicago, 111.
14:
5.
13. 14.
15. 16. 17.
329,
1962.
S. J.: J. Clin.
