Introduction

During pregnancy, cells of maternal origin enter the fetal circulation as early as 13 weeks of gestation (Petit et al., 1997; Lo et al., 1998). Beyond circulation, maternal chimeric cells have been identified in various tissues from a small series of newborn patients, such as thymus, liver, thyroid, and also skin (Srivatsa et al., 2003). Interestingly, maternal cells can persist until adult life up to 49 years (Maloney et al., 1999). The consequences of this chimerism are currently under investigation. Maternal and fetal cells are semi-allogenic. Therefore, one hypothesis is that maternal cells may induce a reaction against the child's various tissues, triggering features of an "auto-immune" disease. In accordance, investigators have reported the presence of maternal microchimeric cells in lesional muscle and skin specimens of juvenile dermatomyositis, but not in controls (Artlett et al., 2000; Reed et al., 2000; Selva-O'Callaghan et al., 2001). Maternal CD4+ cells have been identified in the peripheral blood lymphocyte of juvenile dermatomyositis patients (Artlett et al., 2000; Reed et al., 2004). However, in the affected muscle, the precise type of maternal infiltrating cells was not analyzed in these studies. Similarly, adult patients affected with systemic sclerosis were more prone for having maternal cell microchimerism in their peripheral blood (Lambert et al., 2004). In contrast, in the myocardium from foetuses with neonatal lupus syndrome, maternally derived cardiomyocytes were identified (Stevens et al., 2003). This unexpected result suggested that maternal stem cells may have migrated to a damaged myocardium and adopted a myocardial phenotype. This raises an alternate hypothesis: maternal stem cells participate in the child's tissue regeneration process and do not trigger the disease.

Pityriasis lichenoides (PL) is an eruption of unknown cause, characterized clinically by successive flares of inflammatory skin lesions. The disease occurs more frequently among children and young adults. Some investigators have suggested that viral agents such as parvovirus B19 (Tomasini et al., 2004), varicella-zoster (Boralevi et al., 2003), or Epstein–Barr virus (Klein et al., 2003) could initiate these lesions. Pathological findings consist in a dermal peri-vascular and peri-adnexial lymphocytic infiltration in wedge shape, extending to the deep reticular dermis associated with superficial erythrocytes extravasation. The epidermis shows various levels of acanthosis, parakeratosis or scale-crusts, usually housing nuclear debris, keratinocytic necrosis, and exocytosis by a lymphocytic infiltrate of small size that can affect as well the adnexal epithelia or the hair follicles. The lymphocytes have been characterized as mostly CD3+, CD8+, Tia1+, cytotoxic T cells (Tomasini et al., 2004). All these features may also be observed during the cutaneous graft-versus-host disease. We therefore asked whether transplacentally transferred maternal cells could be identified as allogenic cells in the skin of children affected with PL.

Results

In skin biopsies of male children studied by fluorescence in situ hybridization (FISH), the average estimated number of scanned nucleated cells per patient was 145,000 among the PL cases and 571,000 among the controls (P=0.23) (Table 1). Female cells were detected in 11/12 patients affected with PL as compared with 4/7 controls (P=0.15) (Figure 1). The frequency of female microchimeric cells was 99 per million cells in PL patients as compared to 5 per million cells in controls (P=0.005). Most microchimeric cells were detected in the epidermis and only a few were in the dermis.

Figure 1.

Maternal microchimeric cells are common in lesional skin from PL patients. Photomicrographs show FISH experiments on paraffin-embedded sections of skin biopsies from PL patients. We used hybridization probes to X and Y chromosomes labelled respectively with Cy3 (red) and FITC (green). All nuclei were counterstained with 4',6-diamidino-2-phenylindole (blue). (a) Photomicrograph showing a female microchimeric XX cell (two red spots in a single nucleus with intact borders; white arrow) in the epidermis surrounded by male XY cells (black arrows indicate Y chromosomes). All chromosomes may not appear in the same plane of focus due to the thickness of our sections ( 630 magnification). (b) Photomicrograph showing a female microchimeric XX cell (two red spots in a single nucleus with intact borders; white arrow) in the dermis. All chromosomes may not appear in the same plane of focus due to the thickness of our sections. Original magnification: 630.

Full figure and legend (82K)

Table 1 - Patients affected with PL have higher maternal cell microchimerism in their skin.

Full table

We next sought to determine the phenotype of maternal microchimeric cells in the epidermis. In most cases studied, the epidermis was infiltrated with leukocytes. We therefore used CD45 to identify leukocytes, and CD1a and cytokeratin to identify cell types that usually reside in the epidermis such as Langerhans cells and keratinocytes. Combined FISH with immunostaining using anti-CD45 antibody was used on nine sections from patients affected with PL and five sections from controls. None of the 19 microchimeric cells from the PL patients stained positive with CD45. Similarly, on 10 sections from PL patients and seven from the controls, none of the 20 microchimeric cells that were identified stained positively with CD1a. However, using cytokeratin staining, seven out of 10 patients with PL had cytokeratin-positive epidermal female cells. Overall, 14/21 microchimeric cells were cytokeratin positive (Figure 2). The positive cells were located exclusively in follicular or interfollicular epidermis. Most of the female cells that did not stain with anti-cytokeratin antibody were in the dermis.

Figure 2.

Microchimeric maternal cells in the epidermis express cytokeratin. Photomicrographs show combined FISH and immunostaining experiments. We used hybridization probes to X and Y chromosomes labelled respectively with Cy3 (red) and FITC (green). All nuclei were counterstained with 4',6-diamidino-2-phenylindole (blue). (a) CD45-positive (3-amino-9-ethylcarbazole substrate for CD45 staining in red; black arrows) male cells adjacent to a microchimeric female cell that was not stained with anti-CD45 antibody. (b) CD1a-positive (3-amino-9-ethylcarbazole substrate for CD1a staining in red; black arrow) male cell adjacent to a microchimeric female cell that was not stained with anti-CD1a antibody. (c, d) Cytokeratin-expressing (revealed by FITC-labelled secondary antibody in green) microchimeric cells in the epidermis. The morphology is highly suggestive of keratinocytes. (e) Microchimeric cell in the epidermal layers of a hair follicle. (f) Same microchimeric cell stained with cytokeratin (green) showing a keratinocyte phenotype.

Full figure and legend (194K)

Discussion

The results of this study demonstrate the presence of maternally derived cells in lesional skin of male children presenting with inflammatory conditions, and more particularly in PL. Indeed, all the studied male children with PL never had any transfusion, organ transplantation, or a twin sister. Therefore, the female cells detected in this context were most likely of maternal origin. Further on, phenotyping of these maternal chimeric cells showed that these were keratinocytes, as they expressed cytokeratin, but not CD45 and CD1a, and were located almost exclusively in interfollicular or follicular epidermis. Of note, the frequency of maternal cells was higher in PL than in children with other inflammatory skin disorders.

Our study was performed using very stringent criteria for determining the female genotype of cells (see Materials and Methods). However, one cannot exclude that some XX cells were in fact XY cells undergoing mitosis (XXYY) or aneuploid cells. To overcome this problem, we used as controls inflammatory skin diseases where the level of inflammation was similar to our cases. Detecting maternal XX cells requires that the nucleus be individualized. This is hardly achievable in inflammatory areas where the cells are small and overlapping. We therefore think that our study underestimates the possibility that some microchimeric maternal cells may express CD45. However, we believe that our analysis is totally valid for the maternal cells found in the epidermis.

Previous reports have conclusively demonstrated the presence of maternal microchimeric cells in the skin. However, their phenotype was not studied and these may also have been keratinocytes. Indeed, maternal cells have been found in unaffected skin samples from autopsied newborns in two out of four cases (Srivatsa et al., 2001). Similarly, in one fetus affected with neonatal lupus syndrome and in young patients affected with juvenile inflammatory myopathy, maternal cell microchimerism has also been reported in the skin (Artlett et al., 2000; Stevens et al., 2003). In the present study, maternal cell chimerism could be identified in cases and controls in lesional as well as in healthy skin although at a much lower frequency. This difference in the frequency of maternal microchimeric cells was not previously analyzed.

The fact that in inflammatory cutaneous conditions occurring in children, chimeric cells appeared to be mainly keratinocytes – rather than immune effector cells – is in accordance with recent findings on lupus (Stevens et al., 2003). In infants with neonatal lupus erythematosus, maternally derived cells in the myocardium were indeed actin-positive myocytes. This result in lupus – as well as ours in PL – suggests that the maternally derived cells found in the children's damaged tissues are not immune effector cells inducing a graft (mother) versus host (child) disease. Of note, PL has never been associated with any HLA type I or type II alleles. In contrast, these maternal cells appear to be tissue-specific differentiated cells. These results suggest two hypotheses: either the initial step is a primitive existence of maternally derived keratinocytes in the healthy skin of children. In this scheme, the higher density of such cells in PL as compared to healthy skin could trigger a host versus graft type inflammation. Alternatively, maternal stem cells may specifically migrate to damaged skin, after the onset of PL, in order to help repair. There may be a higher need for this role in PL compared to other skin inflammatory disorders, such as atopic dermatitis, because there is more epidermal necrosis. Such phenomenon has already been suggested in fetal cell microchimerism (Khosrotehrani et al., 2004).

Previous studies of bone marrow transplant recipients in humans and animal models have demonstrated the possibility of finding keratinocytes of donor origin (Korbling et al., 2002; Borue et al., 2004; Harris et al., 2004). Therefore, our above data as well as those of Stevens et al. suggest that under pathologic conditions, maternal microchimeric cells can adopt the child's tissue phenotype in a neurectodermal and mesodermal tissue. Maternally derived stem cells could migrate to a damaged tissue and fuse or differentiate to finally adopt the phenotype of keratinocytes or myocytes. Whether the transferred maternal cells responsible for this phenomenon are hemopoietic stem cells (Krause et al., 2001) or another type remains unknown.

In conclusion, we identified maternally derived keratinocytes in pediatric patients with inflammatory skin disorders, especially PL. Further studies would need to investigate whether maternally derived cells contribute to the disease or help repairing it.

Materials and Methods

Patients

We retrieved routine archival paraffin-embedded lesional skin biopsies from 12 male children affected with PL (aged 2–13 years, average 7.2 years) and five male children with other inflammatory skin disorders (four atopic dermatitis and one cutaneous drug reaction) according to local ethical committee procedures with regard to the Helsinki guidelines and patient consent. All PL patients were diagnosed with PL chronica on the basis of a histological examination of a lesional skin biopsy. In addition, two healthy cutaneous specimens adjacent to congenital naevi surgical pieces were obtained. All patients and controls were male, without any history of transfusion or being a twin.

Fluorescence in situ hybridization

FISH was performed as previously described (Johnson et al., 2000) on 5-m-thick paraffin-embedded sections. We used X and Y chromosome probes labelled respectively with Cy3 (red) and FITC (green) mapping to Xp11.1q11.1 – a satellite centromeric region of the X chromosome – and Yq12 satellite III region of the Y chromosome (Vysis, Downers Grove, IL).

Combined FISH with immunostaining

We combined FISH and immunostaining as previously described (Khosrotehrani et al., 2003, 2004). The antibodies used were anti-CD45, anti-CD1a (Dako, Carpintera, CA), and anti-cytokeratin antibodies (Chemicon International, Temecula, CA) that stain leukocytes, Langerhans cells, and keratinocytes or other epithelial cells, respectively. Staining was revealed by the Envision+peroxidase kit (Dako) for CD1a and CD45. For cytokeratin labelling, we used immunofluorescence with secondary goat anti-mouse antibody labelled with fluorescein (Jackson ImmunoResearch, West Grove, PA). All slides were counterstained with 4',6-diamidino-2-phenylindole.

Scoring and statistical analysis

Sections from male patients and controls were all blindly scanned for the presence of female – presumably maternal – cells. Sections were scored if more than 70% of the nuclei had at least one FISH signal. Male cells were recognized as having one X (red) and one Y (green) chromosome. In order to conclusively identify a female cell, the following criteria were requested: (1) presence of two red signals corresponding to the size and shape of adjacent X chromosomes inside a blue-colored nucleus; (2) presence of a minimal distance between the two signals corresponding to at least the size of one of them; (3) presence of intact and recognizable nuclear borders; (4) absence of overlapping nuclei; and (5) absence of a Y chromosome FISH signal (green). When all these criteria were fulfilled, the cell was considered as microchimeric and counted. Within each section, we estimated the total number of nucleated cells: the number of nuclei in three fields ( 630 magnification) in the dermis and the number of nuclei in three fields in the epidermis were counted separately. The total number of epidermal nuclei and dermal nuclei per section was obtained by multiplying the mean nuclei count per field of each skin layer with the number of fields in the dermis and the epidermis. This was the denominator for each section, provided in the last column of Table 1.

For each case, the number of microchimeric cells found was adjusted to the total number of cells examined. The average frequency of microchimeric cells was then compared in the PL and the control group using Student's t-test.

Conflict of Interest

The authors state no conflict of interest.

References

1. Artlett CM, Ramos R, Jiminez SA, Patterson K, Miller FW, Rider LG (2000) Chimeric cells of maternal origin in juvenile idiopathic inflammatory myopathies. Childhood Myositis Heterogeneity Collaborative Group. Lancet 356:2155–2156 | Article | PubMed | ISI | ChemPort |
2. Boralevi F, Cotto E, Baysse L, Jouvencel AC, Leaute-Labreze C, Taieb A (2003) Is varicella-zoster virus involved in the etiopathogeny of pityriasis lichenoides? J Invest Dermatol 121:647–648 | Article | PubMed | ISI |
3. Borue X, Lee S, Grove J, Herzog EL, Harris R, Diflo T et al. (2004) Bone marrow-derived cells contribute to epithelial engraftment during wound healing. Am J Pathol 165:1767–1772 | PubMed | ISI |
4. Harris RG, Herzog EL, Bruscia EM, Grove JE, Van Arnam JS, Krause DS (2004) Lack of a fusion requirement for development of bone marrow-derived epithelia. Science 305:90–93 | Article | PubMed | ISI | ChemPort |
5. Johnson KL, Zhen DK, Bianchi DW (2000) The use of fluorescence in situ hybridization (FISH) on paraffin-embedded tissue sections for the study of microchimerism. Biotechniques 29:1220–1224 | PubMed | ISI | ChemPort |
6. Khosrotehrani K, Johnson KL, Cha DH, Salomon RN, Bianchi DW (2004) Transfer of fetal cells with multilineage potential to maternal tissue. JAMA 292:75–80 | Article | PubMed | ISI | ChemPort |
7. Khosrotehrani K, Stroh H, Bianchi DW, Johnson KL (2003) Combined FISH and immunolabeling on paraffin-embedded tissue sections for the study of microchimerism. Biotechniques 34:242–244 | PubMed | ISI | ChemPort |
8. Klein PA, Jones EC, Nelson JL, Clark RA (2003) Infectious causes of pityriasis lichenoides: a case of fulminant infectious mononucleosis. J Am Acad Dermatol 49:S151–S153 | Article | PubMed | ISI |
9. Korbling M, Katz RL, Khanna A, Ruifrok AC, Rondon G, Albitar M et al. (2002) Hepatocytes and epithelial cells of donor origin in recipients of peripheral-blood stem cells. N Engl J Med 346:738–746 | Article | PubMed | ISI |
10. Krause DS, Theise ND, Collector MI, Henegariu O, Hwang S, Gardner R et al. (2001) Multi-organ, multi-lineage engraftment by a single bone marrow-derived stem cell. Cell 105:369–377 | Article | PubMed | ISI | ChemPort |
11. Lambert NC, Erickson TD, Yan Z, Pang JM, Guthrie KA, Furst DE et al. (2004) Quantification of maternal microchimerism by HLA-specific real-time polymerase chain reaction: studies of healthy women and women with scleroderma. Arthritis Rheum 50:906–914 | Article | PubMed | ISI | ChemPort |
12. Lo ES, Lo YM, Hjelm NM, Thilaganathan B (1998) Transfer of nucleated maternal cells into fetal circulation during the second trimester of pregnancy. Br J Haematol 100:605–606 | Article | PubMed | ISI | ChemPort |
13. Maloney S, Smith A, Furst DE, Myerson D, Rupert K, Evans PC et al. (1999) Microchimerism of maternal origin persists into adult life. J Clin Invest 104:41–47 | Article | PubMed | ISI | ChemPort |
14. Petit T, Dommergues M, Socie G, Dumez Y, Gluckman E, Brison O (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 | Article | PubMed | ISI | ChemPort |
15. Reed AM, McNallan K, Wettstein P, Vehe R, Ober C (2004) Does HLA-dependent chimerism underlie the pathogenesis of juvenile dermatomyositis? J Immunol 172:5041–5046 | PubMed | ISI | ChemPort |
16. Reed AM, Picornell YJ, Harwood A, Kredich DW (2000) Chimerism in children with juvenile dermatomyositis. Lancet 356:2156–2157 | Article | PubMed | ISI | ChemPort |
17. Selva-O'Callaghan A, Boeckh-Behrens TM, Balada-Prades E, Solans-Laque R, Vilardell-Tarres M (2001) Fetal microchimerism and inflammatory myopathies. Lancet 357:887 | Article | PubMed | ChemPort |
18. Srivatsa B, Srivatsa S, Johnson KL, Bianchi DW (2003) Maternal cell microchimerism in newborn tissues. J Pediatr 142:31–35 | Article | PubMed | ISI |
19. Srivatsa B, Srivatsa S, Johnson KL, Samura O, Lee SL, Bianchi DW (2001) Microchimerism of presumed fetal origin in thyroid specimens from women: a case–control study. Lancet 358:2034–2038 | Article | PubMed | ISI | ChemPort |
20. Stevens AM, Hermes HM, Rutledge JC, Buyon JP, Nelson JL (2003) Myocardial-tissue-specific phenotype of maternal microchimerism in neonatal lupus congenital heart block. Lancet 362:1617–1623 | Article | PubMed | ISI |
21. Tomasini D, Tomasini CF, Cerri A, Sangalli G, Palmedo G, Hantschke M et al. (2004) Pityriasis lichenoides: a cytotoxic T-cell-mediated skin disorder. Evidence of human parvovirus B19 DNA in nine cases. J Cutan Pathol 31:531–538 | Article | PubMed | ISI |

Acknowledgments

We thank Caroline Le Danff, for technical help; Sarah Guegan was supported by the Fondation pour la Recherche Médicale, France and the UPRES EA 2396 was funded by Assistance Publique-Hôpitaux de Paris (CRC04026).