SEARCH

SEARCH BY CITATION

Keywords:

  • Species-specific polymerase chain reaction;
  • Human;
  • Mouse;
  • Chimera;
  • Stem cell;
  • Metastasis;
  • Quantification

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Studies of the regenerative potential of human stem cells commonly involve their transplantation into immune-deficient mice or in vitro coculture with mouse cells. The optimal use of such models requires the detection and quantification of relatively low numbers of human cells in a murine background. We report here a duplex polymerase chain reaction (PCR) approach involving the coamplification of human-and mouse-specific repetitive sequences. The determination of product ratios compensates against variations in sample quality and enables quantitation from >50% down to 0.01% human-in-mouse from a single reaction. Product ratios are determined by standard electrophoresis of end-stage PCR reactions followed by image analysis techniques using freely available software, with no requirement for real-time PCR. The approach has been used to analyze tissue from mice transplanted with human cells and cocultures between differentiating mouse embryonal stem cells and human umbilical cord blood cells.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The transplantation of human cells into immune-deficient mice is an increasingly common approach in studies of tumor progression [1] and studies of the in vivo regenerative potential of human stem cells [2]. The quality of the information derived from these models depends on the accuracy with which low numbers of donor-derived cells can be detected in host tissues, and a variety of methods have been developed to this end. These include the use of species-specific antibodies or genetic tags such as ß-galactosidase [3], luciferase [4], or fluorescent protein genes [5] to highlight cells in tissue sections, in cell suspensions, and even in living animals [6, 7]. Although cell imaging techniques can provide detailed information down to the single-cell level in a defined location [8, 9], they are not well suited to the widespread and systematic quantitation of low-level contribution throughout a range of tissues and organs.

Species-specific polymerase chain reaction (PCR) offers a more effective means of detecting and quantifying low levels of donor contribution under these conditions, particularly when using primers to repetitive rather than single-copy target sequences. Hence, a primer pair specific to the α-satellite repeat of human chromosome 17 [10] has been used to detect low levels of human hematopoiesis in xenotransplanted nonobese diabetic/ severe combined immunodeficiency (NOD/SCID) mice [11], whereas a real-time PCR modification of this procedure allows accurate quantitation down to 1 in 105 to 106 cells [12]. However, the exclusive use of human primers in these cases can limit the practical applications. First, accurate quantitation requires absolute consistency in the purity and amount of DNA introduced into the reaction. This can be difficult to achieve, particularly when comparing samples extracted from a range of different organs under different conditions. Second, the methods require the use of real-time PCR analysis, limiting their use to laboratories with direct access to this technology.

With the aim of providing a simple, sensitive, and reliable means of human cell quantification in murine tissues, we have developed a duplex PCR approach involving the coamplification of murine and human repetitive sequences and found that a single reaction condition provides quantification of the ratio of human-to-mouse DNA over a wide range without the need for real-time PCR. The method has been used to analyze xenotrans-planted mice and an in vitro model of human stem cell integration into hanging drop cultures of murine embryonal stem (ES) cells, demonstrating superior contribution by the stem cell–enriched CD133+ fraction.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Isolation of Cord Blood Cells

Umbilical cord/placental blood was collected into sterile collection bags (MacoPharma, Langen, Germany, http://www.macopharma.com) after clamping of the umbilical cord, with the informed consent of the mother and with the approval of the local ethics committee of the University of Leipzig. Within 24 hours of collection, the mononuclear cells were separated over two sequential gradients of lymphoprep (Axis-shield, Oslo, Norway, http://www.axis-shield.com) and the CD133+ cells were purified using antibody-coated magnetic beads (Miltenyi Bio-tec, Bergisch Gladbach, Germany, http://www.miltenyibiotec.com). Cells in the flow-through were collected as CD133, and those retained in the column were washed, eluted, applied to a second column, and washed again before final elution. The purity of the selected population was confirmed by staining for CD34 to avoid problems encountered with the efficiency of fluorescent detection of the bead-bound CD133 antigen immediately after purification. More than 95% of the purified cells were CD34+ by fluorescence-activated cell sorter (FACS) analysis (not shown).

Cell Culture

Before the establishment of cocultures, the primary human cells were cultured overnight in StemSpan medium supplemented with stem cell factor, thrombopoietin, and Flt-3 ligand (CellSystems, St. Katharinen, Germany, http://www.cellsystems.de).

The murine FDCPmix [13] and human KG-1a [14] cell lines were maintained as described. The murine ES cell line R1 [15] was maintained on freshly prepared, inactivated, mouse embryonal fibroblast feeder layers supplemented with 1,000 U/ml leukemia inhibitory factor [16].

The embryoid body and pancreatic differentiation cultures were carried out as described [17, 18]. Pluripotent embryonic stem cells R1 were differentiated by cultivation first in hanging drops (n = 600 cells/drop) as embryoid bodies (EBs) (according to Wobus et al. [16]) for 2 days on the lid of a Petri dish filled with sterile phosphate-buffered saline. EBs were then transferred into bacteriological Petri dishes (Greiner Bio-One, Frickenhausen, Germany, http://www.gbo.com/en) and cultivated for 3 days in Iscove's modified Dulbecco's medium (IMDM) containing 20% fetal calf serum (FCS), 2 mM L-glutamine, 1 × nonessential amino acids stock, and 1 × penicillin-streptomycin mixture (all from Invitrogen, Karl-sruhe, Germany, http://www.invitrogen.com). Shortly before use, monothioglycerol (Sigma, St. Louis, http://www.sigmaaldrich.com) was filter-sterilized through a 0.2-μm filter and added to a final concentration of 450 μM. The EBs were plated onto gelatin-coated, 6-cm tissue culture dishes at day 5 and cultivated for an additional 9 days. On days 5 and 9, the EBs were dissociated with a mixture of 0.1% trypsin and 0.08% EDTA (1:1 vol/vol) and the cells replated onto poly-L-ornithin/laminin-coated Petri dishes in differentiation medium containing 10% FCS. On the next day, the medium was exchanged for serum-free differentiation medium and the cells were cultivated for an additional 6–21 days.

On days 5 plus 15, 5 plus 17, or 5 plus 30, the cell clusters were dissociated and the cells were counted in a hemocytometer. Cell suspensions were prepared in which 95% murine ES-derived cells were mixed with 5% human CD133+ or CD133 cells in IMDM containing 20% FCS and additives. Seventy drops of 20 μl each containing 400 cells were set on the lid of a 10-cm Petri dish. After 2 days, the mixed EBs were collected and transferred to 6-cm bacteriological Petri dishes in 8 ml medium. After an additional 3 days, EBs were collected in microcentrifuge tubes and centrifuged (n = 10 to 15 EBs). The medium was discarded and the pellets were frozen rapidly in liquid nitrogen and stored at −80°C until required for DNA analysis.

Transplantation and FACS Analysis

Cryopreserved mononuclear cells from human umbilical cord blood were used for engraftment studies (5 × 106 mononuclear cells per mouse). Before application, the cells were stained with carboxyfluorescein diacetate succinimidyl ester to relocate the transplanted cells in tissues or periphery by flow cytometry and microscopy. The cells were transplanted intravenously (tail vein) 3–5 hours after sublethal irradiation (350 cGy) of 6- to 8-week-old NOD/SCID mice. To improve the human homing/ engraftment, we administered a conditioned supernatant of the human 358/8 cell line every 2 days until day 14 after the cell transplantation. After 8 weeks, the mice were euthanized and the tissues were collected. Human engraftment in spleen cell suspensions was analyzed by flow cytometry with a human-specific CD45-APC antibody (IQ-Products, Groningen, Holland, http://www.iqproducts.nl).

Sample Preparation

Samples were prepared from cultured cells, EBs, and spleen cell suspensions by overnight digestion in 100 μl of PCR buffer containing 100 ng/ml proteinase K. The proteinase K was then inactivated by heating to 95°C for 10 minutes. For each PCR reaction, 10μl was used.

For optimization of the PCR conditions and determination of the sensitivity range, DNA samples were prepared separately from titrations (100% to 0.0001%) of human (KG1a) in murine (FDCPmix) cells.

The DNA standards used for quantitation, representing dilutions of human-in-mouse DNA from 50%–0.001%, were made by mixing stock DNA solutions prepared from human peripheral blood mononuclear cells and from murine tissues using the Genomic Tip 20/G Kit (Qiagen, Hilden, Germany, http://www.qiagen.com).

PCR Primer Design

See Table 1 for primers. The h7A and h7B PCR primers specific for the α-satellite repeat of human chromosome 7 were as described by Warburton et al. [10]. Primers m8F and m8R were designed to amplify a 728-bp fragment from the mouse chromosome 8 gene repeat sequence [19]. The mouse primers correspond to positions 160 through 167 and 887 through 867, respectively, in GenBank sequence M75701. All primers were obtained from Sigma-Genosys, Cambridge, U.K. (http://www.sigma-genosys.com).

Table Table 1.. Polymerase chain reaction primers and major products
TargetPrimerPrimer sequence (5′-3′)Product size (kb)
Human Chr 7 α-satelliteh7AAgCgATTTgAggACAATTgC1.0, 0.68, 0.34 [10]
 h7BCCACCTgAAAATgCCACAgC 
Mouse Chr 8 centromeric repeatm8FCCACATgTCTATCACTTTgCC0.73
 m8RTCATgAgACgAAgTgATTTCC 

PCR Conditions

PCR reactions were set up in a final volume of 40 μl of PCR buffer (50 mM KCl, 20 mM Tris-Cl [pH 8.3], 2.5 mM MgCl2, 0.5% Tween 20) containing 0.5 mM deoxynucleotide triphosphates (Roche, Mannheim, Germany, http://www.roche.com), 0.25 μM of each primer, 2.5 U Taq DNA polymerase (Roche), and either 10 μl of cell lysate or 1 μl (approximately 10 ng) of DNA standards. All PCR reactions were performed in a Gradient Robocycler 96 (Stratagene, La Jolla, CA, http://www.stratagene.com).

Quantitation

The PCR products were separated by electrophoresis on 1.6% agarose gels and stained with ethidium bromide. Nonsaturated digital images were obtained using an EASY Win32 imaging system (Herolab, Wiesloch, Germany, http://www.herolab.com), and band intensities were quantified using the public domain Image J software (http://rsb.info.nih.gov) to determine peak areas.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Primer Design and Species Specificity

Our aim was to establish a duplex PCR assay that provides a measure of the ratio of human-to-mouse DNA over a range of experimental conditions in which the human component is widely variable but generally low and the amount of material available for assay is often limiting. For the sensitive and specific detection of human cells, we used previously described primers that amplify a major product of approximately 1,000 bp (with minor products of approximately 680 and 340 bp) from the highly repetitive alpha satellite repeats of human chromosome 7 [10]. These tandem repeats are present at many thousands of copies per cell and are highly species-specific. To complement the human alpha satellite target, we designed primers to a 728-bp section of a mouse-specific repetitive sequence that is present at 60 to 100 copies on chromosome 8 [19]. The coamplification of a highly repetitive human and a middle-repetitive murine target sequence provides the high level of sensitivity required for the routine processing of small samples while biasing the relative sensitivity of a duplex PCR toward the detection of human cells. Furthermore, the murine primers were designed to have melting temperatures significantly lower than those of the human primers, to allow further adjustment of the relative reaction efficiencies towards one or other species by choosing an appropriate annealing temperature (see below).

To confirm the effects of annealing temperature, a standard preparation of 0.5% human-in-mouse DNA was amplified with the human and mouse primer sets both alone and in combination in a temperature-gradient PCR using annealing temperatures between 53°C and 63°C (Fig. 1). When used alone, the human primers generated the expected products consistently over the entire range of annealing temperatures, with no evidence of nonspecific products at the less-stringent lower temperatures. The murine primers performed well up to an annealing temperature of 62°C, above which the product yield decreased. In the presence of both primer pairs, higher annealing temperatures favored the human reaction and further reduced the yield of mouse product, whereas the lower annealing temperatures favored the murine reaction and reduced the final yield of human product. This apparent competition effect most likely occurs as a consequence of the feedback inhibition or substrate depletion, which marks the end of the logarithmic phase of PCR amplification. In the case of a duplex PCR, this is triggered by the total sum of products, regardless of which of the two reactions (human or mouse) has been most productive.

thumbnail image

Figure Figure 1.. The effects of annealing temperature on amplification efficiency. Amplification products from temperature-gradient poly-merase chain reactions of 0.5% human-in-mouse DNA using primers specific for (A) human chromosome 7 α-satellite sequences, (B) mouse chromosome 8 centromeric repeat sequence, and (C) a mixture of both primer pairs.

Download figure to PowerPoint

Sensitivity and Range of Quantification

The competition effect apparent from Figure 1 suggests that a duplex PCR reaction may on the one hand be less sensitive than a human-only reaction to very low levels of human DNA but on the other hand may provide quantitation (in terms of the ratio of human-to-murine products) over a much wider range. The limits of detection and range of quantitation of the two approaches were therefore compared directly using samples taken from the same dilution series (Fig. 2). The human-only reaction produced a weak signal from 0.001% human-in-mouse DNA after 35 cycles, and the product yield was increased further by extending the reaction to 40 cycles. In comparison, the duplex reaction detected human-in-mouse down to 0.01% at 35 cycles. Since the duplex reaction accumulates more overall product and reaches an end point by this stage, the detection limit is not extended by further cycles (not shown). Under our standard conditions, the detection limit of the duplex reaction therefore lies between 0.001% and 0.01% human-in-mouse and is up to 10-fold less sensitive than the human-only reaction.

thumbnail image

Figure Figure 2.. The limits of detection and range of quantification under various reaction conditions. Samples prepared from titrated mixtures of human and mouse cells were amplified using (A) human primers only at an annealing temperature of 59°C or (B) a duplex mixture of human and mouse primers under conditions that bias toward the murine (annealing at 53°C) or human (annealing at 59°C) products. After electrophoresis and image analysis, the product yield of the (C) human-only polymerase chain reaction (PCR) and the (D) human-to-mouse product ratio of the duplex PCR were plotted against the input ratio on double logarithmic scales to show the relationship over the entire range.

Download figure to PowerPoint

However, the plots of product yield (human-only PCR) or product ratio (duplex PCR) against input DNA reveal a major advantage of the duplex approach, in that the product ratios reflect closely those of the input target DNA over the entire range, from the lowest limit of detection up to 50% human content. Once again, this most likely occurs as a result of either feedback inhibition or substrate exhaustion, which conserves the product ratios at the end of the logarithmic phase of amplification. Regardless of the cycle number at which this point is reached, the information is therefore maintained throughout the subsequent plateau phase. The end-stage analysis of a single PCR reaction condition therefore suffices to yield a measure of human-in-mouse content from 0.01% up to >50%.

Analysis of Transplanted Mice and Human/Mouse Cocultures

Having established the sensitivity and range of quantitation using titrated DNAs, the duplex PCR method was used to analyze two relevant experimental models. First, spleen tissue from four independent NOD/SCID mice transplanted with human cord blood–derived mononuclear cells was analyzed both by FACS and by duplex PCR. To enable statistical analysis, the PCR reactions were performed in triplicate under double-blind conditions. As shown in Figure 3A, the results of the duplex PCR were highly reproducible and in close agreement with those of FACS analysis.

thumbnail image

Figure Figure 3.. Duplex PCR analysis of chimeric mice and cultures. (A): PCR reactions were performed in triplicate (PCR #1-3) using spleen tissue from four transplanted mice (1–4). Mean and standard deviations are shown for the two spleens that fell within the assay range of 0.01%–50% human contribution. The corresponding coefficients of variation were 16.1% (mouse 3) and 11.8% (mouse 4). (B): Mixed cultures of CD133+ or CD133 human umbilical cord blood cells (5%) and mouse ES cells (95%) were established using either undifferentiated ES cells or ES-derived cells from various stages of a pancreatic differentiation culture (5 plus 16, 5 plus 27, and 5 plus 30 days). After an additional 5 days of coculture, cell lysates were analyzed by duplex PCR (35 cycles, annealing at 59°C). The percent human DNA content of mixed cultures calculated from the scanned gels is shown above each lane. Standard: DNA standards run in parallel. Abbreviations: ES, embryonal stem; FACS, fluorescence-activated cell sorter; PCR, polymerase chain reaction.

Download figure to PowerPoint

One of the original reasons for developing a duplex PCR based on repetitive sequences was to facilitate the rapid and reliable screening of large numbers of small-scale cocultures of murine and human cells. In one such approach, cells purified from human umbilical cord blood were mixed at a frequency of 5% either with undifferentiated mouse ES cells or with ES-derived cells from various stages of a pancreatic differentiation [17, 18]. After 5 days of coculture, individual cultures were harvested and cell lysates were introduced directly into the duplex PCR assay. The results (Fig. 3B) show that whereas CD133+ human cord blood cells are capable of making a detectable contribution to hanging drop cultures of both undifferentiated and differentiating ES cells, the CD133 population contributes relatively poorly to EBs established from undifferentiated ES cells and only rarely to those established from the later stages of pancreatic differentiation.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

We report the development of a PCR technique for the simple and reliable quantitation of human cells in a mouse background. Our determinations are based not on the absolute yield of human product but on the ratio of coamplified human and mouse products from a duplex PCR reaction. A previously reported technique based on the real-time duplex PCR analysis of single-copy target sequences allows quantitation down to 2% human-in-mouse [20]. The technique reported here is based on repetitive targets and enables quantitation down to 0.01%. This corresponds to less than one human genome in the sample size used, presumably due to breakage of the large, tandemly repetitive α-satellite region during sample preparation. Furthermore, the duplex PCR method uses standard reaction conditions, gel electrophoresis, and image analysis using public domain software, making it easily accessible to most laboratories.

In a duplex PCR reaction, the product ratio established during the logarithmic phase of amplification is maintained into the subsequent plateau phase as a result of the feedback inhibition or substrate exhaustion, affecting both products similarly. This produces two major advantages. First, a single end-point analysis of the reaction products is sufficient to quantify human contribution from the lower detection limit (in these experiments, 0.01%) right up to levels of >50% (the highest content tested). Second, the duplex reaction is robust to variations in the quantity or quality of target DNA because these would affect both the murine and human components similarly and therefore have little or no effect on the final product ratios (Fig.4). However, it should be noted that variations in the quantity or quality of target would be expected to affect the overall progress of the duplex reaction. Since the reaction efficiencies of the human and mouse amplifications may differ, it is possible that the product ratio changes during the log phase (Fig. 4). For this reason, it would be unwise to analyze the reaction products before the plateau phase has been reached.

thumbnail image

Figure Figure 4.. Expected variation of the product ratio during logarithmic amplification. Using an annealing temperature of 59°C, the human target sequence (in black) is amplified more efficiently than that of the mouse (in gray). Because the product accumulation curves are not parallel, the product ratio varies throughout the logarithmic phase. Measurements made during this period (point 1) would therefore be sensitive to differences in the quality and quantity of samples and standards (compare continuous and dashed lines). In contrast, such variations have little effect on the final ratio of accumulated products at the end point of the reaction (points 2 and 3).

Download figure to PowerPoint

The duplex PCR assay described here provides a simple means of determining the contribution made to various tissues of immune-deficient mice in xenotransplantation studies either of the metastatic potential of human cancer cells or of the regenerative potential of human stem cells as has been analyzed here. It is important to stress that this DNA-based assay provides no information concerning the phenotype of the cells detected, nor does it distinguish between fused and nonfused cells. However, it does provide a rapid and quantitative screen with which different stem cell populations, experimental conditions, and target tissues may be compared objectively on a large scale. Given the questions currently surrounding the regenerative potential of stem cells from different sources, the low frequencies of donor contribution to solid tissues and the potential influence of ex vivo culture conditions on regenerative potential, quantitative, large-scale comparisons of this nature are likely to provide a useful introduction to more detailed studies of stem cell fates under specific conditions. Indeed, the duplex PCR assay reported here was developed specifically to enable the screening of large numbers of chimeric EBs in in vitro assays comparing the regenerative potential of candidate stem cell populations. The demonstrated contribution of human umbilical cord blood CD133+ cells to various stages of a murine ES cell pancreatic differentiation culture demonstrates the potential benefits of combining these approaches.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

We describe a duplex PCR technique in which human and murine-specific, repetitive sequences are coamplified in a single reaction. Determination of the ratio of human-to-mouse products rather than the absolute amount of human material compensates for variations arising during sample preparation. Because the proportion of human-to-mouse products is maintained into the plateau phase of the PCR reaction, there is no need to use real-time PCR technology, a single reaction being sufficient to quantify human contribution from <0.01% to >50%. This provides a simple, sensitive, and reliable indication of the frequency of human cells in a murine background that should be of widespread use to studies of human stem cell potential both in vivo and in vitro. In this study, the technique has been used to quantify human cells in xenotransplanted mice and to demonstrate the contribution of human CD133+ cord blood stem cells to differentiation cultures established from murine ES cells.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

This work was supported by grants from the Bundesministerium für Bildung und Forschung (01GN0105 and 01GN0106) and by the Interdisciplinary Centre for Clinical Research (IZKF) at the University of Leipzig (Project N02).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References