| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
and 
Immunoglobulin Light Chain Expression
Departments of
1
Molecular Biotechnology and
3
Mathematics, Chalmers University of Technology, 405 30 Gothenburg, Sweden.
2 Department of Pathology, Lundberg Laboratory for Cancer Research, Gothenburg University, 413 45 Gothenburg, Sweden.
4 TATAA Biocenter, Chalmers University of Technology, 405 30 Gothenburg, Sweden.
aAddress correspondence to this author at: TATAA Biocenter, Medicinarg. 9C, Chalmers University of Technology, 405 30 Gothenburg, Sweden. Fax 46-31-7733948; e-mail mikael.kubista{at}tataa.com.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
:IgL
ratio has long been used as a clinical criterion for non-Hodgkin B-cell lymphomas. As a first step toward a quantitative real-time PCR-based multimarker diagnostic analysis of lymphomas, we have developed a method for determination of IgL:IgL ratio in clinical samples.
Methods: Light-up probe-based real-time PCR was used to quantify IgL and IgL cDNA from 32 clinical samples. The samples were also investigated by routine immunohistochemical analysis and flow cytometry.
Results: Of 32 suspected non-Hodgkin lymphoma samples analyzed, 28 were correctly assigned from real-time PCR measurements assuming invariant PCR efficiencies in the biological samples. Four samples were false negatives. One was a T-cell lymphoma, one was a diffuse large B-cell lymphoma, and one was reanalyzed and found lymphoma-positive by in situ calibration, which takes into account sample-specific PCR inhibition. Twelve of the samples were fine-needle aspirates, and these were all correctly assigned.
Conclusions: This work is a first step toward analyzing clinical samples by quantitative light-up probe-based real-time PCR. Quantitative real-time PCR appears suitable for high-throughput testing of cancers by measuring expression of tumor markers in fine-needle aspirates.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
The vast majority of NHLs develop from a B-lymphocyte lineage. B-lymphocytes produce immunoglobulins with a heavy chain and either a (IgL) or a (IgL) light chain. Which light chain will be produced is determined early in the development of each B-lymphocyte. In healthy humans, 
60% of B-cells produce chains and the rest produce chains. Normal lymphoid tissues therefore contain a mixture of B-cells with an IgL:IgL ratio of 60:40 (1)(2).
Lymphomas, like all malignant tumors, are clonal and arise from one transformed cell. Lymphoma tissues are dominated by the tumor cells; consequently, the IgL:IgL ratio is changed. With -producing tumors, the IgL:IgL ratio is higher, whereas with -producing tumors, the ratio is lower. Assuming that the translation efficiency and stability of the IgL and IgL mRNAs are similar, clonality may be detected by measuring the IgL:IgL expression ratio.
Real-time PCR can detect and quantify target nucleic acids in biological samples with light-up probes (3)(4)(5). We have used this approach to assess the expression of IgL and IgL in B-cell lymphoid tumors. Although it cannot yet replace traditional immunohistochemistry (IHC) and flow cytometry in diagnosis, this assay, which takes into account reaction efficiency, demonstrates the potential of quantitative reverse transcription-PCR in cancer diagnostics.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
and chains (IHC); and in some cases, flow cytometry. The samples were classified as lymphadenitis or malignant lymphoma according to the Revised European-American Lymphoma classification system (6).
rna extraction and cDNA SYNTHESIS
RNA was extracted by use of the Fast Prep System (FastRNA Green; Qbiogene). We mixed 10 µg of total RNA with 2 µg of poly(dT) oligomers (Pharmacia) and incubated the mixture at 65 °C for 5 min. First-strand cDNA synthesis was then performed by adding 0.05 mol/L Tris-HCl (pH 8.3), 0.075 mol/L KCl, 3 mmol/L MgCl2, 0.01 mol/L dithiothreitol, 10 U/mL Moloney murine leukemia virus reverse transcriptase (Life Technologies), 0.05 units/mL RNA guard (Life Technologies), and 10 mmol/L of each deoxyribonucleotide (Life Technologies) to a final volume of 20 µL and incubating the samples at 37 °C for 1 h. The reaction was terminated by incubation at 65 °C for 5 min, and samples were stored at -80 °C.
light-up probes
Two homopyrimidine light-up probes, H-CCTTTTTCCC-NH2 (IgLLUP) and CCTCCTCTCT-NH2 (IgLLUP), directed against PCR amplification products of the constant regions in the human IgL and IgL light chains, respectively, were designed. Both probes are homopyrimidine sequences, which are known to exhibit very large signal enhancement on target binding (7). Both probes had the thiazole orange derivative, N-carboxypentyl-4-[(3'-methyl-1',3'-benzothiazol-2'-yl)methylenyl]quinolinium iodide, as label attached to the peptide nucleic acid. They were synthesized by solid-phase synthesis and purified twice by reversed-phase HPLC as described previously (8). Probe concentrations were determined spectroscopically assuming molar absorptivities at 260 nm of 83 100 M-1 cm-1 for IgLLUP and 81 100 M-1 cm-1 for IgLLUP (8). The probes were designed to have melting temperatures of 65–70 °C, which is between the annealing (55 °C) and elongation (74 °C) temperatures of the PCRs.
pcr products
PCR products were purified with a QIAquickTM PCR purification reagent set (Qiagen), and their concentrations were determined spectroscopically, assuming a molar absorptivity at 260 nm of 13 200 M-1 cm-1 per base pair (9). Primer (Medprobe Inc.) concentrations were estimated assuming: 
260/103 = 12.0 · nG + 7.1 · nC + 15.2 · nA + 8.4 · nT M-1 cm-1, where nX is the total number of base X (9).
real-time pcr
PCR systems were designed for a 231-bp fragment of human IgL (GenBank accession no. AK024974) and a 223-bp fragment of human IgL (GenBank accession no. X51755), which comprise the IgLLUP and IgLLUP target sequences, respectively.
Reaction conditions were optimized as described elsewhere (10). IgL and IgL PCRs both contained 75 mM Tris (pH 8.8), 20 mM (NH4)2SO4, 1 mL/L Tween 20, 1 U of JumpStartTM Taq DNA polymerase (with antibody; Sigma-Aldrich), and 200 ng/µL bovine serum albumin (Fermentas). Specific components for the IgL PCR were 5 mM MgCl2, 0.2 mM deoxyribonucleotides (Sigma-Aldrich), 800 nM each of the primers (MedProbe Inc.), and 800 nM IgLLUP; for the IgL PCR, specific components were 3.5 mM MgCl2, 0.4 mM deoxyribonucleotides, 600 nM each of the primers, and 600 nM IgLLUP. Primer sequences for IgL were 5'-TGA GCA AAG CAG ACT ACG AGA-3' (forward) and 5'-GGG GTG AGG TGA AAG ATG AG-3' (reverse); for IgL, primer sequences were 5'-GAG CCT GAC GCC TGA G-3'(forward) and 5'-ATT GAG GGT TTA TTG AGT GCA G-3' (reverse).
Real-time PCR was performed in a LightCycler (Roche Diagnostics) using the thermocycler program: 3 min of preincubation at 95 °C followed by 50 cycles for 0 s at 95 °C, 10 s at 55 °C, and 11 s at 74 °C. Fluorescence was monitored (excitation at 470 nm and emission at 530 nm) at the end of the annealing phase (the LightCycler F1 channel). All amplification curves were baseline-adjusted by subtracting the arithmetic average of the five lowest fluorescence read-out values in each sample (arithmetic baseline adjustment in the LightCycler software). The threshold was set to a value of 1.00, which was significantly above background noise, and the number of cycles required to reach this values, CT, was determined (11).
model for the quantification of relative expression of two genes
A mathematical model was developed to determine the ratio of the expression of two genes by real-time PCR. The model is general and applied here to the IgL and IgL genes. The basic equation describing real-time PCR amplification is:
 | (1) |
 | (2) |
 | (3) |
and IgL genes is obtained as:
 | (4) |
 | (5) |
 | (6) |
 | (7) |
= IIgL. Equating Eq. 5
with Eq. 7
and rearranging we obtain:
 | (8) |
and 6
and rearranging we get:
 | (9) |
and CTIgL are the CT values obtained from the PCR amplifications of the IgL and IgL cDNAs; EIgL and EIgL are the efficiencies of the two PCRs, which are obtained as slopes from plots of CT vs log N0 in dilution series of the samples; and KRS is the relative sensitivity constant of the two PCR assays, determined using test samples with known cDNA concentrations.
The fractions of IgL and IgL mRNA, expressed as percentages, are:
 | (10) |
 | (11) |
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
:IgL expression ratio or positive with a deviating expression ratio, we must know with what accuracy CT can be determined. We therefore designed experiments to measure the variation in CT attributable to experimental error and biological variability. We first studied the reproducibility of the PCR by splitting a sample into aliquots that were analyzed in parallel runs (intraassay). We then also included variation attributable to sample handling by analyzing the same sample in independent runs (interassay). To minimize variation in template concentration between the two assays being compared, we prepared a master mixture containing template and all common PCR components and split it into two aliquots to which the unique components for the IgL and the IgL reactions were added. Each experiment was performed eight times with patient sample BR0 (Fig. 1
).
|
In most reports, PCR reproducibility is expressed as the SD in CT. The variance, SD2, is:
 | (12) |
CT
is the mean of the measured CT, and SD is the square root of the variance. However, because we are interested in determining the amount of cDNAs in the sample, the SD of (1 + E)-CT, which is proportional to the number of cDNA molecules (Eq. 1
):
 | (13) |
 | (14) |
(1 + E)-CT is the mean of (1 + E)-CT. To obtain the relative uncertainty in the number of cDNA molecules, we normalize the SD with the average value to obtain the CV, which we express as a percentage:
 | (15) |
The intraassay CV, which reflects the reproducibility of the PCR, was 3.0% for the IgL reaction and 4.9% for the IgL reaction (Table 1
). The interassay CVs, to which experimental errors also contribute, were only slightly larger: 8.1% for the IgL reaction and 5.0% for the IgL reaction. Although it was not possible to calculate a CV for the ratio of the two cDNAs, we can estimate how much the IgL:IgL expression ratio in a negative sample could deviate from 60:40 because of experimental uncertainty in a bad case {CV(a/b) is in general not equal to [CV(a)2 + CV(b)2]1/2, as is frequently inferred in PCR user bulletins}. If we suppose that the number of IgL cDNA molecules is overestimated, because of experimental error, by 2 SD and that the number of IgL cDNA molecules is also underestimated by 2 SD, then the measured ratio would be (60/40) x (1 + 0.162)/(1 - 0.100) = 1.94 = 66/34. If instead the amount of IgL cDNA is underestimated and that of IgL cDNA is overestimated, then the measured ratio would be (60/40) x (1 - 0.162)/(1 + 0.100) = 1.14 = 53/47. Hence, because of experimental uncertainty and variation in PCR efficiency attributable to the added components, we expect negative samples to display an IgL:IgL expression ratio of 53:47 < N0IgL:N0IgL < 66:34.
|
determination of pcr efficiencies in patient samples
Biological samples frequently contain PCR inhibitors, such as heme (12), heparin (13), and IgG (14), and inhibition may vary substantially among samples (15)(16). To correctly interpret results of real-time PCR measurements, it is crucial to estimate the PCR efficiency with high precision in the particular biological sample studied; even a small error in the assumed PCR efficiency may cause large errors in the estimated number of cDNA molecules. We have therefore designed an approach to determine the efficiencies of PCR reactions in situ in biological samples. By diluting the test sample in steps and measuring the CT value at each dilution, we could construct an intrinsic calibration curve from which the PCR efficiency could be determined (Fig. 2
). We chose to dilute the samples 64 times, in three fourfold dilution steps. The dilutions were performed in duplicate, and the CT values were measured for both the IgL and IgL reactions to determine the efficiencies of the two assays separately. Seven patient samples, four negative and three positive, were characterized this way, as well as purified template that should not contain any inhibitors.
|
The PCR efficiencies obtained when we amplified purified template were EIgL = 95% and EIgL = 93%, signifying that both reactions proceeded with very high efficiencies as expected for optimized PCR assays. Six of the patient samples exhibited efficiencies that were 10% lower; the IgL PCR efficiency was 75% < EIgL < 86% with EIgL = 79%, and the IgL efficiency was 79% < EIgL < 91% with EIgL = 85% (Table 2
). The seventh sample, BR17, exhibited typical IgL efficiency (83%), whereas the IgL efficiency was only 59%. The reason for the extremely low efficiency of the IgL reaction in this sample is unclear. It was considered an outlier and was not included in the calculation of average efficiencies.
|
When comparing the yields of two reactions, the efficiency ratio:
 | (16) |
). For the six samples, 1.01 < XER < 1.07 with XER = 1.03 (Table 2
). Hence, after 25 amplification cycles, which was typically required to reach threshold with the patient samples (Fig. 1
), twice (1.0325 = 2) as many DNA molecules had been formed compared with DNA because of the difference in PCR efficiencies.
Finally, to relate the measured CT values of the two real-time PCRs to the ratio between the numbers of corresponding cDNA molecules, we must also determine the relative sensitivity, KRS, of the two probing systems (Eq. 8
).
 | (17) |
and IgL calibrators. For our purpose, however, it is better to calculate KRS from the CT values (CTIgL and CTIgL) and PCR efficiencies (EIgL and EIgL) determined for the four negative samples (Table 2
), assuming an IgL:IgL expression ratio of 60:40. The reason is that the assumption of a particular IgL:IgL expression ratio will cancel when calculating the relationship between CTIgL and CTIgL for negative samples (see below). This gave 1.4 
KRS 1.9 with KRS = 1.5 (Table 2
). Hence, assuming an IgL:IgL expression ratio of 60:40, the probing of IgL DNA is 50% more sensitive than the probing of IgL DNA with the probes and conditions here.
characterization of patient samples
A total of 32 patient samples were analyzed for B-cell lymphoma by the Q-PCR assay. Twelve of the samples were collected as fine-needle aspirates. All samples were run in duplicate, including negative controls. The data are summarized in Table 3
and in Table 1
of the Data Supplement (available with the online version of this article athttp://www.clinchem.org/content/vol49/issue1/) and are plotted in Fig. 3
. In the plot, each symbol represents one sample and is positioned on the coordinates CTIgL, CTIgL. The corresponding number of cDNA molecules of purified template, calculated assuming EIgL = 95% and EIgL = 93%, is indicated in logarithmic scale on the opposite axes. Samples considered negative by IHC analysis are shown as circles, and positive samples are shown as squares.
|
|
Negative samples with an IgL:IgL gene expression ratio of 60:40 are expected to lie on a straight line. Rewriting Eq. 9
, we get:
 | (18) |
 | (19) |
 | (20) |
and CTIgL with slope b and intercept m. Inserting EIgL = 0.85, EIgL = 0.79, and KRS = 1.5, which are the average values determined for the six samples above (Table 2
), and 
= 60:40 = 1.5, we obtain b = 0.95 and m = 0.021. Note that the relative sensitivity, KRS, was calculated from measurements on negative samples assuming a 60:40 expression ratio (Eq. 17
). This cancels the 
ratio in the second term. Hence, the calculated slope and intercept of the relationship between CTIgL and CTIgL for negative samples is independent of the assumption of a particular IgL:IgL expression ratio. A line with b = 0.95 and m = 0.021 is shown in Fig. 3
.
Some negative samples were slightly off the line representing 60:40 expression (Fig. 3
). This may be attributable to variations in PCR efficiencies among the samples. Such variations can cause an error in the estimation of the number of cDNA molecules from the measured CT values when mean PCR efficiencies are assumed. If the efficiencies of the two PCR assays in a sample deviate from the mean values to about the same degree, the measured CT values will still correctly reflect the expression ratio and negative samples will fall on the 60:40 line, although they will be displaced diagonally from where they would be if their efficiencies were EIgL and EIgL. However, if the efficiency of one of the reactions deviates more than the other from the mean values, a negative sample may be off the 60:40 line. For the seven samples characterized by in situ calibration (Table 2
), the measured CT values could be corrected for the differences between their specific PCR efficiencies and the mean efficiencies:
 | (21) |
with open symbols, and they are connected to the measured CT values by arrows. Indeed, most arrows, sample BR17 being the only exception, are diagonal, indicating that the two reactions are inhibited to the same degree, which does not affect classification.
To account for experimental error and variations in PCR efficiencies in classification of samples, we estimated limits within which negative samples should be found. Keeping the intercept fixed in Eq. 20
:
 | (22) |
)/log (1 + EIgL), from the efficiencies determined for the six samples (BR17 was excluded) characterized by in situ calibration. This gave an SD of 0.031. For a gaussian distribution, the 95% confidence region is given by the mean ± 1.96 · SD. In Fig. 3
, the dashed lines enclose the region defined by:
 | (23) |
:IgL expression ratio among healthy individuals. These factors would broaden the confidence region further. Hence, the region indicates where negative samples are expected to be found with at least 95% probability. All negative samples in this study fell within this region (Fig. 3
).
Positive samples with IgL clonality were below the 60:40 line, whereas those with IgL clonality were above it. This separation of clonal and normal samples was as good for fine-needle aspirates as for the surgical biopsies, although the former contained less material and had higher CT values. Most positive samples fell outside the confidence region. However, samples BR5 and BR17 were within the region although they were positive. BR17 had anomalous IgL PCR efficiency, and when corrected for that, it fell outside the region. We do not know why sample BR5 was within the region. It may have also been attributable to anomalous PCR efficiencies, or it could have been the result of limited clonality. Three samples, indicated * by in Fig. 3
, were classified as diffuse large B-cell lymphomas by IHC analysis. One of them, BR29, was close to the 60:40 line. The other two were correctly assayed as positive; one with IgL clonality and the other with IgL clonality. BR23 was a T-cell lymphoma, as judged by IHC analysis, and the sample therefore exhibited a normal IgL:IgL ratio. An interactive Excel sheet to analyze IgL and IgL expression data has been made available as part of the Data Supplement accompanying the online version of this article (http://www.clinchem.org/content/vol49/issue1/) and athttp://www.molbiotech.chalmers.se/research/mk/anders/index.htm.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
and IgL mRNA. The ratio between IgL- and IgL-producing cells in healthy individuals was 60:40, and we expect a similar IgL:IgL ratio for negative samples in our real-time PCR assay. In lymphoma-derived samples, the IgL:IgL ratio is expected to deviate substantially from 60:40 because of clonal expansion. In contrast to most other assays based on relative gene expression measurements, this one does not rely on comparison with the expression of housekeeping genes, which has been seriously questioned (17)(18)(19).
pcr in biological samples is substantially inhibited and the degree of inhibition varies among samples and depends on template
In most studies of gene expression by Q-PCR, the CT values determined for test samples have been related to CT values of standard samples containing known amounts of cDNA. This approach relies on the crucial assumption that PCR efficiencies in the test and standard samples are the same. If this is not true, a CT value measured in a test sample will correspond to a different number of cDNA copies than the same CT value measured in the standard sample. The error introduced by such an assumption may be substantial because of accumulation effects. For example, 80% efficiency in the test sample and 85% efficiency in the standard sample produces a 50% difference in the number of DNA copies after 25 cycles (Eq. 1
). The common procedure to account for differences in PCR efficiencies between test and standard samples is to amplify a reference gene, usually a housekeeping gene, in parallel and to relate the expression of the gene being studied to the expression of the housekeeping gene. This, of course, relies on the assumption that expression of the housekeeping gene is constant among the samples being compared, which has also been questioned (17)(18)(19). Furthermore, which is rarely acknowledged, it also assumes that the efficiencies of the two reactions, i.e., the PCR of the studied gene and the PCR of the housekeeping gene, are inhibited to the same degree in the standard sample as well as in the test sample (20):
 | (24) |
The PCR efficiencies of both reactions in seven samples and of purified template were determined in situ from CT values measured in dilution series. The total dilution was 64 times, which was sufficient to make the effect of experimental variation in CT values on the slopes of the CT-vs-log(cDNA concentration) plots negligible. Significant PCR inhibition in the biological samples was evident from the significantly higher PCR efficiencies obtained with purified template compared with those obtained in patient samples: purified template had efficiencies >93% compared with efficiencies in the range 75–95% for 13 of the 14 reactions in the patient samples. The IgL reaction in sample BR17 was extensively inhibited, with an efficiency of only 59%. The IgL reaction in the same sample had the expected efficiency (83%; see Table 2
), suggesting that the sample contained components that predominantly interfered with the IgL PCR. We did not investigate the cause of inhibition but considered the sample an outlier and did not include it when calculating average efficiencies.
classification of samples
The most accurate classification of samples as lymphoma-negative or -positive is expected when sample-specific inhibition of both PCR reactions is determined by in situ calibration and accounted for. For the seven samples characterized this way, the fractions of IgL expression calculated directly from the measured CT values, i.e., without correcting for sample-specific effects, were 52% (BR0), 59% (BR46), 60% (BR42), and 63% (BR37) in the negative samples and 24% (BR18), 43% (BR17), and 89% (BR52) in the positive samples (Table 1
in the Data Supplement). Because lymphoma-negative samples should express IgL to 60% (1)(2), BR46, BR42, and BR37 are clearly negative, whereas BR18 and BR52 are positive. Classification of BR0 and BR17 is less clear, although the negative BR0 sample expressed IgL to a higher extent than the positive BR17 sample (52% vs 43%). When sample-specific efficiencies, as determined by in situ calibration, were taken into account, IgL expression in the negative samples was 56%, 62%, 62%, and 62%, whereas it was 6%, 16%, and 93% in the positive samples. The negative samples are thus clearly distinguished from the positive ones, as also seen in Fig. 3
, where they are clearly separated by the confidence region based on at least 95% probability.
We did not optimize the in situ calibration protocol in this work, but decided rather arbitrarily to dilute 64 times in steps of 4 and to choose samples containing sufficient material to avoid any statistical uncertainty (21)(22). Note that the in situ calibration protocol is economic with sample because it is based on dilution; less material is required, for example, than when running regular duplicate or triplicate samples. We performed the dilution in four steps, which means that four measurements were made on each sample. Although the samples were run in parallel, some additional hands-on time was required. For high-throughput analyses, it may therefore be time-efficient to first perform a standard single or duplicate measurement on each sample, do a gross classification, and then further characterize uncertain samples by in situ calibration. The gross classification can be quite accurate if any differences in average efficiencies and sensitivities of the two assays are considered. Fig. 3
shows such classification of the samples studied here. All negative samples are within the confidence region, whereas most positive ones are outside it. Three samples in Fig. 3
are false negatives. Two of them are at the periphery of the confidence region and therefore candidates for reinvestigation by the more accurate in situ calibration protocol. Indeed, BR17 was reinvestigated and found to be positive. The positive sample that appears as false negative and is not at the periphery is a diffuse large B-cell lymphoma. The reason for its misclassification has not been identified. It may be attributable to tumor sampling error or highly anomalous PCR efficiencies. However, some diffuse large B-cell lymphomas express very low amounts of immunoglobulins, and the monoclonal immunoglobulin production of the tumor cells may not be detected against the background of polyclonal immunoglobulin produced by nonlymphoma B-lymphocytes in the sample. The classification presented takes into account experimental uncertainty, but to evaluate the individual variation that may be present in various types of NHL a larger series of tumors must be characterized.
The described real-time PCR assay compares the expression of IgL and IgL and is designed to detect B-cell monoclonality in NHL. This is not sufficient for general tumor screening of patient samples, but demonstrates the potential, accuracy, and effectiveness of real-time PCR-based assays. When more markers are available, real-time PCR-based methods may have great potential as an important complement to morphologic and immunologic methods in clinical tumor diagnostics. Accurate analysis of fine-needle aspirates with a yield of 1000–100 000 representative cells is clearly sufficient for successful analysis. Indeed, we estimate that the material in a fine-needle aspirate will be sufficient for up to 50 different tests to detect various tumor cell characteristics, including B-cell monoclonality.
|
Acknowledgments |
|---|
:IgL expression ratio. This work was supported by the Swedish Research Council and the Swedish Cancer Society. |
Footnotes |
|---|
and IgL, immunoglobulin, and light chains, respectively; IHC, immunohistochemistry; CT, threshold cycle number; and dNTP, deoxynucleotide triphosphate. 
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
- or -type immunoglobulins. Blood 1976;47:79-89.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  W. Si, J. Gong, Y. Han, H. Yu, J. Brennan, H. Zhou, and S. Chen Quantification of Cell Proliferation and Alpha-Toxin Gene Expression of Clostridium perfringens in the Development of Necrotic Enteritis in Broiler Chickens Appl. Envir. Microbiol., November 1, 2007; 73(21): 7110 - 7113. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Ellerstrom, R. Strehl, K. Noaksson, J. Hyllner, and H. Semb Facilitated Expansion of Human Embryonic Stem Cells by Single-Cell Enzymatic Dissociation Stem Cells, July 1, 2007; 25(7): 1690 - 1696. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Treeck, G. Pfeiler, D. Mitter, C. Lattrich, G. Piendl, and O. Ortmann Estrogen receptor {beta}1 exerts antitumoral effects on SK-OV-3 ovarian cancer cells J. Endocrinol., June 1, 2007; 193(3): 421 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Montooth, K. T. Siebenthall, and A. G. Clark Membrane lipid physiology and toxin catabolism underlie ethanol and acetic acid tolerance in Drosophila melanogaster J. Exp. Biol., October 1, 2006; 209(19): 3837 - 3850. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Noaksson, N. Zoric, X. Zeng, M. S. Rao, J. Hyllner, H. Semb, M. Kubista, and P. Sartipy Monitoring Differentiation of Human Embryonic Stem Cells Using Real-Time PCR Stem Cells, October 1, 2005; 23(10): 1460 - 1467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ghosh, A. C. Steere, B. D. Stollar, and B. T. Huber In Situ Diversification of the Antibody Repertoire in Chronic Lyme Arthritis Synovium J. Immunol., March 1, 2005; 174(5): 2860 - 2869. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. G. Rutledge Sigmoidal curve-fitting redefines quantitative real-time PCR with the prospective of developing automated high-throughput applications Nucleic Acids Res., December 15, 2004; 32(22): e178 - e178. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Stahlberg, M. Kubista, and M. Pfaffl Comparison of Reverse Transcriptases in Gene Expression Analysis Clin. Chem., September 1, 2004; 50(9): 1678 - 1680. [Full Text] [PDF]
|
|
|
|
|
|
A. Stahlberg, J. Hakansson, X. Xian, H. Semb, and M. Kubista Properties of the Reverse Transcription Reaction in mRNA Quantification Clin. Chem., March 1, 2004; 50(3): 509 - 515. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Pastinen, R. Sladek, S. Gurd, A.'a Sammak, B. Ge, P. Lepage, K. Lavergne, A. Villeneuve, T. Gaudin, H. Brandstrom, et al. A survey of genetic and epigenetic variation affecting human gene expression Physiol Genomics, January 15, 2004; 16(2): 184 - 193. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Bar, A. Stahlberg, A. Muszta, and M. Kubista Kinetic Outlier Detection (KOD) in real-time PCR Nucleic Acids Res., September 1, 2003; 31(17): e105 - e105. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
, B-cell lymphomas; *, diffuse large B-cell lymphomas; •, negative samples. 
and 
indicate corrected CT values for samples for which specific PCR efficiencies were determined. 
and 
indicate fine-needle aspirates. Arrows connect measured and corrected CT values for individual samples.