Main

Bidirectional passage of maternal and fetal cells across the placenta can give rise to a chimeric state that persists for years in both the mother and her offspring (1,2). The function and the significance of these semiallogenic cells are not known. Although the consequences of fetal microchimerism (a small amount of fetal cells in the mother) are not unknown, fetal cells have been speculated to be involved in triggering autoimmune disorders (3–7) or to act as a rejuvenating source involved in tissue repair processes (8,9). Other reports point to fetal chimeric cells as “innocent bystanders” in inflammatory or healing processes (10).

The passage of a small amount of maternal cells to the offspring during pregnancy, maternal microchimerism (MMc), has been studied to a lesser extent than fetal microchimerism. Cells of maternal origin have been detected at various concentrations in fetal blood from week 13 (11) and in up to 100% in cord blood samples depending on the method used for detection of maternal cells (12–14).

We recently reported the detection of maternal immunocompetent cells in 11-s trimester fetuses (15) that were aborted due to malformations and/or trisomy 21 or for social reasons. Maternal cells were found in seven of the 11 fetuses, and these cells expressed surface markers for T- and B-lymphocytes, leukocytes, and hematopoietic progenitor cells. The maternal cells were present in all organs investigated and were found in both normal fetuses and in those with malformations and/or trisomy 21.

Furthermore, Stevens et al. (16) recently showed that maternal cells were present in various tissue types in seven male infants. They also showed that the maternal cells were differentiated into specific organ cell types. Maloney et al. (2) showed that cells of maternal origin were present in the blood of healthy individuals through middle-age.

As with fetal microchimerism, maternal cells in the offspring have been associated with different autoimmune diseases especially conditions that arise in children (17–25). Because the fate of those semiallogenic maternal cells is largely unknown, particularly in tissues of healthy individuals, we sought to investigate whether maternal cells were present in lymphoid tissues of children during infancy in an attempt to elucidate an eventual role of MMc in immune development. In previous studies, we demonstrated the occurrence of mature maternal T and B cells as well as hematopoietic progenitors in various fetal tissues. In this study, we investigated MMc in tonsil and adenoid tissues, which are inherently rich in lymphoid cells and may even represent a potential site of postnatal lymphopoiesis (26).

METHODS

Patients.

Tissue samples from the tonsils and/or adenoids of 20 children undergoing tonsillectomy and/or adenoidectomy were collected prospectively. From the literature and our previous work, and we assumed a 30% MMc prevalence in various tissues. Thus, we speculated that 20 cases should provide a sample size to test for the presence of maternal cells in tonsil and adenoid tissues of the offspring.

The indications for surgery were respiratory obstruction (snoring) and/or recurrent tonsillitis (Table 1). A questionnaire about the health conditions of both mother and child was given to the mother of the child. Questions included the length of breast-feeding time, the obstetric history of the mother, whether the mother or child had ever received a blood transfusion, and whether the mother or child was a twin. The characteristics of the children and their mothers are shown in Table 1. None of the children or mothers was a twin or had ever received a blood transfusion, and all the participants were healthy except one mother with hypothyreosis and one with fibromyalgia. The obstetric history of the children was normal in all except in one case where the child was delivered in gestational week 33 because of abruption of the placenta (case 17).

Table 1 Characteristics of children and mothers
Full size table

Tissue collection.

Peripheral blood was collected from the mothers and the children. In the children, both tonsils and adenoids were removed by curettage under general anesthesia and sent to the Pathology Department immediately after the operation for dissection and further analysis.

The tissues were washed with sterile saline and dissected under sterile conditions. Each tonsil was divided into two halves. From each half, a piece of tissue measuring approximately 5 × 5 × 5 mm was collected for PCR analysis. Thus, a total of four fractions from each tonsil pair were included in the analysis (denoted T1 to T4). These fractions were harvested from the central area of the tonsil to avoid contamination by epithelial cells and possible inflammatory exudates on the surface of the crypts. The adenoid was similarly divided into two halves and a small (central) piece of tissue from each half was taken for PCR analysis. No histopathological examination of the tissues was performed. This study was approved by the Ethics Committee at Karolinska University Hospital (Dnr 2008/3:3). Institutional Review Board consent was obtained from all patients.

Quantitative RT-PCR.

DNA from blood, tonsil, and adenoid samples was extracted using Qiagen Mini Kit (Qiagen, Hilden, Germany). The concentration of DNA in the tissues was initially diluted to 100 ng/μl, to yield a total amount of 500 ng in each PCR reaction.

The primers and probes used for chimerism analysis were adapted from others (27,28). In short, biallelic genetic systems were used to screen DNA samples. For each biallelic system, one of the primers was from the polymorphic region to specifically amplify each allele, whereas the second primer and the probe were common to both alleles. An allele was considered informative when it was positive for maternal DNA and negative for the child's DNA. Detection and quantification were performed with an ABI 7000 Sequence Detection System (Applied Biosystems, Foster City, CA) using TaqMan technology. The PCR parameters were as follows: 2 min at 50°C, 10 min at 95°C, followed by 40 cycles of 15 s at 95°C and 1 min at 60°C. RQ-PCR reactions were performed in a total volume of 25 μL containing 1× TaqMan Universal PCR Master Mix (Applied Biosystems, containing dNTPs with dUTP, MgCl2, AmpliTaq Gold DNA Polymerase, AmpErase UNG, Passive Reference), 5 μL of DNA sample, 300 nM of each primer, and 200 nM probe. The amount of amplifiable DNA in each sample was assessed by parallel amplification of the reference gene glyceraldehyde phosphate dehydrogenase (GAPDH). All samples were run in duplicate and both maternal and child DNA samples were included in each run. For a positive result, both reactions were required to yield amplification signals before cycle 40. The negative controls included water and a low concentration (10 ng) blood-DNA sample from the corresponding child. Relative quantification of recipient DNA was calculated according to the ΔΔCt method (Applied Biosystems, user bulletin 2) using GAPDH as a reference gene and the maternal DNA sample as a calibrator. The formula used was 2−(ΔCtU−ΔCtC), where ΔCt = Ct target gene − Ct reference gene. U was the unknown sample and C was the calibrator sample.

The specificity and sensitivity of the RQ-PCR method was determined for all markers included in this study using artificial DNA mixtures and varying DNA amounts. The PCR assay was validated by mixing DNA from one individual (positive for the three markers) in different concentration in a second individual (negative for the markers). The total amount of DNA was 500 ng, which was demonstrated by a stable Ct value for GAPDH (data not shown). We detected no false positive results using 500 ng DNA and 40 cycles of PCR amplification.

HLA determination.

HLA-A, -B and -DRB1 polymorphisms were determined using standard PCR-based methods (LabType, One Lambda, Inc, CA) including sequence-specific oligonucleotide probes (PCR-SSOP) using xMAP technology (Luminex System, Luminex Corporation, Austin, TX).

RESULTS

Maternal cells were identified in four of 20 children (Table 2). The children that were positive for MMc in the tonsils and/or adenoids were two boys (cases 9 and 15) and two girls (cases 10 and 14) between 4 and 6 y of age. Cases 9 and 14 were positive in all tissue fractions as well as in the blood tested. Case 10 was positive for maternal cells in all fractions tested except in one of the tonsil fractions (T4) and in the two fractions of the adenoid. Case 15 was positive in only one of the four fractions of the tonsils and in blood. All cases positive for MMc in blood were also found positive in tonsil or adenoid tissue, in approximately similar rates. Detailed RT-PCR data for samples with MMc are shown in Table 3. No apparent correlation between the presence of maternal cells in the tonsils and/or adenoids and the various combinations of mother/child HLA classes was found (data not presented).

Table 2 Detection of MMc in tonsils and adenoids
Full size table
Table 3 RT-PCR data on samples with MMc
Full size table

DISCUSSION

This study confirms previous reports indicating that MMc is a rather common phenomenon. Most studies have evaluated the incidence of MMc in patients affected by different diseases, although limited information has been reported regarding MMc in healthy individuals, especially in tissues. In second trimester, fetuses from socially motivated terminations, we demonstrated MMc in various organs in about 50% of the cases from the 13th week of gestation (15). Recently, Mold et al. (29) reported even higher MMc rates in fetal lymph nodes acquired from second trimester abortions. Up to 100% of all children at birth, harbors maternal cells in cord blood (12–14), and the estimated frequency of maternal cells varies between 0.02 and 5% (9,11,12). Data on MMc after birth are primarily based on blood samples. MMc can persist to middle age, and depending on the method used for detection, MMc rates range from 20 to 39% in different cell populations in peripheral blood (30). Thus, MMc seems to be frequent and persists for many years.

Information on the amount and distribution of maternal chimeric cells in organs and tissues of healthy individuals may be crucial in understanding the eventual physiologic role of these cells; however, very few studies have addressed this issue. The results of these reports are summarized in Table 4. In studies on tissues from abortions, biopsies, and autopsy material, MMc have been demonstrated in a wide range of tissues such as skin, heart, liver, spleen, thymus, thyroid, and pancreas (15–25,29,30,31), and no particular organ has an apparent predilection for MMc. These reports vary greatly with regard to the frequency of MMc and cite frequencies from 0 to 100%, although several limitations impacted the validity of the published results. In many cases, the patient material was limited in size, and the methods for detection were very disparate. Moreover, most of the controls used in the studies were patients affected with severe disease in the specific organ studied (i.e., skin, muscle, liver, etc) rather than healthy individuals.

Table 4 Studies showing MMc in fetuses, children, and adolescents
Full size table

This study is the first report on MMc in tonsil and adenoid tissues. Our investigation of these tissues was prompted, in part, by the fact that we could study maternal cells in lymphoid organs that are readily available in tissues of relatively healthy children. The tonsil and adenoid tissues develop after birth and constitute a first line of defense against microbial invasion of the upper aero-digestive system. Both B- and T-cell-related immunologic processes are initiated in different specialized compartments of the palatine tonsils (32,33). Also, the adenoid is an active immunologic site in the nasopharyngeal region that harbors cells from both the innate and adaptive immune systems (34,35).

Here, we analyzed fresh tissue, targeting non-shared maternal polymorphic regions in the extracted DNA. Several limitations were inherent to our study. Although we found MMc in 20% of the cases, the number of patients included in the study was limited. Furthermore, in agreement with Alizadeh et al. (27), the sensitivity of the RQ-PCR chimerism method is 0.01% for most markers. Below 0.1%, the linearity of the method is disrupted, but the sensitivity may still be as low as 0.01%. The level of maternal chimerism varies between individuals, and although this method was very sensitive, an even more sensitive method might have yielded more positive cases.

Thus, in theory, this method may underestimate MMc. Moreover, because we did not carry out any in situ analysis, we were unable to evaluate the anatomical location and integration of these cells. We, therefore, cannot claim that the maternal cells are integrated into the tonsil tissue or that these cells are circulating in the small vessels associated with the crypts of the tonsil. This issues require further studies combining immunohistochemistry with other techniques for identification of maternal cells, such as in situ hybridization for X and Y chromosomes. Similarly, although our data suggests good correlation between blood and tonsillar tissue MMc, the issue is not resolved because we have not provided evidence for engraftment of maternal chimeric cells.

Questions remain as to why MMc persists in some individuals despite HLA disparity between mother and child. The mechanism by which maternal cells are recognized and tolerated by the fetus is unknown, but it is plausible that acceptance of maternal cells is a function of the immature immune system of the unborn child. In this study, we were unable to demonstrate any obvious relationship between HLA types of mother and child and the presence of MMc; however, the relatively limited material in our study did not allow for a definitive answer about whether HLA compatibility between mother and child may influence the appearance (or type) of MMc.

The biologic significance of MMc is not known. MMc appears to be a part of normal biology and some evidences suggest that MMc plays an important role in the mother's progeny. Recently, Mold et al. (29) presented evidence that maternal cells cross the placenta and reside in fetal lymph nodes. Moreover, some of these cells induced the development of fetal cells into CD4+CD25 high Fox p3+Tregs that serve to suppress fetal antimaternal immunity and thereby the reaction to noninherited maternal alleles (NIMAs). In addition, some data suggest that maternal cells and antigen modulate the antigen-specific reactivity of the fetal immune system and thereby induce a long-lasting form of tolerance. In a study of kidney transplantation, Burlingham et al. (36) demonstrated that for a kidney from a sibling donor who is mismatched with the recipient for one HLA haplotype, graft survival is higher when the donor has maternal HLA antigens not inherited by the recipient. Furthermore, the presence of NIMAs in other types of transplantations is associated with fewer rejections than transplants carrying noninherited paternal alloantigens (37,38). Thus, these studies suggest that T cell tolerance to alloantigens during fetal life may be maintained after birth. Similar findings have been described with regard to B-cell tolerance (39). The prospect of a close interaction between mother and fetus with regard to immunologic function and development may have far-reaching consequences in understanding pregnancy, fetal complications, tolerance to antigens, and transmission of infective agents during pregnancy. In accordance with such speculation, a recent study by Mackelprang et al. (40) demonstrated that the risk for in utero and breast milk HIV-1 transmission is increased with HLA concordance and homozygosity.

Another consideration when speculating about the function of maternal cells is whether these cells are able to demonstrate cellular plasticity. Stevens et al. (19) in studies on MMc in myocardium of children with neonatal lupus syndrome demonstrated evidence that maternal cells had differentiated into cardiac myocytes. In our previous studies, we demonstrated that CD 34+ and CD 45+ cells can be transferred across the placenta from the mother to the fetus. Both these cell types might exhibit plasticity properties (41).

Another consideration when speculating about the function of maternal cells is whether these cells are able to demonstrate cellular plasticity. In our previous studies, we demonstrated that CD 34+ and CD 45+ cells can be transferred across the placenta from the mother to the fetus. These cells express a cell marker for hematopoietic and endothelial cell progenitors. Stevens et al. (19) studied similar cell markers in heart tissue of children with neonatal lupus and congenital heart block and demonstrated that 86% of cells of maternal origin found in the myocardium also expressed sarcomeric α-actin, a specific marker of cardiomyocytes. Thus, maternal cells in tissues of the offspring appear to be able to exhibit plasticity. This natural plasticity may play an important role in the influence of the mother on her offspring. This mechanism may also explain how acquired abilities, both beneficial and/or detrimental to the child, may be transferred from one generation to another as an epigenetic effect.

In conclusion, MMc was found in 20% of tonsil and adenoid tissue samples from healthy children from 2 to 15 y of age. To our knowledge, this study is the first to investigate MMc in these tissues. Because these organs have central roles in immune regulation of the developing child, further studies are needed to explore the biologic effects of MMc in these tissues.