Abnormal immunoglobulin subclasspatterns in women with a history ofrecurrent miscarriage

Rhoda Wilson, Ph.D., Marjorie A. Maclean, M.D., Carol Jenkins, M.Sc.,Dennis Kinnane, M.D., John Mooney, M.Sc., and James J. Walker, M.D.

University Department of Medicine, Glasgow Royal Infirmary, and Department of Obstetrics and Gynaecology,St. James Hospital, Leeds, United Kingdom

Objective: To determine whether IgG subclass patterns differed between nonpregnant women, healthypregnant women, and pregnant women with a history of recurrent miscarriage.

Design: Controlled clinical study.

Setting: An academic setting.

Patient(s): Group 1 was comprised of 10 nonpregnant women, group 2 of 10 healthy pregnant women, group3 of eight pregnant women with a history of recurrent miscarriage and whose pregnancies on this occasionwent to term, and group 4 of 10 women with a history of recurrent miscarriage whose pregnancies again failedlater in the first trimester.

Intervention(s): None of the patients received any medication.

Main Outcome Measure(s): Serum levels of total IgG and IgG 1, 2, 3, and 4.

Result(s): The results obtained showed that normal pregnancy was associated with a significant increase intotal IgG production and an increase in IgG subclasses 1, 2, and 3. Women with a history of miscarriage, but whohad a successful pregnancy on this occasion, showed a similar pattern of IgG subclasses. Women with a history ofmiscarriage and whose pregnancy again ended in miscarriage showed a different IgG subclass pattern.

Conclusion(s): Pregnancies that ended in miscarriage showed a different pattern of IgG subclasses than thosethat continued to term. The changes seen in immunoglobulin patterns could be linked to changes in cytokineproduction. (Fertil Steril� 2001;76:915–7. ©2001 by American Society for Reproductive Medicine.)

Key Words: Miscarriage, IgG subclasses, cytokines

Miscarriage affects approximately 1% of allpregnancies. While many of these women willnot have any further problems, a number willhave repeated miscarriages (recurrent miscar-riage). Although the causes of recurrentmiscarriage are not fully understood, immuno-logical mechanisms are thought to play an im-portant role. The effector T cell response isdivided into two subsets: the TH1 cells, whichproduce interferon (IFN) and IL-12, and TH2cells, which produce IL-4, 5, and 13 (1). Mu-rine and human studies have shown that TH2type cytokines are associated with a successfulpregnancy, while TH1 type cytokines are asso-ciated with miscarriage (2, 3). It is thought thatthis cytokine production emanates from activa-tion of the endometrial lymphocytes (4, 5).

Cytokines exert many different effects and

act on many different cell types. They havebeen shown to act on immunoglobulin subclassproduction (6). IL-4 causes human B cells toswitch to IgG 4 production, while IFN� inhib-its this process (7). IL-10 has been shown toincrease total immunoglobulin production (8).While changes in immunoglobulin subclasseshave been reported in normal pregnancy (9),little is known about changes that occur inwomen with a history of recurrent miscarriage.

We now report our findings of IgG subclasspatterns in normal pregnancy and recurrentmiscarriage.

MATERIALS AND METHODS

Patient GroupsThe ethical committee approved the study,

and all women gave informed consent. There

Received March 12, 2001;revised and accepted May30, 2001.Reprint requests: RhodaWilson, Ph.D., Departmentof Medicine, GlasgowRoyal Infirmary, 10Alexandra Parade,Glasgow G31 2ER,Scotland, United Kingdom(FAX: 0141-211-0414;E-mail:gcl025@clinmed.gla.ac.uk).

FERTILITY AND STERILITY�VOL. 76, NO. 5, NOVEMBER 2001Copyright ©2001 American Society for Reproductive MedicinePublished by Elsevier Science Inc.Printed on acid-free paper in U.S.A.

0015-0282/01/$20.00PII S0015-0282(01)02857-6

915

were four patient groups enrolled into the study: Group 1was comprised of 10 healthy nonpregnant women. Group 2was comprised of 10 first trimester (8.1 � 1.3 weeks gesta-tion) pregnant women. All pregnancies went to term. Group3 was comprised of eight first trimester pregnant women(7.6 � 1.5 weeks gestation) with a history of recurrentmiscarriage (at least three previous miscarriages) whosepregnancies on this occasion went to term. Group 4 wascomprised of 10 first trimester pregnant women (7.1 � 1.1weeks gestation) with a history of recurrent miscarriage (atleast 3 previous miscarriages) whose pregnancies againfailed later in the first trimester.

MethodsSerum samples were collected from the patients and

stored at �20°C. Titres of IgG subclasses were measured byELISA (10). Wells of the ELISA plate were coated with amonoclonal antihuman IgG 1, 2, 3, or 4 (Sigma, Poole,Dorset). This was followed by incubation with a standardhuman serum, the IgG 4 subclass concentration of which hadbeen determined by nephelometry (Department of ClinicalImmunological Western Infirmary, Glasgow). After block-ing of the unreacted sites in the wells with 0.1% BSA inphosphate-buffered saline (PBS), the wells were washedwith PBS containing 0.05% Tween 20. Five hundred micro-liters of patient serum were added to the wells and incubatedfor 90 minutes at 37°C, after which any unbound materialwas washed out. One hundred microliters of biotin-antihuman IgG (Sigma) was added and incubated for 1 hourat 37°C. After washing, the reaction is viewed with 100 �Lof tetramethylbenzidine substrate for 8 minutes at roomtemperature and stopped with 50 �L of 2.5% sodium fluo-ride. Plates were read at 630 and 490 nm reference usingan automated plate reader MR 500 (Dynex Laboratory,Middlesex, UK). IgG levels were calculated by reference tolinear regression analysis of the standard serum line.

Statistical MethodsResults are given as mean � 1SD and were analyzed for

statistical significance using a Mann-Whitney analysis.

RESULTS

The results (Table 1) show that compared with nonpreg-nant women, women having a normal pregnancy had asignificant increase in total IgG production. IgG subclasses1, 2, and 3 were also significantly increased in early preg-nancy. Women in group 3, who had a past history of recur-rent miscarriage, but whose pregnancies on this occasioncontinued to term, showed IgG subclass patterns that did notdiffer significantly from those in group 2. Women in group4, who had a history of recurrent miscarriage and whosepregnancies on this occasion again failed, showed signifi-cantly reduced levels of total IgG and IgG subclasses 1, 2, 3,and 4 compared with women in group 2. IgG 1 levels werefound to be significantly higher in group 3 compared withgroup 4 patients, and there was no overlap between the twogroups (Fig. 1). It is worth noting that all patients werepregnant at the time of sampling, although all women ingroup 4 miscarried later in the first trimester.

DISCUSSION

This study has shown that normal pregnancy is associatedwith significant changes in IgG subclasses. Pregnant womenwho miscarried later in the first trimester showed differentIgG subclass patterns.

Pregnancy has been associated with the production ofTH2 type cytokines (IL-10 and IL- 4), while miscarriage hasbeen associated with the production of TH1 type cytokines(IFN �, IL-12) (2, 3). IL-4 has been found to cause humanperipheral blood cells to become activated and switch to IgG4 production (7). IFN� inhibits this process (5). IL-10 hasbeen show in vitro to increase total IgG production (8).

T A B L E 1

Immunoglobulin levels in pregnant and nonpregnant women.

Group no. Immunoglobulin levels (mg/mL)

IgG 1 P IgG 2 P IgG 3 P IgG 4 P Total IgG P

1 1.7 � 0.9 1.4 � 0.9 0.4 � 0.2 0.6 � 0.2 4.0 � 22 3.1 � 0.8a �.03a 5.7 � 3.7a �.001a 0.8 � 0.8 0.5 � 0.1 10.4 � 4.1 �.02a

3 3.9 � 1.1 5.4 � 0.5 0.7 � 0.1 0.3 � 0.2 10.6 � 0.44 2.4 � 0.4bc �.002b 3.2 � 0.9b �.04b 0.5 � 0.1b �.004b 0.1 � 0.1b �.04b 6.8 � 4.2

�.001c

Note: Group 1 � nonpregnant; group 2 � pregnant; group 3 � miscarriage, successful; group 4 � miscarriage not successful. Results are � 1SD.a Group 2 vs. group 1.b Group 4 vs. group 2.c Group 4 vs. group 3.

Wilson. IgG subclasses in pregnancy and miscarriage. Fertil Steril 2001.

916 Wilson et al. IgG subclasses in pregnancy and miscarriage Vol. 76, No. 5, November 2001

Levels of IL-10 have been found to be elevated in serumfrom women with ongoing pregnancies (9). The high circu-lating levels of IL-10 may cause the observed increase intotal IgG production in the groups 2 and 3. Unsuccessfulpregnancies are associated with reduced levels of IL-10 (11),and in these women levels of total IgG were significantlyreduced.

IgG 4 levels were significantly increased in ongoing preg-nancies, while miscarriage was associated with significantlyreduced levels of IgG 4. This is presumably because levels ofIL-4 are known to be reduced and levels of IFN� are in-creased in these women (2, 3). Elevated levels of IFN� areknown to inhibit IgG 4 production.

Pregnancy is known to cause changes within the immunesystem, which are not observed in women who have recur-rent miscarriage. This study has shown that pregnancy is alsoassociated with changes in the production of IgG subclassesand these changes can be linked to the immunological eventsthat occur. Women who were pregnant at the time of sam-pling but who miscarried later in the first trimester showeddifferent IgG subset patterns compared with healthy preg-nant women. This would suggest that in these women,changes leading to miscarriage occur at an early stage in thepregnancy.

References1. Wegnman TG, Lin H, Guilbert L, Mossman TR. Bidirectional cytokine

interaction in maternal fetal relationship: is successful pregnancy a TH1phenomenon? Immunol Today 1993;11:353–6.

2. Jenkins C, Roberts, Wilson R, MacLean MA, Shilto J, Walker JJ.Evidence of a TH1 type response being associated with recurrentmiscarriage. Fertil Steril 2000;73:1206–8.

3. Lim KJ, Odukoya OA, Ajjan RA, Li TC, Weetman AP, Cooke ID. Therole of T-helper cytokines in human reproduction. Fertil Steril 2000;73:136–42.

4. Johnson PM, Christmas SE, Vince GS. Immunological aspects ofimplantation and implantation failure. Hum Reprod 1999;14:26–36.

5. Kitaya K, Yasuda J, Yagi I, Fushika S, Honjo H. IL-15 expression athuman endometrium and decidua. Biol Hum Reprod 2000;63:683–7.

6. Milburn HJ, Potter LW, Dilmec A, Cochrane GM, Kennedy DM.Corticosteroid restores the balance between locally produced TH1 andTH2 cytokines and immunological isotypes in sarcoid lung. Clin ExpImmunol 1997;108:105–13.

7. Del Prete G, Maggi E, Parronchi P, Chretien I, Tiri A, Macchia D, et al.IL-4 is an essential factor for the IgE synthesis induced in vitro byhuman T cell clones and their supernatants. J Immunol 1988;140:4193–8.

8. Lorente L, Zou W, Levy Y, Richaud-Patin Y, Wijdenes J, Alcocer-Varela J, et al. Role of interleukin 10. I. The B lymphocyte hyperac-tivity and autoantibody production of human systemic lupus erythem-atosus. J Exp Med 1995;181:839–44.

9. Yasuhara M, Tamaki H, Lyama S, Yamaguchi Y, Tachi J, Amino N.Reciprocal changes in serum levels of immunoglobulins (IgG, IgA,IgM) and complements (C3, C4) in normal pregnancy and after deliv-ery. J Clin Lab Immunol 1992;38:137–41.

10. Kinnane DF, Mooney J, MacFarlane TW, McDonald M. Local andsystemic antibody response to putative periodontitis: correlation withclinical indices. Oral Microbiol 1993;8:65–8.

11. Karhukorpi J, Laitinen T, Karttunenr R, Tiilikainen AS. The function-ally important IL-10 promotor polymorphism (-1082G A) is not a majorgenetic regulator in Recurrent spontaneous abortion. Mol Hum Reprod2001;7:201–3.

F I G U R E 1

IgG 1 levels in pregnant women with a history of recurrentmiscarriage.

Wilson. IgG subclasses in pregnancy and miscarriage. Fertil Steril 2001.

FERTILITY & STERILITY� 917