Influence of Maternal-Fetal Histocompatibility and MHC Zygosity on Maternal Microchimerism1
http://www.100md.com
免疫学杂志 2005年第11期
Abstract
To investigate the relationship between maternal-fetal histocompatibility and maternal microchimerism, we developed a sensitive quantitative PCR assay for the neomycin resistance gene (neoR), and, in a mouse model system, used neoR as a noninherited maternal allele marker of maternal cells to detect and quantitate maternal microchimerism in tissues of neoR–/– N2 backcross progeny of (neoR+/–)F1 females mated with neoR–/– males. Using this approach, we obtained evidence for the presence of chimeric maternal cells in the brain, spleen, and thymus of all weanling and adult mice so tested. The numbers of chimeric maternal cells present in the spleen did not differ significantly from those in the thymus regardless of age or maternal-fetal histocompatibility. At all ages, brain tissue had higher level of maternal microchimerism than lymphoid tissue in mice MHC identical with their mothers, but the levels were similar in mice MHC disparate with their mothers. The levels of chimeric maternal cells in both brain and lymphoid tissue of mice with homozygous syngenicity and maternal allogenicity were similar, and tended to be higher than tissue-specific levels in mice with either combined maternal-fetal allogenicity or heterozygous syngenicity. Thus, MHC homozygous progeny had higher levels of maternal microchimerism than MHC heterozygous progeny. We conclude that normal mice possess small numbers of maternal cells in spleen, thymus, brain, and probably most other tissues, and that maternal-fetal histocompatibility influences the levels of these cells by mechanisms related to MHC zygosity of the progeny.
Introduction
Maternal microchimerism, the presence of very small fractional concentrations of chimeric maternal cells, occurs in at least half of all humans from fetal life into adulthood (1, 2, 3, 4). The factors that govern maternal microchimerism are poorly understood, but may be of considerable clinical relevance in view of the fact that maternal microchimerism has been associated with autoimmunity (5) and allograft tolerance (6), and may serve as a vehicle for vertical transmission of pathogenic microbes and allergens (7).
The notion that maternal-fetal histocompatibility plays a role in the regulation of persistent maternal microchimerism derives from the well-known importance of donor-host histocompatibility in regulating donor cell chimerism in allogeneic bone marrow transplant recipients (8). Much of the evidence for this has come from studies that have taken advantage of the ready availability of inbred mouse strains with well-defined MHC genes. There have been several previously reported investigations of maternal microchimerism in mouse model systems that, taken together, demonstrate the frequent occurrence in fetal, neonatal, and adult lymphoid and hemopoietic tissues of chimeric cells transferred in utero and/or through breastfeeding from allogeneic (9), semiallogeneic (10), and syngeneic (11, 12) mothers. However, mouse models have not been used to systematically investigate the possible influence of maternal-fetal histocompatibility on maternal microchimerism. To help fill this gap in knowledge, we have developed a sensitive and accurate quantitative PCR (QPCR)3 assay for the neomycin resistance gene (neoR), and have used neoR as a noninherited maternal allele (NIMA) in mice to quantitate chimeric maternal cells in neoR–/– N2 backcross progeny of (neoR+/–)F1 females mated with neoR–/– males. In the studies described in this work, we have used this approach to examine the relationship between maternal-fetal histocompatibility and the prevalence and levels of maternal microchimerism in lymphoid tissues and brains of weanling and adult mice. The results indicate that small amounts of NIMA-specific DNA probably representing maternal cells are present in lymphoid and nonlymphoid tissues of virtually all neonatal and adult mice regardless of their pattern of maternal-fetal histocompatibility, with a tendency for higher levels in MHC homozygous mice. Consistent with previously reported evidence of the normalcy of maternal cell transfer and long-term engraftment in immunocompetent progeny (9), the current findings lend support to the possibility that maternal cells have a normal physiological role.
Materials and Methods
Statistical analysis
Differences in the means of the log-transformed fractional levels of NIMA-specific DNA were compared by ANOVA or by the Mann-Whitney U test. Fisher’s protected least significant difference test was used for multiple comparisons with significance set at p < 0.05. All statistical analyses were conducted using the StatView 4.5 software program for Macintosh computers.
Results
Development of a 5' nuclease (TaqMan) QPCR assay for the neoR gene
Our overall strategy in these experiments was to use the neoR gene as a NIMA to quantitate the levels of chimeric maternal cells in neoR–/– N2 backcross progeny of neoR+/– females mated with neoR–/– wild-type males. As a first step, we sought to develop a sensitive and accurate QPCR method for detection of the neoR gene. We first generated obligatory neoR+/– heterozygotes by mating neoR+/+ homozygous B6 FVIII-knockout females with wild-type B6 males. A known amount of genomic DNA isolated from tissue derived from a neoR+/– mouse was then serially diluted in a fixed known amount of genomic DNA from a wild-type neoR–/– mouse, and each dilution was used as genomic DNA in QPCR assays for the neoR gene. To standardize the assay for genomic DNA, a TaqMan assay for rodent GAPDH available from Applied Biosystems was run in parallel with the neoR TaqMan assays on each of the DNA samples. The results showed a tight fit to linearity (Fig. 1), sensitivity to the level of detection of one neoR genome equivalent, and neoR specificity because PCR product was only detected with DNA from neoR+ mice; none was detected after 40 cycles of real-time PCR using DNA from wild-type neoR– B6 mice. Based on these results, subsequent experiments used QPCR assays instead of standard PCR assays to determine the neoR genotype of backcross progeny of all neoR+/– x neoR–/– matings.
Generation of NIMA informative neoR–/– progeny differing in maternal-fetal histocompatibility
Four different backcross mating schemes were used to generate neoR–/– progeny of neoR+/– mothers with different maternal-fetal histocompatibility relationships across the entire H-2 complex (Table I). The terms used in this work to describe these maternal-fetal histocompatibility relationships, homozygous syngenicity, heterozygous syngenicity, fetal semiallogenicity, maternal semiallogenicity, and combined maternal and fetal semiallogenicity, are based on those described by Hoff et al. (14). All members of a given litter of N2 progeny were killed on the same day, and samples of their spleen, thymus, and brain were snap frozen and stored at –80°C. Aliquots of genomic DNA isolated from frozen tissues were used in H-2-specific PCR to determine the H-2 genotypes of the tissue donors, and in neoR-specific QPCR assays to determine their neoR genotypes, and to detect and quantitate NIMA-specific DNA in tissues of mice found to be neoR negative.
Effects of maternal-fetal histocompatibility on levels of NIMA-specific DNA in brain and lymphoid tissue
The results presented in Table IIand Figs. 3 and 4 compare the fractional levels of NIMA-specific DNA in brain and lymphoid tissue from mice generated by pregnancies with different types of maternal-fetal histocompatibility. The data for each maternal-fetal histocompatibility type represent pooled data from the different mating schemes generating each of these types because we detected no significant within-type differences. The fractional levels of NIMA-specific DNA in both brain and lymphoid tissue of mice with homozygous syngenicity and maternal semiallogenicity were similar, and tended to be higher than in mice with either combined maternal-fetal semiallogenicity or heterozygous syngenicity. By ANOVA, the mean fractional level of NIMA-specific DNA in the brain was significantly higher in mice with homozygous syngenicity than in mice with either combined maternal-and-fetal semiallogenicity (p = 0.016) or heterozygous syngenicity (p = 0.05). In lymphoid tissue, as in the brain, the mean fractional level of NIMA-specific DNA in mice with maternal-fetal homozygous syngenicity was significantly higher than in mice with maternal-fetal heterozygous syngenicity (p = 0.02), and tended to be higher than in mice with combined maternal-and-fetal semiallogenicity (p = 0.08). Lymphoid tissue levels were also significantly higher in mice with maternal semiallogenicity than in mice with either combined maternal-and-fetal semiallogenicity (p = 0.02) or heterozygous syngenicity (p = 0.004). As pointed out above, across all mice tested, brain tissue had significantly higher fractional levels of NIMA-specific DNA than lymphoid tissue. However, when analyzed for the effect of maternal-fetal histocompatibility, this difference was only significant in the case of mice with histocompatible mothers, i.e., those with maternal-fetal homozygous syngenicity (p < 0.0001) or heterozygous syngenicity (p = 0.0013), and not in those with maternal semiallogenicity (p = 0.94) or maternal-fetal semiallogenicity (p = 0.69).
Discussion
In the studies described in this work, we developed a highly sensitive QPCR assay for quantitation of the neoR gene, and used the neoR gene as a NIMA in mouse model experiments designed to test for the effects of maternal-fetal histocompatibility on the presence and levels of maternal microchimerism. Our strategy was to use NIMA-specific DNA as a marker of chimeric maternal cells to assess the prevalence and fractional amount of maternal microchimerism in the brain and lymphoid tissue of informative homozygous neoR–/– backcross progeny of heterozygous neoR+/– female mice mated with male wild-type neoR–/– mice. The mating partners were chosen to be homozygous or heterozygous at the MHC locus for either shared or disparate alleles to create progeny with four possible patterns of maternal-fetal histocompatibility. Using this approach, we detected NIMA-specific DNA in the brain and lymphoid tissue of all adult mice so tested regardless of the presence or absence of maternal-fetal histocompatibility. However, comparison of the fractional levels of NIMA-specific DNA in mice derived from pregnancies with different types of maternal-fetal histocompatibility revealed a consistent pattern: regardless of their age at testing, the levels in mice derived from pregnancies with maternal-fetal homozygous syngenicity and maternal semiallogenicity were similar, and tended to be higher than in mice derived from pregnancies with either combined maternal-fetal semiallogenicity or heterozygous syngenicity.
It seems likely that most, if not all, of the NIMA-specific DNA detected in these studies derives from intact chimeric cells, and thereby provides an accurate reflection of the presence and level of maternal microchimerism. However, in the absence of direct evidence for chimeric cells using such techniques as in situ immunohistochemistry and immunofluorescence, it remains possible that at least some of the NIMA-specific DNA represents cell-free DNA released from dead cells either in situ or in the circulation. This is a particularly important consideration with regard to the data obtained in this study with brain tissue, because, while direct evidence for chimeric maternal cells in lymphoid tissues of adult progeny has been previously reported (10), the current findings are the first evidence supporting the presence of chimeric maternal cells in the brain. The existence of the blood barrier to large molecules and the extremely short t1/2 of foreign DNA in the circulation (15) makes it unlikely that the NIMA-specific DNA detected in this study in brain represents circulating DNA released into the circulation by dead cells outside the brain. Moreover, even if the NIMA-specific DNA derives in part from DNA released in situ by chimeric cells, the data still support the presence of such cells in the brain. The only previously reported data bearing on this question is that of Piotrowski and Croy (11), who, using histochemical methods, failed to detect maternal cells in sections of fetal mouse brain tissue obtained from gestation day 8 to term. The discrepancy between the results of that previous study and the data obtained in this study could be due to a lower sensitivity of NIMA detection by histochemistry compared with PCR.
The current data do not permit estimation of the extent to which the NIMA-specific DNA detected in this study represents maternal cells derived from prenatal transplacental passage or maternal cells derived from postnatal lactational transfer. However, previously reported evidence that prenatally acquired maternal cells have a greater capacity for long-term postnatal persistence than those acquired postnatally (10, 16) taken together with the observation that chimeric maternal cells first appear and are most prevalent in the bone marrow (17) favors the notion that the NIMA-specific DNA detected in this study in adult lymphoid tissue and brain represents maternal cells that had originally seeded and successfully engrafted the fetal bone marrow in utero.
Our detection of NIMA-specific DNA in progeny with maternal semiallogenicity fits with findings previously reported by Zhang and Miller (10), who, using immunofluorescence assays, detected small numbers of maternal cells expressing the noninherited maternal H-2d allele in lymph nodes of many, but not all, 6-wk-old H-2b backcross offspring of H-2b/d B6D2F1 females mated with H-2b/b B6 males. The greater prevalence of maternal microchimerism suggested by our results is probably related to a higher sensitivity of detection of NIMA by PCR than by direct immunofluorescence.
The occurrence of maternal microchimerism in mice in the face of maternal-fetal histoincompatibility is also entirely consistent with similar findings in humans (18, 19, 20). In both species, this can be explained, in part, by the fact that partial tolerance to noninherited maternal alloantigens is a normal occurrence. Thus, both human and mouse studies indicate that in utero and neonatal exposure to NIMA-expressing cells frequently induces long-lasting NIMA-specific transplant tolerance in vivo (6, 10, 21, 22) and reductions in the frequency and functional activity of NIMA-specific T cells (22, 23). In the current study, support for a role for NIMA-specific T cells in regulating maternal microchimerism comes from the observation that brain tissue had significantly higher fractional levels of NIMA-specific DNA than lymphoid tissue in mice with histocompatible mothers, but not in mice with histoincompatible mothers. Consistent with the findings of Marleau et al. (9) that the prevalence and level of maternal microchimerism in extramedullary tissues are lower and more transient than in bone marrow in immunocompetent offspring of MHC disparate mothers, but not in immunodeficient or histocompatible offspring, this finding also suggests that NIMA-specific T cell regulation of chimeric maternal cells may play a more important regulatory role in some tissues than others.
Although classic T and B cell alloreactivity to noninherited maternal MHC Ags clearly plays a role in regulating chimeric maternal cells, the overall pattern of similarities and differences in the levels of maternal microchimerism detected in this study in mice derived from pregnancies with different types of maternal-fetal histocompatibility cannot be readily explained by any single mechanism involving unidirectional or bidirectional maternal-fetal T or B cell alloreactions. The expected T and B cell tolerance to self MHC Ags in adult mice derived from pregnancies with homozygous syngenicity and the expected T and B cell alloreactivity to noninherited maternal MHC Ags of adult mice derived from pregnancies with combined maternal and fetal semiallogenicity could readily account for the higher levels of maternal microchimerism detected in the former than in the latter, but none of these mechanisms explains why progeny of homozygous syngeneic pregnancies have higher levels of persistent maternal microchimerism than progeny of heterozygous syngeneic pregnancies in both brain and lymphoid tissue. Homozygous syngeneic progeny and heterozygous syngeneic progeny should both be tolerant to maternal cells because such cells presumably express exactly the same complement of MHC Ags as cells from the progeny themselves. Differences in the presence or absence of T or B cell self-tolerance and alloreactivity also fail to explain why the level of maternal microchimerism is higher in mice with maternal allogenicity than in mice with either heterozygous syngenicity or combined maternal-fetal allogenicity.
Further analysis of the data suggests that zygosity at the MHC locus is the one histocompatibility characteristic that distinguished progeny with relatively high vs those with relatively low levels of maternal microchimerism. Thus, as shown in Fig. 4, compared with progeny with MHC heterozygosity (i.e., those with H-2b/d, H-2b/z, or H-2z/d genotypes), mice with MHC homozygosity (i.e., those with H-2b/b or H-2d/d genotypes) had significantly higher levels of maternal microchimerism in both brain (p = 0.05) and lymphoid tissue (p = 0.004). It can be argued that the zygosity effects on maternal microchimerism observed in this study are due to zygosity effects of non-MHC genes rather than zygosity effects of MHC genes. However, this seems unlikely because identical MHC zygosity effects were observed in comparisons of several different types of MHC homozygous and heterozygous progeny even though the relative maternal and fetal zygosity of most of their non-MHC-linked genes should not necessarily correspond to the relative maternal and fetal zygosity of their MHC genes. Formal confirmation of a role for MHC zygosity in regulating maternal microchimerism could come from studies similar to those described in this work comparing the levels of chimeric maternal cells in MHC congenic mice with different genetic backgrounds.
The apparent importance of MHC zygosity in regulating in vivo levels of persistent maternal microchimerism highlights the likely importance of host NK cells in this phenomenon because, in contrast to T and B cell specificity, NK cell specificity is determined in many cases by the relative MHC zygosity of NK cells and their targets (24, 25). For example, transplanted fully allogeneic cells are rapidly removed by mouse and rat NK cells, whereas semiallogeneic cells usually are not (26, 27, 28). Based on such findings, Zhang and Miller (10) previously pointed out that this pattern of NK specificity favors survival of maternal cells in offspring because such cells are normally semiallogeneic rather than fully allogeneic, and presented evidence directly supporting that hypothesis.
All of the mice tested in this study resulted from pregnancies yielding NeoR-positive as well as NeoR-negative progeny, and derived from matings of mothers with at least two similar pregnancies. Therefore, while most of the NIMA-specific DNA detected in this study almost certainly represents intact chimeric maternal cells, at least some of the NIMA-specific DNA detected in this study could be one or more of three other possible chimeric cells: cells derived from NeoR-positive cohort embryos; cells present in the maternal circulation as a result of transfer from NeoR-positive embryos of previous pregnancies; and cells in the maternal circulation that derived perinatally from a neoR-positive maternal grandmother. The presence and levels, if any, of such alternative chimeric cells remain to be determined.
The recent report of myocardial cells bearing maternal markers in a human neonate with neonatal lupus syndrome (29) suggests that at least some chimeric maternal cells either develop or intrinsically possess properties of somatic tissue cells. Together with our demonstration of the ubiquity and widespread tissue distribution of chimeric maternal cells in both weanling and adult mice, this fits with the possibility that small numbers of chimeric maternal cells may have the beneficial effect throughout an offspring’s life of participating in tissue repair and regeneration.
If maternal microchimerism occurs throughout life, and has the potential for exerting both beneficial tissue reparative effects and deleterious autoimmune effects, what might determine the balance between the two? That scleroderma in women is associated with increased levels of fetal microchimerism, but not increased prevalence of fetal microchimerism (30), suggests that autoimmunity is associated with relatively high levels of normally occurring microchimerism. If so, the relatively lower levels of maternal microchimerism observed in this study in association with MHC heterozygosity may account in part for the recently described protective effect of MHC heterozygosity on autoimmunity (31). In light of this possibility, and the association of autoimmunity with both maternal and fetal microchimerism, it will be of interest to determine whether MHC zygosity affects fetal microchimerism in the same manner as it affects maternal microchimerism.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This research was supported by a grant from the Children’s Research Center of Michigan.
2 Address correspondence and reprint requests to Dr. Joseph Kaplan at the current address: Children’s Hospital of Michigan, 3901 Beaubien Boulevard, Detroit, MI 48201. E-mail address: jkaplan{at}med.wayne.edu
3 Abbreviations used in this paper: QPCR, quantitative PCR; NIMA, noninherited maternal allele.
Received for publication October 12, 2004. Accepted for publication March 8, 2005.
References
Socie, G., E. Gluckman, E. Carosella, Y. Brossard, C. Lafon, O. Brison. 1994. Search for maternal cells in human umbilical cord blood by polymerase chain reaction amplification of two minisatellite sequences. Blood 83: 340-344.
Scaradavou, A., C. Carrier, N. Mollen, C. Stevens, P. Rubinstein. 1996. Detection of maternal DNA in placental/umbilical cord blood by locus-specific amplification of the noninherited maternal HLA gene. Blood 88: 1494-1500.
Petit, T., M. Dommergues, G. Socie, Y. Dumez, E. Gluckman, O. Brison. 1997. Detection of maternal cells in human fetal blood during the third trimester of pregnancy using allele-specific PCR amplification. Br. J. Haematol. 98: 767-771.
Maloney, S., A. Smith, D. E. Furst, D. Myerson, K. Rupert, P. C. Evans, J. L. Nelson. 1999. Microchimerism of maternal origin persists into adult life. J. Clin. Invest. 104: 41-47.
Nelson, J. L.. 1998. Microchimerism and autoimmune disease. N. Engl. J. Med. 338: 1224-1225.
Burlingham, W. J., A. P. Grailer, D. M. Heisey, F. H. Claas, D. Norman, T. Mohanakumar, D. C. Brennan, H. de Fijter, T. van Gelder, J. D. Pirsch, et al 1998. The effect of tolerance to noninherited maternal HLA antigens on the survival of renal transplants from sibling donors. N. Engl. J. Med. 339: 1657-1664.
Bianchi, D. W., R. Romero. 2003. Biological implications of bi-directional fetomaternal cell traffic: a summary of a National Institute of Child Health and Human Development-sponsored conference. J. Matern. Fetal Neonatal Med. 14: 123-129.(Joseph Kaplan2,* and Susa)
To investigate the relationship between maternal-fetal histocompatibility and maternal microchimerism, we developed a sensitive quantitative PCR assay for the neomycin resistance gene (neoR), and, in a mouse model system, used neoR as a noninherited maternal allele marker of maternal cells to detect and quantitate maternal microchimerism in tissues of neoR–/– N2 backcross progeny of (neoR+/–)F1 females mated with neoR–/– males. Using this approach, we obtained evidence for the presence of chimeric maternal cells in the brain, spleen, and thymus of all weanling and adult mice so tested. The numbers of chimeric maternal cells present in the spleen did not differ significantly from those in the thymus regardless of age or maternal-fetal histocompatibility. At all ages, brain tissue had higher level of maternal microchimerism than lymphoid tissue in mice MHC identical with their mothers, but the levels were similar in mice MHC disparate with their mothers. The levels of chimeric maternal cells in both brain and lymphoid tissue of mice with homozygous syngenicity and maternal allogenicity were similar, and tended to be higher than tissue-specific levels in mice with either combined maternal-fetal allogenicity or heterozygous syngenicity. Thus, MHC homozygous progeny had higher levels of maternal microchimerism than MHC heterozygous progeny. We conclude that normal mice possess small numbers of maternal cells in spleen, thymus, brain, and probably most other tissues, and that maternal-fetal histocompatibility influences the levels of these cells by mechanisms related to MHC zygosity of the progeny.
Introduction
Maternal microchimerism, the presence of very small fractional concentrations of chimeric maternal cells, occurs in at least half of all humans from fetal life into adulthood (1, 2, 3, 4). The factors that govern maternal microchimerism are poorly understood, but may be of considerable clinical relevance in view of the fact that maternal microchimerism has been associated with autoimmunity (5) and allograft tolerance (6), and may serve as a vehicle for vertical transmission of pathogenic microbes and allergens (7).
The notion that maternal-fetal histocompatibility plays a role in the regulation of persistent maternal microchimerism derives from the well-known importance of donor-host histocompatibility in regulating donor cell chimerism in allogeneic bone marrow transplant recipients (8). Much of the evidence for this has come from studies that have taken advantage of the ready availability of inbred mouse strains with well-defined MHC genes. There have been several previously reported investigations of maternal microchimerism in mouse model systems that, taken together, demonstrate the frequent occurrence in fetal, neonatal, and adult lymphoid and hemopoietic tissues of chimeric cells transferred in utero and/or through breastfeeding from allogeneic (9), semiallogeneic (10), and syngeneic (11, 12) mothers. However, mouse models have not been used to systematically investigate the possible influence of maternal-fetal histocompatibility on maternal microchimerism. To help fill this gap in knowledge, we have developed a sensitive and accurate quantitative PCR (QPCR)3 assay for the neomycin resistance gene (neoR), and have used neoR as a noninherited maternal allele (NIMA) in mice to quantitate chimeric maternal cells in neoR–/– N2 backcross progeny of (neoR+/–)F1 females mated with neoR–/– males. In the studies described in this work, we have used this approach to examine the relationship between maternal-fetal histocompatibility and the prevalence and levels of maternal microchimerism in lymphoid tissues and brains of weanling and adult mice. The results indicate that small amounts of NIMA-specific DNA probably representing maternal cells are present in lymphoid and nonlymphoid tissues of virtually all neonatal and adult mice regardless of their pattern of maternal-fetal histocompatibility, with a tendency for higher levels in MHC homozygous mice. Consistent with previously reported evidence of the normalcy of maternal cell transfer and long-term engraftment in immunocompetent progeny (9), the current findings lend support to the possibility that maternal cells have a normal physiological role.
Materials and Methods
Statistical analysis
Differences in the means of the log-transformed fractional levels of NIMA-specific DNA were compared by ANOVA or by the Mann-Whitney U test. Fisher’s protected least significant difference test was used for multiple comparisons with significance set at p < 0.05. All statistical analyses were conducted using the StatView 4.5 software program for Macintosh computers.
Results
Development of a 5' nuclease (TaqMan) QPCR assay for the neoR gene
Our overall strategy in these experiments was to use the neoR gene as a NIMA to quantitate the levels of chimeric maternal cells in neoR–/– N2 backcross progeny of neoR+/– females mated with neoR–/– wild-type males. As a first step, we sought to develop a sensitive and accurate QPCR method for detection of the neoR gene. We first generated obligatory neoR+/– heterozygotes by mating neoR+/+ homozygous B6 FVIII-knockout females with wild-type B6 males. A known amount of genomic DNA isolated from tissue derived from a neoR+/– mouse was then serially diluted in a fixed known amount of genomic DNA from a wild-type neoR–/– mouse, and each dilution was used as genomic DNA in QPCR assays for the neoR gene. To standardize the assay for genomic DNA, a TaqMan assay for rodent GAPDH available from Applied Biosystems was run in parallel with the neoR TaqMan assays on each of the DNA samples. The results showed a tight fit to linearity (Fig. 1), sensitivity to the level of detection of one neoR genome equivalent, and neoR specificity because PCR product was only detected with DNA from neoR+ mice; none was detected after 40 cycles of real-time PCR using DNA from wild-type neoR– B6 mice. Based on these results, subsequent experiments used QPCR assays instead of standard PCR assays to determine the neoR genotype of backcross progeny of all neoR+/– x neoR–/– matings.
Generation of NIMA informative neoR–/– progeny differing in maternal-fetal histocompatibility
Four different backcross mating schemes were used to generate neoR–/– progeny of neoR+/– mothers with different maternal-fetal histocompatibility relationships across the entire H-2 complex (Table I). The terms used in this work to describe these maternal-fetal histocompatibility relationships, homozygous syngenicity, heterozygous syngenicity, fetal semiallogenicity, maternal semiallogenicity, and combined maternal and fetal semiallogenicity, are based on those described by Hoff et al. (14). All members of a given litter of N2 progeny were killed on the same day, and samples of their spleen, thymus, and brain were snap frozen and stored at –80°C. Aliquots of genomic DNA isolated from frozen tissues were used in H-2-specific PCR to determine the H-2 genotypes of the tissue donors, and in neoR-specific QPCR assays to determine their neoR genotypes, and to detect and quantitate NIMA-specific DNA in tissues of mice found to be neoR negative.
Effects of maternal-fetal histocompatibility on levels of NIMA-specific DNA in brain and lymphoid tissue
The results presented in Table IIand Figs. 3 and 4 compare the fractional levels of NIMA-specific DNA in brain and lymphoid tissue from mice generated by pregnancies with different types of maternal-fetal histocompatibility. The data for each maternal-fetal histocompatibility type represent pooled data from the different mating schemes generating each of these types because we detected no significant within-type differences. The fractional levels of NIMA-specific DNA in both brain and lymphoid tissue of mice with homozygous syngenicity and maternal semiallogenicity were similar, and tended to be higher than in mice with either combined maternal-fetal semiallogenicity or heterozygous syngenicity. By ANOVA, the mean fractional level of NIMA-specific DNA in the brain was significantly higher in mice with homozygous syngenicity than in mice with either combined maternal-and-fetal semiallogenicity (p = 0.016) or heterozygous syngenicity (p = 0.05). In lymphoid tissue, as in the brain, the mean fractional level of NIMA-specific DNA in mice with maternal-fetal homozygous syngenicity was significantly higher than in mice with maternal-fetal heterozygous syngenicity (p = 0.02), and tended to be higher than in mice with combined maternal-and-fetal semiallogenicity (p = 0.08). Lymphoid tissue levels were also significantly higher in mice with maternal semiallogenicity than in mice with either combined maternal-and-fetal semiallogenicity (p = 0.02) or heterozygous syngenicity (p = 0.004). As pointed out above, across all mice tested, brain tissue had significantly higher fractional levels of NIMA-specific DNA than lymphoid tissue. However, when analyzed for the effect of maternal-fetal histocompatibility, this difference was only significant in the case of mice with histocompatible mothers, i.e., those with maternal-fetal homozygous syngenicity (p < 0.0001) or heterozygous syngenicity (p = 0.0013), and not in those with maternal semiallogenicity (p = 0.94) or maternal-fetal semiallogenicity (p = 0.69).
Discussion
In the studies described in this work, we developed a highly sensitive QPCR assay for quantitation of the neoR gene, and used the neoR gene as a NIMA in mouse model experiments designed to test for the effects of maternal-fetal histocompatibility on the presence and levels of maternal microchimerism. Our strategy was to use NIMA-specific DNA as a marker of chimeric maternal cells to assess the prevalence and fractional amount of maternal microchimerism in the brain and lymphoid tissue of informative homozygous neoR–/– backcross progeny of heterozygous neoR+/– female mice mated with male wild-type neoR–/– mice. The mating partners were chosen to be homozygous or heterozygous at the MHC locus for either shared or disparate alleles to create progeny with four possible patterns of maternal-fetal histocompatibility. Using this approach, we detected NIMA-specific DNA in the brain and lymphoid tissue of all adult mice so tested regardless of the presence or absence of maternal-fetal histocompatibility. However, comparison of the fractional levels of NIMA-specific DNA in mice derived from pregnancies with different types of maternal-fetal histocompatibility revealed a consistent pattern: regardless of their age at testing, the levels in mice derived from pregnancies with maternal-fetal homozygous syngenicity and maternal semiallogenicity were similar, and tended to be higher than in mice derived from pregnancies with either combined maternal-fetal semiallogenicity or heterozygous syngenicity.
It seems likely that most, if not all, of the NIMA-specific DNA detected in these studies derives from intact chimeric cells, and thereby provides an accurate reflection of the presence and level of maternal microchimerism. However, in the absence of direct evidence for chimeric cells using such techniques as in situ immunohistochemistry and immunofluorescence, it remains possible that at least some of the NIMA-specific DNA represents cell-free DNA released from dead cells either in situ or in the circulation. This is a particularly important consideration with regard to the data obtained in this study with brain tissue, because, while direct evidence for chimeric maternal cells in lymphoid tissues of adult progeny has been previously reported (10), the current findings are the first evidence supporting the presence of chimeric maternal cells in the brain. The existence of the blood barrier to large molecules and the extremely short t1/2 of foreign DNA in the circulation (15) makes it unlikely that the NIMA-specific DNA detected in this study in brain represents circulating DNA released into the circulation by dead cells outside the brain. Moreover, even if the NIMA-specific DNA derives in part from DNA released in situ by chimeric cells, the data still support the presence of such cells in the brain. The only previously reported data bearing on this question is that of Piotrowski and Croy (11), who, using histochemical methods, failed to detect maternal cells in sections of fetal mouse brain tissue obtained from gestation day 8 to term. The discrepancy between the results of that previous study and the data obtained in this study could be due to a lower sensitivity of NIMA detection by histochemistry compared with PCR.
The current data do not permit estimation of the extent to which the NIMA-specific DNA detected in this study represents maternal cells derived from prenatal transplacental passage or maternal cells derived from postnatal lactational transfer. However, previously reported evidence that prenatally acquired maternal cells have a greater capacity for long-term postnatal persistence than those acquired postnatally (10, 16) taken together with the observation that chimeric maternal cells first appear and are most prevalent in the bone marrow (17) favors the notion that the NIMA-specific DNA detected in this study in adult lymphoid tissue and brain represents maternal cells that had originally seeded and successfully engrafted the fetal bone marrow in utero.
Our detection of NIMA-specific DNA in progeny with maternal semiallogenicity fits with findings previously reported by Zhang and Miller (10), who, using immunofluorescence assays, detected small numbers of maternal cells expressing the noninherited maternal H-2d allele in lymph nodes of many, but not all, 6-wk-old H-2b backcross offspring of H-2b/d B6D2F1 females mated with H-2b/b B6 males. The greater prevalence of maternal microchimerism suggested by our results is probably related to a higher sensitivity of detection of NIMA by PCR than by direct immunofluorescence.
The occurrence of maternal microchimerism in mice in the face of maternal-fetal histoincompatibility is also entirely consistent with similar findings in humans (18, 19, 20). In both species, this can be explained, in part, by the fact that partial tolerance to noninherited maternal alloantigens is a normal occurrence. Thus, both human and mouse studies indicate that in utero and neonatal exposure to NIMA-expressing cells frequently induces long-lasting NIMA-specific transplant tolerance in vivo (6, 10, 21, 22) and reductions in the frequency and functional activity of NIMA-specific T cells (22, 23). In the current study, support for a role for NIMA-specific T cells in regulating maternal microchimerism comes from the observation that brain tissue had significantly higher fractional levels of NIMA-specific DNA than lymphoid tissue in mice with histocompatible mothers, but not in mice with histoincompatible mothers. Consistent with the findings of Marleau et al. (9) that the prevalence and level of maternal microchimerism in extramedullary tissues are lower and more transient than in bone marrow in immunocompetent offspring of MHC disparate mothers, but not in immunodeficient or histocompatible offspring, this finding also suggests that NIMA-specific T cell regulation of chimeric maternal cells may play a more important regulatory role in some tissues than others.
Although classic T and B cell alloreactivity to noninherited maternal MHC Ags clearly plays a role in regulating chimeric maternal cells, the overall pattern of similarities and differences in the levels of maternal microchimerism detected in this study in mice derived from pregnancies with different types of maternal-fetal histocompatibility cannot be readily explained by any single mechanism involving unidirectional or bidirectional maternal-fetal T or B cell alloreactions. The expected T and B cell tolerance to self MHC Ags in adult mice derived from pregnancies with homozygous syngenicity and the expected T and B cell alloreactivity to noninherited maternal MHC Ags of adult mice derived from pregnancies with combined maternal and fetal semiallogenicity could readily account for the higher levels of maternal microchimerism detected in the former than in the latter, but none of these mechanisms explains why progeny of homozygous syngeneic pregnancies have higher levels of persistent maternal microchimerism than progeny of heterozygous syngeneic pregnancies in both brain and lymphoid tissue. Homozygous syngeneic progeny and heterozygous syngeneic progeny should both be tolerant to maternal cells because such cells presumably express exactly the same complement of MHC Ags as cells from the progeny themselves. Differences in the presence or absence of T or B cell self-tolerance and alloreactivity also fail to explain why the level of maternal microchimerism is higher in mice with maternal allogenicity than in mice with either heterozygous syngenicity or combined maternal-fetal allogenicity.
Further analysis of the data suggests that zygosity at the MHC locus is the one histocompatibility characteristic that distinguished progeny with relatively high vs those with relatively low levels of maternal microchimerism. Thus, as shown in Fig. 4, compared with progeny with MHC heterozygosity (i.e., those with H-2b/d, H-2b/z, or H-2z/d genotypes), mice with MHC homozygosity (i.e., those with H-2b/b or H-2d/d genotypes) had significantly higher levels of maternal microchimerism in both brain (p = 0.05) and lymphoid tissue (p = 0.004). It can be argued that the zygosity effects on maternal microchimerism observed in this study are due to zygosity effects of non-MHC genes rather than zygosity effects of MHC genes. However, this seems unlikely because identical MHC zygosity effects were observed in comparisons of several different types of MHC homozygous and heterozygous progeny even though the relative maternal and fetal zygosity of most of their non-MHC-linked genes should not necessarily correspond to the relative maternal and fetal zygosity of their MHC genes. Formal confirmation of a role for MHC zygosity in regulating maternal microchimerism could come from studies similar to those described in this work comparing the levels of chimeric maternal cells in MHC congenic mice with different genetic backgrounds.
The apparent importance of MHC zygosity in regulating in vivo levels of persistent maternal microchimerism highlights the likely importance of host NK cells in this phenomenon because, in contrast to T and B cell specificity, NK cell specificity is determined in many cases by the relative MHC zygosity of NK cells and their targets (24, 25). For example, transplanted fully allogeneic cells are rapidly removed by mouse and rat NK cells, whereas semiallogeneic cells usually are not (26, 27, 28). Based on such findings, Zhang and Miller (10) previously pointed out that this pattern of NK specificity favors survival of maternal cells in offspring because such cells are normally semiallogeneic rather than fully allogeneic, and presented evidence directly supporting that hypothesis.
All of the mice tested in this study resulted from pregnancies yielding NeoR-positive as well as NeoR-negative progeny, and derived from matings of mothers with at least two similar pregnancies. Therefore, while most of the NIMA-specific DNA detected in this study almost certainly represents intact chimeric maternal cells, at least some of the NIMA-specific DNA detected in this study could be one or more of three other possible chimeric cells: cells derived from NeoR-positive cohort embryos; cells present in the maternal circulation as a result of transfer from NeoR-positive embryos of previous pregnancies; and cells in the maternal circulation that derived perinatally from a neoR-positive maternal grandmother. The presence and levels, if any, of such alternative chimeric cells remain to be determined.
The recent report of myocardial cells bearing maternal markers in a human neonate with neonatal lupus syndrome (29) suggests that at least some chimeric maternal cells either develop or intrinsically possess properties of somatic tissue cells. Together with our demonstration of the ubiquity and widespread tissue distribution of chimeric maternal cells in both weanling and adult mice, this fits with the possibility that small numbers of chimeric maternal cells may have the beneficial effect throughout an offspring’s life of participating in tissue repair and regeneration.
If maternal microchimerism occurs throughout life, and has the potential for exerting both beneficial tissue reparative effects and deleterious autoimmune effects, what might determine the balance between the two? That scleroderma in women is associated with increased levels of fetal microchimerism, but not increased prevalence of fetal microchimerism (30), suggests that autoimmunity is associated with relatively high levels of normally occurring microchimerism. If so, the relatively lower levels of maternal microchimerism observed in this study in association with MHC heterozygosity may account in part for the recently described protective effect of MHC heterozygosity on autoimmunity (31). In light of this possibility, and the association of autoimmunity with both maternal and fetal microchimerism, it will be of interest to determine whether MHC zygosity affects fetal microchimerism in the same manner as it affects maternal microchimerism.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This research was supported by a grant from the Children’s Research Center of Michigan.
2 Address correspondence and reprint requests to Dr. Joseph Kaplan at the current address: Children’s Hospital of Michigan, 3901 Beaubien Boulevard, Detroit, MI 48201. E-mail address: jkaplan{at}med.wayne.edu
3 Abbreviations used in this paper: QPCR, quantitative PCR; NIMA, noninherited maternal allele.
Received for publication October 12, 2004. Accepted for publication March 8, 2005.
References
Socie, G., E. Gluckman, E. Carosella, Y. Brossard, C. Lafon, O. Brison. 1994. Search for maternal cells in human umbilical cord blood by polymerase chain reaction amplification of two minisatellite sequences. Blood 83: 340-344.
Scaradavou, A., C. Carrier, N. Mollen, C. Stevens, P. Rubinstein. 1996. Detection of maternal DNA in placental/umbilical cord blood by locus-specific amplification of the noninherited maternal HLA gene. Blood 88: 1494-1500.
Petit, T., M. Dommergues, G. Socie, Y. Dumez, E. Gluckman, O. Brison. 1997. Detection of maternal cells in human fetal blood during the third trimester of pregnancy using allele-specific PCR amplification. Br. J. Haematol. 98: 767-771.
Maloney, S., A. Smith, D. E. Furst, D. Myerson, K. Rupert, P. C. Evans, J. L. Nelson. 1999. Microchimerism of maternal origin persists into adult life. J. Clin. Invest. 104: 41-47.
Nelson, J. L.. 1998. Microchimerism and autoimmune disease. N. Engl. J. Med. 338: 1224-1225.
Burlingham, W. J., A. P. Grailer, D. M. Heisey, F. H. Claas, D. Norman, T. Mohanakumar, D. C. Brennan, H. de Fijter, T. van Gelder, J. D. Pirsch, et al 1998. The effect of tolerance to noninherited maternal HLA antigens on the survival of renal transplants from sibling donors. N. Engl. J. Med. 339: 1657-1664.
Bianchi, D. W., R. Romero. 2003. Biological implications of bi-directional fetomaternal cell traffic: a summary of a National Institute of Child Health and Human Development-sponsored conference. J. Matern. Fetal Neonatal Med. 14: 123-129.(Joseph Kaplan2,* and Susa)