Endogenous murine leukemia virus DNA sequences in murine cell lines: implications for gene therapy safety testing by PCR

Proteomic profiling of human embryonic stem cell-derived microvesicles reveals a risk of transfer of proteins of bovine and mouse origin

BACKGROUND AIMS

Microvesicles (MV) shed from the plasma membrane of eukaryotic cells, including human embryonic stem cells (hESC), contain proteins, lipids and RNA and serve as mediators of cell-to-cell communication. However, they may also contain immunogenic membrane domains and infectious particles acquired from xenogenic components of the culture milieu. Therefore, MV represent a potential risk for clinical application of cell therapy.

METHODS

We tested the ability of hESC and their most commonly used feeder cells, mouse embryonic fibroblasts (MEF), to produce MV. We found that hESC are potent producers of MV, whereas mitotically inactivated MEF do not produce any detectable MV. We therefore employed a combined proteomic approach to identify the molecules that constitute the major components of MV from hESC maintained in a standard culture setting with xenogenic feeder cells.

RESULTS

In purified MV fractions, we identified a total of 22 proteins, including five unique protein species that are known to be highly expressed in invasive cancers and participate in cellular activation, metastasis and inhibition of apoptosis. Moreover, we found that hESC-derived MV contained the immunogenic agents apolipoprotein and transferrin, a source of Neu5Gc, as well as mouse retroviral Gag protein.

CONCLUSIONS

These findings indicate that MV represent a mechanism by which hESC communicate; however, they also serve as potential carriers of immunogenic and pathogenic compounds acquired from environment. Our results highlight a potential danger regarding the use of hESC that have previously been exposed to animal proteins and cells.

In utero injection of alpha-L-iduronidase-carrying retrovirus in canine mucopolysaccharidosis type I: infection of multiple tissues and neonatal gene expression

Canine alpha-L-iduronidase (alpha-ID) deficiency is caused by a single base pair mutation in the alpha-ID gene, resulting in no enzyme activity in homozygous affected pups. The disease clinically resembles human mucopolysaccharidosis type I (MPSI). We used the canine MPSI model system to address the efficacy of a new retroviral vector, MND-MFG, containing the human alpha-ID cDNA (MND-MFG-alpha-ID) for direct in utero gene delivery to MPSI cells. In vitro, the MND-MFG-alpha-ID vector showed high-level, long-term expression of the transgene in both canine and human alpha-ID-deficient fibroblasts. The effectiveness of this vector for in utero gene transfer and expression in multiple tissues was assessed by injecting viral supernatants into MPSI fetuses and evaluating transduction efficiency and enzyme expression at various times after birth. Transduction of a spectrum of cell types and tissues was observed in all seven live-born pups and in one stillborn pup. Although enzyme activity was not detected in adult tissues from the seven surviving pups, significant alpha-ID enzyme activity was detected in both the liver and kidney of the deceased pup. Our combined gene delivery vector and in utero transfer approach, while encouraging in terms of overall gene transfer efficiency to multiple tissues and successful short-term gene expression, was unable to meet the important requirement of sustained in vivo gene expression.

Synthesis of virus-specific high-mobility DNA after temperature upshift of SC-1 cells chronically infected with moloney murine leukemia virus mutant ts1

Premature termination products of reverse transcription that consist physically of viral minus-sense single-stranded DNA that is shorter than one long terminal repeat and partial DNA duplexes are dramatically increased in the central nervous system (CNS) of FVB/N mice that are infected by ts1, a temperature-sensitive mutant of Moloney murine leukemia virus. Due to their migration in agarose gels, these incomplete physical forms of DNA have been designated high-mobility (HM) DNA. In non-CNS tissues, the level of HM DNA is either low or not detectable. In order to determine the conditions that are necessary for the synthesis of HM DNA in vivo, we have characterized the physical forms of HM DNA that were synthesized in vitro in chronically infected SC-1 cells after temperature upshift. At the permissive temperature of 34 degrees C, the chronically infected SC-1 cells did not synthesize HM DNA. After temperature upshift of the cultured cells from 34 to 37 degrees C, the chronically infected SC-1 cells developed extremely high levels of HM DNA. Following temperature downshift of the cultured cells from 37 to 34 degrees C, a decrease in the level of HM DNA and an increase in the level of unintegrated linear proviral DNA occurred simultaneously. These results suggested that the accumulation of HM DNA both in vitro and in vivo may be the result of superinfection.

Engraftment of gene-marked hematopoietic progenitors in myeloma patients after transplant of autologous long-term marrow cultures

We conducted a phase I hematopoietic stem cell (HSC) gene-marking trial in patients undergoing autologous blood or marrow stem cell transplant for the treatment of multiple myeloma. Between 500 and 1000 ml of bone marrow was harvested from each of 14 myeloma patients and 1 syngeneic donor. A mean of 3.3x10(9) cells per patient were plated in 20 to 50 long-term marrow culture (LTMC) flasks and maintained for 3 weeks. LTMCs were exposed on days 8 and 15 to clinical-grade neo(r)-containing retrovirus supernatant (G1Na). A mean of 8.23x10(8) day-21 LTMC cells containing 5.2x10(4) gene-marked granulocyte-macrophage progenitor cells (CFU-GM) were infused along with an unmanipulated peripheral blood stem cell graft into each patient after myeloablative therapy. Proviral DNA was detected in 71% of 68 tested blood and bone marrow samples and 150 of 2936 (5.1%) CFU-GM derived from patient bone marrow samples after transplant. The proportion of proviral DNA-positive CFU-GM declined from a mean of 9.8% at 3 months to a mean of 2.3% at 24 months postinfusion. Southern blots of 26 marrow and blood samples were negative. Semiquantitative PCR analysis indicated that gene transfer was achieved in 0.01-1% of total bone marrow and blood mononuclear cells (MNCs). Proviral DNA was also observed in EBV-transformed B lymphocytes, in CD34+ -enriched bone marrow cells, and in CFUs derived from the latter progenitors. Gene-modified cells were detected by PCR in peripheral blood and bone marrow for 24 months after infusion of LTMC cells. Sensitivity and specificity of the PCR assays were independently validated in four laboratories. Our data confirm that HSCs may be successfully transduced in stromal based culture systems. The major obstacle to therapeutic application of this approach remains the overall low level of genetically modified cells among the total hematopoietic cell pool in vivo.

Gene therapy for canine alpha-L-iduronidase deficiency: in utero adoptive transfer of genetically corrected hematopoietic progenitors results in engraftment but not amelioration of disease

Canine alpha-L-iduronidase (iduronidase) deficiency is a model of the human lysosomal storage disorder mucopolysaccharidosis type I (MPS I). We used this canine model to evaluate the therapeutic potential of hematopoietic stem cell (HSC) gene therapy for enzyme deficiencies. In previous studies, iduronidase-deficient dogs infused with autologous marrow cells genetically modified to express iduronidase had long-term engraftment with provirally marked cells, but there was no evidence of proviral iduronidase expression or clinical improvement. The presence of humoral and cellular immune responses against iduronidase apparently abrogated the therapeutic potential of HSC gene therapy in these experiments. To evaluate HSC gene therapy for canine MPS I in the absence of a confounding immune response, we have now performed in utero adoptive transfer of iduronidase-transduced MPS I marrow cells into preimmune fetal pups. In three separate experiments, 17 midgestation fetal pups were injected with 0.5-1.5 x 10(7) normal or MPS I allogeneic long-term marrow culture (LTMC) cells transduced with neo(r)- or iduronidase-containing retroviral vectors. Nine normal and three MPS I pups survived the neonatal period and demonstrated engraftment of provirally marked progenitors at levels of up to 12% for up to 12 months. However, the proportion of provirally marked circulating leukocytes was approximately 1%. Neither iduronidase enzyme nor proviral-specific transcripts were detected in blood or marrow leukocytes of any MPS I dog. Humoral immune responses to iduronidase were not detected in neonates, even after "boosting" with autologous iduronidase-transduced LTMC cells. All MPS I dogs died at 8-11 months of age from complications of MPS I disease with no evidence of amelioration of MPS I disease. Our results suggest that iduronidase-transduced primitive hematopoietic progenitors can engraft in fetal recipients, contribute to hematopoiesis, and induce immunologic nonresponsiveness to iduronidase in MPS I dogs. However, the therapeutic potential of HSC gene transfer in this model of iduronidase deficiency appears to be limited by poor maintenance of proviral iduronidase gene expression and relatively low levels of genetically corrected circulating leukocytes.

A replication-competent retrovirus arising from a split-function packaging cell line was generated by recombination events between the vector, one of the packaging constructs, and endogenous retroviral sequences

Previously we reported the presence of a replication-competent retrovirus in supernatant from a vector-producing line derived from a widely used split-function amphotropic packaging cell line. Rigorous routine screening of all retroviral stocks produced in our laboratory has not, previously or since, indicated the presence of such a virus. Replication-competent retroviruses have never previously been used in our laboratory, and stringent screening of all routinely used cell lines has not revealed the presence of any helper viruses. Therefore, it is highly unlikely that this virus represents an adventitious cross-contaminant or had been imported unknowingly with our cell line stocks. PCR studies with DNA from infected cell lines and Northern blot analysis and reverse transcriptase PCR with RNA from infected cells suggest that the helper virus arose by recombination events, at sites of partial homology, between sequences in the vector, one of the packaging constructs, and endogenous retroviral elements. These recombinations were not present in stocks of the packaging cell line or in an initial stock of the vector-producing line, indicating that these events occurred while the vector-producing line was being passaged for harvest of supernatant stocks.