JMD Am. J. Pathology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Originally published online as doi:10.2353/jmoldx.2008.080021 on June 13, 2008

Published online before print June 13, 2008
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jmoldx.2008.080021v1
10/4/355    most recent
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vaughn, C. P.
Right arrow Articles by Samowitz, W. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vaughn, C. P.
Right arrow Articles by Samowitz, W. S.
Journal of Molecular Diagnostics 2008, Vol. 10, No. 4
Copyright © 2008 American Society for Investigative Pathology & Association for Molecular Pathology
DOI: 10.2353/jmoldx.2008.080021

Confirmation of Single Exon Deletions in MLH1 and MSH2 Using Quantitative Polymerase Chain Reaction

Cecily P. Vaughn*, Elaine Lyon*{dagger} and Wade S. Samowitz*{dagger}

From the ARUP Institute for Clinical and Experimental Pathology, * Salt Lake City; and the Department of Pathology, {dagger} University of Utah Health Sciences Center, Salt Lake City, Utah


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Deletions of one or more exons in the mismatch repair genes MLH1 and MSH2 have been implicated in a significant fraction of hereditary nonpolyposis colorectal cancer (HNPCC or Lynch syndrome). Multiplex ligation-dependent probe amplification (MLPA) detection of deletions of multiple sequential exons is widely accepted; however, there is concern over the reliability of MLPA results showing single exon deletions. Given the clinical implications of a diagnosis of Lynch syndrome, it is important to use an alternative technique to confirm single exon deletions. To verify single exon deletions, we developed a SYBR Green-based quantitative polymerase chain reaction (PCR) assay. Clinical DNA samples containing deletions in 33 of the 35 exons in MLH1 and MSH2, previously screened by MLPA, were evaluated by quantitative PCR. Gene dosage ratios were determined by both the relative standard curve method and by the 2–{Delta}{Delta}CT method. Deleted exons had gene dosage ratios of 0.4 to 0.6, while nondeleted exons exhibited ratios of 0.8–1.3. We found 100% concordance between the quantitative PCR and MLPA results, including confirmation of all single exon deletions. The 2–{Delta}{Delta}CT method was as accurate as using standard curves for the calculation of ratios. Single exon deletions in MLH1 and MSH2 can be verified using quantitative PCR. Assays using this method are simple to design and easy to perform, making them ideal for confirmatory testing.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The detection of relatively large intragenic deletions has become increasingly important in patient screening for many disorders, including cystic fibrosis,1, 2, 3 Rett syndrome,4 and hereditary nonpolyposis colorectal cancer (HNPCC or Lynch syndrome).5, 6 In Lynch syndrome, 10 to 27% of pathogenic mutations have been attributed to large deletions in the MLH1 and MSH2 genes.5, 7, 8 Genomic rearrangements are particularly prevalent in the MSH2 gene, where they comprise up to 45% of mutations.9 Since large deletions cannot be detected by standard exonic sequencing, other methods such as Southern blotting,5, 10 long-range polymerase chain reaction (PCR),11 quantitative multiplex PCR,8 and multiplex ligation-dependent probe amplification (MLPA)12, 13, 14, 15 have been used to detect deletions spanning one or more exons.

The use of MLPA for detection of large deletions has recently become widespread because it provides a robust and relatively simple method to simultaneously look at gene dosage of up to 40 targets.12 However, while MLPA detection of deletions of multiple sequential exons is widely accepted, there is concern over the reliability of MLPA results showing single exon deletions.10, 13 Given the clinical implications of diagnoses of diseases such as Lynch syndrome, it is important that any detection method have built-in redundancy to guard against false-positive results. For samples with multiple contiguous deleted exons, dosage ratios of adjacent deleted exons provide internal confirmation of results. However, for MLPA results showing a single exon deletion, it is important to use an alternative technique for confirmation.

To this end, we examined the feasibility of using SYBR Green-based quantitative PCR to detect deletions as a rapid confirmation of single exon deletions in the MLH1 and MSH2 genes. Samples shown by MLPA to contain deletions were tested to verify whether quantitative PCR could distinguish samples containing heterozygous deletions from samples with nondeleted exons. We also compared two alternative gene dosage calculation methods.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
DNA Samples and Multiplex Ligation-Dependent Probe Amplification
Ten DNA samples previously determined by MLPA to contain large deletions in either the MLH1 or MSH2 gene were generously provided by the Mayo Clinic. Collectively, the samples represented heterozygous deletions in 18 of the 19 MLH1 exons and 15 of the 16 MSH2 exons. The deletions ranged in size between 1 and 13 exons (Table 1)Go . All single exon deletions were further confirmed by the Mayo Clinic using Southern blot analysis.


View this table:
[in this window]
[in a new window]

  		Table 1. Deletions Determined by MLPA

 
MLPA analysis was also performed at our institution to verify the deletions. This analysis was performed using 300 ng of each DNA sample with the SALSA MLPA Kit P003 for MLH1 and MSH2 (MRC-Holland, Amsterdam, The Netherlands). All steps were performed according to the manufacturer’s recommendations. Products were separated by capillary electrophoresis, and analysis of the results was performed using the GeneMarker software, version 1.6 (SoftGenetics, State College, PA). Exons with peak height ratios of less than 0.7 relative to the corresponding peak of the control samples were scored as deletions.

Quantitative Polymerase Chain Reaction
Primer Design
Primers for each exon in the MLH1 and MSH2 genes and for a fragment of the β2-microglobulin gene were designed using Primer Designer 5 (Scientific and Educational Software, Cary, NC). The primers were designed to have a TM of ~60°C and generate amplicons approximately 110 bp in length. Initially, one to three primer sets were chosen for each exon and standard curves were generated, as described below. Following amplification, melting curve analysis was performed to verify the absence of nonspecific products. Primers for analysis of patient samples were chosen based on the following two criteria: absence of nonspecific products and a PCR efficiency of ~2. Sequences of the chosen primers are listed in Table 2Go .


View this table:
[in this window]
[in a new window]

  		Table 2. Primer Sequences for Quantitative PCR

 
Standard Curves
DNA from a wild-type sample was serially diluted and amplified for each target exon at concentrations of 100, 50, 25, and 12.5 ng per reaction. Each concentration was run in duplicate. Crossing threshold (CT) values were determined by the LightCycler software, version 3.5 using the second derivative maximum. The efficiency of each PCR reaction was determined by the following formula: E = 10–1/slope.16

Amplification
Fifty nanograms of each DNA sample were amplified in a 20-µl reaction containing 0.5 µmol/L each primer, 3 mmol/L MgCl2, and 1X LightCycler FastStart DNA Master SYBR Green I (Roche Applied Science, Indianapolis, IN). Amplification was performed on the Roche LightCycler and entailed a preincubation of 95°C for 10 minutes followed by 40 cycles of denaturation at 95°C for 1 second, annealing at 62°C for 5 seconds, and extension at 72°C for 10 seconds. Samples were amplified in duplicate for both the target exon and the reference gene (β2-microglobulin).

Quantitative PCR for the each of the MLH1 exons was performed on samples 1 to 5 and for each of the MSH2 exons on samples 6 to 10. Given the location of deleted and nondeleted exons (as shown in Table 1Go ), a total of 54 deleted and 121 nondeleted exons were evaluated by quantitative PCR.

Calculation of Ratios
Gene dosage ratios for each exon were calculated using two methods: relative quantification using standard curves and the 2–{Delta}{Delta}CT method. CT values were determined by LightCycler software and the average CT of the duplicate reactions was used in the gene dosage ratio calculations. Both the standard curve and the 2–{Delta}{Delta}CT calculations are derived from the same equation, as previously described17, 18, 19 ; however, the first method accounts for differences in PCR efficiencies, whereas the 2–{Delta}{Delta}CT method assumes all efficiencies to be 2.

For relative quantification using standard curves, the concentration of the target exon and β2-microglobulin are determined by comparison of the resulting CT values to their respective standard curves. The ratio of target exon to β2-microglobulin for the patient sample is then divided by the ratio achieved for a control sample.

In the 2–{Delta}{Delta}CT method the PCR efficiencies for all reactions are assumed to be 2, and the gene dosage ratio is calculated using the following equation: 2–[{Delta}CT(target) – {Delta}CT(ref)], where {Delta}CT(target) equals the difference between the CT values for the patient sample and the control sample for the target exon and {Delta}CT(ref) equals the difference between the CT values for the patient sample and the control sample for the reference gene. Calculated gene dosage ratios were compared using a paired t-test.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Multiplex Ligation-Dependent Probe Amplification
All previously identified mismatch repair gene deletions were confirmed by MLPA at our institution (Figure 1Go for representative examples of multiple and single exon deletions).


Figure 1
View larger version (28K):
[in this window]
[in a new window]

  		Figure 1. MLPA results showing a multiexon deletion of MSH2 exons 8–15 (sample 10) (A) and a single exon deletion of MSH2 exon 8 (sample 9) (B). Nondeleted exons have ratios of ~1 (green squares), whereas deleted exons have ratios of ~0.5 (red squares).

 
Quantitative PCR
Standard curves were used to calculate PCR efficiencies (see Materials and Methods). The PCR efficiency of the reaction for β2-microglobulin was 2.04. PCR efficiencies of the reactions for MLH1 and MSH2 exons ranged from 1.80 to 2.27. Representative real-time amplification curves are shown in Figure 2Go .


Figure 2
View larger version (15K):
[in this window]
[in a new window]

  		Figure 2. Real-time amplification curves for samples 1 to 5 for β2-microglobulin (A) and MLH1 exon 13 (B). Each sample was run in duplicate. The starting concentration of all samples was 50 ng per reaction; thus, similar CT values for the reference gene, β2-microglobulin, are observed. In the MLH1 exon 13 amplification, samples without a deletion in this exon (samples 3 to 5) have CT values of ~22.5. Samples with a deletion in this exon (samples 1 and 2) have CT values that are approximately one cycle later at ~23.5.

 
The combined 10 patient samples had a total of 54 deleted exons (as determined by MLPA). Quantitative PCR results of these exons showed gene dosage ratios of 0.38 to 0.56 as calculated by relative quantification using standard curves. There were a combined 121 nondeleted exons, and quantitative PCR results of these exons gave gene dosage ratios of 0.79 to 1.29. By the 2–{Delta}{Delta}CT methodology, deleted exons showed gene dosage ratios of 0.42 to 0.55, and nondeleted exons showed ratios of 0.78 to 1.28 (Table 3)Go .


View this table:
[in this window]
[in a new window]

  		Table 3. Gene Dosage Ratios Determined by Quantitative PCR

 
Comparison of Calculation Methods
A comparison of the ranges, means, and standard deviations of the ratios achieved using the standard curve and the 2–{Delta}{Delta}CT methodologies is shown in Table 3Go . The gene dosage ratios calculated by the two methods differed by <0.08 for individual data points in all cases, and a paired t-test showed no statistical difference between the two calculation methods.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This report demonstrates the utility of using SYBR Green-based quantitative PCR for the verification of single exon deletions. Calculated gene dosage ratios provided an unambiguous distinction between deleted and nondeleted exons, with 100% concordance between the results of quantitative PCR and MLPA. Deletions were successfully detected in all of the exons in MLH1 and MSH2 for which samples with deletions were available (33 of the 35 exons). Gene dosage ratios of all nondeleted exons were clearly differentiated from the ratios of deleted exons.

Multiplex ligation-dependent probe amplification is a very powerful tool for simultaneously evaluating coding regions of entire genes for large deletions.6, 10 Due to the complex nature of the MLPA assay, however, this technique can be sensitive to DNA quality, concentration, and extraction method.10 Given the implications of an incorrect disease diagnosis for individuals and potentially affected family members, some form of confirmation of a deletion is desirable. In the case of multiexon deletions, this confirmation is provided by the results from neighboring exons, since the MLPA probe sets for each exon bind independently of each other. For example, in the multiexon deletion depicted in Figure 1AGo , the finding that exon 10 is deleted is supported by the finding that exons 9 and 11 are deleted; in this way deletion of contiguous exons can be seen to confirm the deletion found for each individual exon. However, the internal validation present in the detection of contiguous multiexon deletions is lacking for single exon deletions, as demonstrated by the MLPA analysis in Figure 1BGo . Sequencing the exon in question can be performed to verify the absence of variants that may hinder the binding of the MLPA probes and cause a false-positive result; however, this method does not provide confirmation of a true-positive result. Thus, for single exon deletions, it is important to employ an alternative technique for confirmation of results.

Assays using the SYBR Green quantitative PCR method presented here are well suited to confirming single exon deletions because they can be rapidly designed and performed. Suitable primers are easily designed since amplification products need only span a small (~110 bp) region of each exon. Additional specificity can be achieved by using hybridization probes; however, we were able to attain sufficient specificity by designing primers with high TMs and using an annealing temperature of 62°C. Real-time PCR on the LightCycler can be performed in under an hour and requires few control samples, resulting in a simple and rapid confirmatory test.

Alternative detection methods, which can also be used in the initial detection of deletions, arguably impose unnecessary difficulties for a confirmatory test. Southern blots are both time-consuming and expensive to perform. Methods using multiplex quantitative PCR require extensive optimization to achieve similar exponential phases of the PCR reactions of all targets in a single tube.20 Methods using long-range PCR or RT-PCR, which are often used in research studies, may not be amenable to use in a clinical setting. Long-range PCR across a possibly deleted region may be technically difficult or impossible, and RT-PCR requires a sample of patient RNA. Additionally, RT-PCR may not be possible in cases involving a deletion of the extreme 5' or 3' exon, and RNA has been shown to be subject to nonsense-mediated decay in the presence of knockout mutations.21 Finally, MLPA assays using either multiple probe sets per exon or secondary MLPA assays using alternative probe sets directed at the same exons are more expensive and time-consuming to perform than the quantitative PCR described in this study.

Quantitative PCR is further simplified if, as in this study, the 2–{Delta}{Delta}CT method can be used for calculating ratios of patient samples. While the standard curve methodology, which corrects for differences in PCR efficiency between the target exon and the reference gene, is the gold standard for quantification calculations, both methods produced near identical results in our data set. For assays quantifying gene dosage ratios in germline defects, the assumption of efficiencies of 2 used in the 2–{Delta}{Delta}CT method may have been appropriate for multiple reasons. First, small amplicons with similar PCR efficiencies can be designed. Second, there is a relative lack of complexity present in the quantitative detection of germline deletions in contrast to other quantitative PCR assays, such as those used for the detection of RNA transcript levels or infectious diseases. In particular, assays detecting germline deletions do not need to cover large dynamic ranges; rather, they need only determine twofold differences. Therefore the differences in PCR efficiency may be less consequential in this application of quantitative PCR. The effect of PCR efficiency differences on the calculation is further minimized by standardizing the input DNA concentration of the clinical and control samples. In fact, Pfaffl18 states that when input DNA is standardized, {Delta}CT(ref) can assumed to be 0, further simplifying the equation. However, the inclusion of a reference gene amplification provides for correction of DNA quality or any inhibitors that may be present. While not a problem in this study, changes in copy number and primer site polymorphism in the reference gene could affect the results of quantitative PCR. This could be addressed by using a nondeleted exon in the gene of interest as the reference.

In summary, the detection of large deletions is becoming increasingly important as a component of full gene analysis in the detection of pathogenic mutations in a myriad of diseases. Many methods are well suited to the detection of large deletions, and MLPA in particular has recently come into widespread use. For samples with MLPA results showing a single exon deletion, quantitative PCR can be used to verify results. We have presented here the primers and PCR conditions necessary to detect deletions in MLH1 and MSH2; however, the same design principles can be applied to other targets such as CFTR, MECP2 (Rett syndrome), and ENG and ALK1 (hereditary hemorrhagic telangiectasia), which also may require confirmation of single exon deletions. The simple design, accuracy, and rapid manner in which SYBR Green-based quantitative PCR can be performed makes it particularly ideal for confirmatory testing in a clinical setting.


   Acknowledgments
 
We thank Dr. Stephen Thibodeau of the Mayo Clinic and Rong Mao of ARUP Laboratories for assistance in the procurement of DNA samples for analysis. We also thank Alison Millson, Scott Reading, and Andrew Wilson of ARUP Laboratories.


   Footnotes
 
Address reprint requests to Wade S. Samowitz, M.D., ARUP Laboratories, 500 Chipeta Way, Salt Lake City, UT 84108. E-mail: wade.samowitz{at}aruplab.com

Accepted for publication April 3, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Bombieri C, Bonizzato A, Castellani C, Assael BM, Pignatti PF: Frequency of large CFTR gene rearrangements in Italian CF patients. Eur J Hum Genet 2005, 13:687-689[CrossRef][Medline]
  2. Niel F, Martin J, Dastot-Le Moal F, Costes B, Boissier B, Delattre V, Goossens M, Girodon E: Rapid detection of CFTR gene rearrangements impacts on genetic counselling in cystic fibrosis. J Med Genet 2004, 41:e118[Free Full Text]
  3. Hantash FM, Redman JB, Starn K, Anderson B, Buller A, McGinniss MJ, Quan F, Peng M, Sun W, Strom CM: Novel and recurrent rearrangements in the CFTR gene: clinical and laboratory implications for cystic fibrosis screening. Hum Genet 2006, 119:126-136[CrossRef][Medline]
  4. Schollen E, Smeets E, Deflem E, Fryns JP, Matthijs G: Gross rearrangements in the MECP2 gene in three patients with Rett syndrome: implications for routine diagnosis of Rett syndrome. Hum Mutat 2003, 22:116-120[CrossRef][Medline]
  5. Wijnen J, van der Klift H, Vasen H, Khan PM, Menko F, Tops C, Meijers Heijboer H, Lindhout D, Moller P, Fodde R: MSH2 genomic deletions are a frequent cause of HNPCC. Nat Genet 1998, 20:326-328[CrossRef][Medline]
  6. Armour JA, Barton DE, Cockburn DJ, Taylor GR: The detection of large deletions or duplications in genomic DNA. Hum Mutat 2002, 20:325-337[CrossRef][Medline]
  7. Wagner A, Barrows A, Wijnen JT, van der Klift H, Franken PF, Verkuijlen P, Nakagawa H, Geugien M, Jaghmohan-Changur S, Breukel C, Meijers-Heijboer H, Morreau H, van Puijenbroek M, Burn J, Coronel S, Kinarski Y, Okimoto R, Watson P, Lynch JF, de la Chapelle A, Lynch HT, Fodde R: Molecular analysis of hereditary nonpolyposis colorectal cancer in the United States: high mutation detection rate among clinically selected families and characterization of an American founder genomic deletion of the MSH2 gene. Am J Hum Genet 2003, 72:1088-1100[CrossRef][Medline]
  8. Charbonnier F, Raux G, Wang Q, Drouot N, Cordier F, Limacher JM, Saurin JC, Puisieux A, Olschwang S, Frebourg T: Detection of exon deletions and duplications of the mismatch repair genes in hereditary nonpolyposis colorectal cancer families using multiplex polymerase chain reaction of short fluorescent fragments. Cancer Res 2000, 60:2760-2763[Abstract/Free Full Text]
  9. Baudhuin LM, Ferber MJ, Winters JL, Steenblock KJ, Swanson RL, French AJ, Butz ML, Thibodeau SN: Characterization of hMLH1 and hMSH2 gene dosage alterations in Lynch syndrome patients. Gastroenterology 2005, 129:846-854[Medline]
  10. Baudhuin LM, Mai M, French AJ, Kruckeberg KE, Swanson RL, Winters JL, Courteau LK, Thibodeau SN: Analysis of hMLH1 and hMSH2 gene dosage alterations in hereditary nonpolyposis colorectal cancer patients by novel methods. J Mol Diagn 2005, 7:226-235[Abstract/Free Full Text]
  11. Nakagawa H, Hampel H, de la Chapelle A: Identification and characterization of genomic rearrangements of MSH2 and MLH1 in Lynch syndrome (HNPCC) by novel techniques. Hum Mutat 2003, 22:258[Medline]
  12. Schouten JP, McElgunn CJ, Waaijer R, Zwijnenburg D, Diepvens F, Pals G: Relative quantification of 40 nucleic acid sequences by multiplex ligation-dependent probe amplification. Nucleic Acids Res 2002, 30:e57[Abstract/Free Full Text]
  13. Wehner M, Mangold E, Sengteller M, Friedrichs N, Aretz S, Friedl W, Propping P, Pagenstecher C: Hereditary nonpolyposis colorectal cancer: pitfalls in deletion screening in MSH2 and MLH1 genes. Eur J Hum Genet 2005, 13:983-986[CrossRef][Medline]
  14. Sellner LN, Taylor GR: MLPA and MAPH: new techniques for detection of gene deletions. Hum Mutat 2004, 23:413-419[CrossRef][Medline]
  15. Gille JJ, Hogervorst FB, Pals G, Wijnen JT, van Schooten RJ, Dommering CJ, Meijer GA, Craanen ME, Nederlof PM, de Jong D, McElgunn CJ, Schouten JP, Menko FH: Genomic deletions of MSH2 and MLH1 in colorectal cancer families detected by a novel mutation detection approach. Br J Cancer 2002, 87:892-897[CrossRef][Medline]
  16. Rassmussen R: Quantification on the LightCycler. Meuer S Wittwer CT Nakagawara K eds. Rapid Cycle Real-Time PCR, Methods and Applications. 2001:pp 21-34 Springer Press New York
  17. Livak KJ, Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2(-delta delta C(T)) method. Methods 2001, 25:402-408[CrossRef][Medline]
  18. Pfaffl MW: A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001, 29:e45[Abstract/Free Full Text]
  19. Schneider M, Joncourt F, Sanz J, von Kanel T, Gallati S: Detection of exon deletions within an entire gene (CFTR) by relative quantification on the LightCycler. Clin Chem 2006, 52:2005-2012[Abstract/Free Full Text]
  20. Diebold R, Bartelt-Kirbach B, Evans DG, Kaufmann D, Hanemann CO: Sensitive detection of deletions of one or more exons in the neurofibromatosis type 2 (NF2) gene by multiplexed gene dosage polymerase chain reaction. J Mol Diagn 2005, 7:97-104[Abstract/Free Full Text]
  21. Kuzmiak HA, Maquat LE: Applying nonsense-mediated mRNA decay research to the clinic: progress and challenges. Trends Mol Med 2006, 12:306-316[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J. Mol. Diagn.Home page
E. Lyon and C. T. Wittwer
LightCycler Technology in Molecular Diagnostics
J. Mol. Diagn., March 1, 2009; 11(2): 93 - 101.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jmoldx.2008.080021v1
10/4/355    most recent
Right arrow Purchase Article
Right arrow View Shopping Cart
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vaughn, C. P.
Right arrow Articles by Samowitz, W. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vaughn, C. P.
Right arrow Articles by Samowitz, W. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS