Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Donor DNA in a renal cell carcinoma metastasis from a bone marrow transplant recipient

Individuals receiving either allogeneic bone marrow transplants,1 or organ transplants,2 are at an increased risk of developing de novo malignancies. Radiation and immunosuppression are both risk factors. As unfortunate as this is, it could allow for detection of tumor cell–hematopoietic cell hybrids in human cancer. In animal models, tumor hybrids were documented through the use of heterologous genetic markers for the tumor and host genotypes, in some cases revealing hybridization with hematopoietic cells.3, 4 In noncancer systems, heterologous markers were used to demonstrate fusion of bone marrow-derived stem cells with tissue cells.5, 6, 7

We looked for BMT donor DNA in a paraffin-embedded metastasis from a child who, after allogeneic liver and bone marrow transplants, developed renal cell carcinoma, and then metastases.8 The pathology report described this specimen as showing portions of a right caval lymph node involved with metastatic renal cell carcinoma. The primary tumor was unavailable. Both the patient and his liver donor were immunotyped as O+; the BMT donor was A+.8 Laser microdissection, sample preparation, and PCR were as described.9 Two primer pairs were designed to distinguish between A and O alleles (Table 1A).10 The primer pairs were redundant and were used in different reaction tubes. Restriction fragments were generated by KpnI digestion. In all, 14 of the 21 tumor samples were microdissected by a pathologist (RL), and seven under the supervision of a pathologist (FP). For a BMT recipient, any blood lineage cells in the tumor will be of donor genotype; thus, it is important to note that the tumor contained fields in which tumor cells could readily be microdissected free of normal cells. Carcinoma cells were distinguished by their large nuclei as compared to normal cells with smaller nuclei (Figure 1).

Table 1 cDNA sequence of a portion of the blood group A transferase gene and the corresponding 176 bp (1F+2R) and 193 bp (3F+4R) primers used herein
Figure 1
figure1

A histological section of the renal cell carcinoma metastasis studied herein. Large arrowheads: nuclei of malignant cells; small arrowheads: nuclei of normal stromal cells.

Buccal DNA samples from donor and recipient were subjected to PCR with both 176 and 193 bp primers (Table 1A). Following KpnI, amplified donor DNA yielded bands of 176 bp (A allele) and 150 bp (O allele) from the 176 bp primers (Figure 2a, upper); and 193 bp (A allele) and 174 bp (O allele) from the 193 bp primers (Figure 2a, lower). In contrast, KpnI digestion of recipient DNA yielded only 150 and 174 bp bands from the respective primer pairs (O allele). Thus, the donor genotype was A/O, and the recipient was O/O, consistent with the clinical immunotyping.

Figure 2
figure2

(a) Donor/recipient genotyping through analyses of buccal DNA. (Upper) Amplification of 176 bp fragments. (Lower) Amplification of 193 bp fragments. (Left to right): lane 1: 25 bp ladder DNA size markers; lane 2: donor DNA; lane 3: donor DNA after KpnI digestion; lane 4: recipient DNA after KpnI digestion. (b) Tumor DNA genotyping through analyses of microdissected DNA. Tumor DNA fragments were amplified with the 176 bp primer set, and aliquots were either mock digested (−) or digested with KpnI (+) before gel electrophoresis. (Left to right): lanes 1 and 2: tumor sample T2; lanes 3 and 4: tumor sample T6; lane 5: DNA size markers, 100 bp ladder. (c) Tumor DNA fragments were amplified with the 193 bp primer set, and aliquots were either mock digested (−) or digested with KpnI (+) before gel electrophoresis. (Left to right): lane 1: DNA size markers, 100 bp ladder; lanes 2 and 3: tumor sample T5; lanes 3 and 4: tumor sample T6; lanes 6 and 7: tumor sample T19. (d) KpnI digestion of 176 bp-amplified tumor DNA fragments isolated by needle-prick from an agarose gel. The needle-prick samples were re-amplified using the 176-bp primer set, and aliquots were either mock digested (−) or digested with KpnI (+), followed by gel electrophoresis. (Left to right): lane 1: 25 bp ladder DNA size markers; lanes 2 and 3: tumor sample T7; lanes 4 and 5: tumor sample T8; lanes 6 and 7: tumor sample T9; lanes 8 and 9: tumor sample T10; lanes 10 and 11: tumor sample T11; lanes 12 and 13: tumor sample T12; lanes 14 and 15: tumor sample T13; lanes 16 and 17: tumor sample T14. (e) Repeated KpnI digestion of 176 bp generated fragments, to determine whether the initial KpnI digestion had been complete. Previously digested 176 bp DNA fragments were isolated from a gel by needle-prick. This DNA was again amplified with the 176 bp primer set and aliquots were either mock digested (+) or digested a second time with KpnI (++), followed by gel electrophoresis. (Left to right): lane 1: 25 bp ladder DNA size markers; lanes 2 and 3: donor buccal DNA; lanes 4 and 5: tumor sample T2; lanes 6 and 7: tumor sample T3; lanes 8 and 9: tumor sample T4; lanes 10 and 11: tumor sample T5. For needle prick re-amplification of DNA, ethidium bromide-stained fragments were located on the gel by UV light and isolated by inserting a 23 g needle into the center of the band. The needle was briefly immersed in a PCR reaction mixture with a primer set, and PCR was carried out. The mixtures were then re-run on agarose gels with or without KpnI digestion.

Tumor cell DNA was amplified using the 176 bp primers, aliquots were subjected to KpnI digestion, and the mixtures were run on agarose gels. Results for tumor samples T2 and T6 are in Figure 2b. Without KpnI (−), a single 176 bp band was seen, whereas with KpnI (+) two bands of 176 and 150 bp were seen, consistent with being A and O allele fragments. Similarly, amplification of tumor samples T5 and T6 with the 193 bp primers and KpnI digestion yielded bands of 193 and 174 bp, as predicted for A and O alleles (Figure 2c). KpnI digestion of tumor sample T19 produced a 193 bp A allele fragment, but no O allele fragment (Figure 2c).

In a second experiment, tumor samples T7–T14 were amplified with the 176 bp primers and run on an agarose gel without KpnI. The 176 bp fragments were isolated by needle prick, re-amplified with the 176 bp primers, digested with KpnI, and run again on a gel (Figure 2d). After KpnI, all tumor samples showed both 176 bp (A allele) and 150 bp (O allele) fragments. This was confirmed by sequencing of KpnI-generated 176 and 150 bp DNA fragments from tumor DNA sample T8, from the gel in Figure 2d (Table 1B). The sequences of the 176 and 150 bp bands from tumor sample T8 were identical to the published sequences for the A and O alleles, as were the corresponding bands of donor DNA.

Table 2 Sequences of 176 and 150 bp DNA fragments following KpnI digestion

To show that the putative A allele bands were truly KpnI-resistant, and not due to incomplete KpnI digestion, previously digested 176 bp fragments from donor and tumor DNA were isolated from gels via needle prick and re-amplified. Aliquots were either mock-digested or digested a second time with KpnI. The once-digested (+) and twice-digested (++) samples were again run on a gel (Figure 2e). After the second KpnI digestion, donor DNA and tumor DNA samples T2, T3, T4, and T5 showed only the 176 bp product, and no further 150 bp was generated. This indicated that the original KpnI digestion had been complete, and that the remaining 176 bp fragments were truly KpnI resistant, consistent with the tumor cells containing the A allele.

In summary, of 21 tumor DNA samples from diverse areas, 16 yielded primer products. Of these, 16 of 16 contained the A allele, while 14 of 16 contained the O allele as well. PCR reactions also included no-DNA controls as a check against cross-contamination. These were negative. Therefore, the donor A allele was present throughout the metastasis, and the tumor genotype was A/O. The source of the O allele in the tumor cells could not be determined, since it could have originated from the recipient (O/O), the BMT donor (A/O), or the previous liver donor (O/O). (Further information regarding the liver donor was unavailable.8)

Although this is a complex case, we propose BMT-tumor cell hybridization as the most probable mechanism to account for donor DNA in the tumor cells. The alternative, that tumor cells were transferred via the transplant, is not supported by the case histories. That is, since the BMT donor was a healthy child at the time of transplant, and remains healthy more than 10 years later,8 it is improbable that his donated BMT would have contained renal carcinoma cells. Further, the pathology diagnosis indicated the primary tumor to be a de novo renal carcinoma. Although the recipient was a child, due to prior treatment he was indeed at an increased risk of developing de novo malignancies.1, 2, 8 Thus, in our opinion, tumor hybridization best explains the molecular genetic data herein. Continued genetic analyses of pathology specimens from allogeneic BMT patients who develop secondary malignancies would seem to be a useful approach to the question of tumor hybridization in cancer progression.

References

  1. 1

    Ades L, Guardiola P, Socie G . Second malignancies after allogeneic hematopoietic stem cell transplantation: new insight and current problems. Blood Rev 2000; 16: 135–146.

    Article  Google Scholar 

  2. 2

    Sheil AGR . Cancer in immune-suppressed organ transplant recipients: Aetiology and evolution. Transplant Proc 1998; 30: 2055–2057.

    CAS  Article  Google Scholar 

  3. 3

    Pawelek JM . Tumor cell hybridization and metastasis revisited. Melanoma Res 2000; 10: 507–514.

    CAS  Article  Google Scholar 

  4. 4

    Chakraborty A, Sodi S, Rachkovsky M et al. A spontaneous murine melanoma lung metastasis comprised of host × tumor hybrids. Cancer Res 2000; 60: 2512–2519.

    CAS  PubMed  Google Scholar 

  5. 5

    Vassilopoulos G, Wang P-R, Russell DW . Transplanted bone marrow regenerates liver by cell fusion. Nature 2003; 422: 901–904.

    CAS  Article  Google Scholar 

  6. 6

    Wang X, Willenbring H, Akkari Y et al. Cell fusion is the principle source of bone-marrow-derived hepatocytes. Nature 2003; 422: 897–901.

    CAS  Article  Google Scholar 

  7. 7

    Alvarez-Dolado M, Pardal R, Garcia-Verdugo JM et al. Fusion of bone-marrow-derived cells with Purkinje neurons, cardiomyocytes, and hepatocytes. Nature 2003; 425: 968–973.

    CAS  Article  Google Scholar 

  8. 8

    Perkins JL, Neglia JP, Ramsay NKC et al. Successful bone marrow transplantation for severe aplastic anemia following orthotopic liver transplantation: long-term follow-up, outcome. Bone Marrow Transplant 2001; 2: 523–526.

    Article  Google Scholar 

  9. 9

    Persson H, Bäckvall H, Ponten F et al. Single cell gene mutation analysis using laser-assisted microdissection of tissue sections. Meth Enzymol 2002; 356: 334–343.

    CAS  Article  Google Scholar 

  10. 10

    Lee JC-I, Chang J-G . ABO genotyping by polymerase chain reaction. J Forensic Sci 1992; 37: 1269–1275.

    CAS  Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Chakraborty, A., Lazova, R., Davies, S. et al. Donor DNA in a renal cell carcinoma metastasis from a bone marrow transplant recipient. Bone Marrow Transplant 34, 183–186 (2004). https://doi.org/10.1038/sj.bmt.1704547

Download citation

Further reading

Search

Quick links