An Urgent Use for Hemoglobin A1c?Tormey, William, P;Hickey, David, P
doi: 10.1093/clinchem/43.8.1463apmid: N/A
To the Editor: The deciding factor in whether to measure an analyte in the emergency out-of-hours period is whether the result will affect the immediate clinical management. Hemoglobin (Hb) A1c is not included in the list of tests usually provided (1)(2), but the unforeseen may occur. A transplant team was informed that kidneys and pancreas were available in another hospital from a 14-year-old girl who had suffered a cardiac arrest after a grand mal seizure. She had become hypernatremic and was infused with a 50 g/L dextrose solution. The surgeon was informed that the plasma glucose was 21 mmol/L and the question arose whether the hyperglycemia reflected undiagnosed diabetes mellitus. A Hb A1c done on call immediately was 7.1%, indicating a very high probability that the patient had diabetes (3). Normal oral glucose tolerance was found in only 4% of subjects with a Hb A1c value of at least 7% in a metaanalysis study (4). The pancreas was thus unsuitable, but the heart, liver, and kidneys were successfully transplanted. Decisions regarding the suitability of organs for transplantation must be made quickly. Hb A1c was the appropriate choice in that context. 1 Lester E. A new strategy for out-of-hours laboratory investigations. Ann Clin Biochem 1986 ; 23 : 497 -500. Crossref Search ADS PubMed 2 Legg EF. The on-call service: a regional survey. Ann Clin Biochem 1989 ; 26 : 19 -25. Crossref Search ADS PubMed 3 Stewart M. Will glycosylated haemoglobin replace the oral glucose-tolerance test?. Lancet 1997 ; 349 : 223 -224. Crossref Search ADS PubMed 4 Peters AL, Davidson MB, Schriger DL, Hasselblad V. A clinical approach for the diagnosis of diabetes mellitus: an analysis using glycosylated hemoglobin levels. Meta-analysis Research Group on the diagnosis of diabetes using glycosylated hemoglobin levels. JAMA 1996 ; 276 : 1246 -1252. Crossref Search ADS PubMed © 1997 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Samples from Patients to Laboratories. Walter Guder, Sheshadri Narayanan, Hermann Wisser, Bernd Zawta. Darmstadt: Git Verlag, 1996, 101 pp., 48 DM. ISBN 3–928865-22-6.Calam, Roger, R
doi: 10.1093/clinchem/43.8.1470apmid: N/A
Abstract An Analytical and Clinical Comparison of ACST BR with Two Other Breast Tumor Marker Assays Houser, S.M., Maimonis, P.J. Chiron Diagnostics, East Walpole, MA As the Preface indicates, this book is intended for the broad audience of laboratorians, physicians, nurses, and anyone involved with the proper application of knowledge as it relates to the preanalytical phase of laboratory testing. Visually rich with colorful illustrations, neatly numbered and captioned, 114 tables, graphs, diagrams, and photographs explicitly reinforce and clarify the text. In addition, the authors cleverly emphasize to the reader their recommendations and warnings by highlighting the former in green print, the latter in red. These printing contrasts make for efficient skimming for emphasized text. The book is well referenced, and an excellent glossary is included. An added bonus is a booklet attached to the inside back cover titled, List of analytes, Preanalytical variables. The presentation of information in user-friendly table form makes this a valuable, transportable ready reference for physicians and laboratorians alike. Before the first chapter, the authors take a novel approach intended to enhance the reader’s awareness of the importance of the multiple steps in the preanalytical process. They compare two short case studies for the same patient, demonstrating an ideal path to diagnosis vs a pathway with significant diagnostic confusion because preanalytical variables were not properly controlled. Having set the tone and purpose for the book, the first chapter launches into an adequate discussion of several biological variables and their influence on test results. These range from age and gender to diet and exercise to intake of caffeine, nicotine, and alcohol. The next three chapters provide an excellent discussion of specimen collection, transport, storage, and preparation of specimens for analysis. The reader is thoroughly informed about the importance of collection timing (influences of circadian rhythms, diagnostic/therapeutic procedures), result differences attributable to posture and to tourniquet time, and concerns associated with collection sites, e.g., indwelling lines. There is a discussion of the advantages and disadvantages of using plasma vs serum, as well as a comparison of analyte concentrations for these two matrixes. The illustrations for venipuncture and skin-puncture are exceptional. Stability schematics define the importance of preanalytical time and temperature for several analytes. However, one recommendation conflicts with the current NCCLS Approved Guideline for the Handling and Processing of Blood Specimens (H18-A), which states a requirement for a 2-h limit from the time of collection to the separation of serum/plasma from cells. The authors recommend a 1-h time limit, which is referenced to an earlier, since revised, NCCLS document. Chapter 4 covers multiple laboratory areas. Preanalytical considerations are defined for transfusion medicine, hematology, coagulation, chemistry, immunoassay, and molecular biology. Possibly in some areas there is not enough depth; however, the manner of information presentation is valuable because it serves up an awareness of the multiple potential factors that can influence result accuracy. A final chapter on interferences encapsulates information that will not be particularly new for many laboratorians. Laboratory scientists have become increasingly aware of the important influence of preanalytical variables on patient test results. This publication would make an excellent addition to one’s library of reference textbooks. It will have particular appeal to those involved with training and instruction. © 1997 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
The Errors of Our WaysBlumenthal,, David
doi: 10.1093/clinchem/43.8.1305pmid: N/A
Two articles in this issue of Clinical Chemistry provide additional evidence of the potential benefits of applying modern quality theory to clinical medicine. Though understandably focused on laboratory process, these papers nevertheless suggest lessons that are widely applicable to reducing error rates and improving quality of care throughout medical practice. The manuscripts complement each other nicely. Witte et al. (1) concentrate on rates of error within the analytical process of the laboratory, and find that widely discrepant values (>7 SD from expected) are relatively rare, occurring in 98 of 219 353 analyses they conducted. To compare error rates in the laboratory with error rates in other settings, they translate the figure 98 of 219 353 into a standard metric for error measurement: errors per million episodes (parts per million in industry jargon). Their data suggest an error rate of 447 ppm. They note, for example, that mishaps in anesthesia include 2.5 deaths per million cases, and aviation deaths are 0.18 per million passenger enplanements. Though their finding is seemingly much higher than for these other fields, the error rate in laboratory analyses is not really comparable, since the outcomes measured by Witte et al. are so different. Only 9 of the 98 laboratory errors were judged likely to have affected clinical decision-making, and whether any serious harm was likely to have resulted is unclear. Plebani and Carraro (2) provide valuable perspective on error rates in laboratory medicine by exploring a somewhat different question: What proportion of laboratory errors are attributable to problems outside the analytical process itself? Their conclusion is that the great majority of errors result from problems in the preanalytical or postanalytical “phases.” These problems include erroneous specification of the hospital unit, collection of the specimen from an infusion route, or failure to notify the physician of the laboratory finding. They conclude that improving the quality of the clinical chemistry system requires increased focus on error in the pre- and postanalytical phases of testing. Correcting such problems, they argue, depends on increased cooperation between laboratory and nonlaboratory personnel. From the standpoint of laboratory practice, several points about these two studies stand out. First, there is room for improvement in the analytical phase of the clinical chemistry process. Second, however, the quantitatively largest reductions in laboratory error are likely to result from interdepartmental cooperation designed to improve the quality of specimen collection and data dissemination. These observations provide valuable guidance for laboratory directors and medical administrators, who must deploy scarce resources to reduce medical error. From the standpoint of quality improvement more generally, other lessons emerge from these papers. First, it is clear that modern quality theory and practice—based on methods pioneered in industrial settings over the last 80 years—are gaining ground in medicine. Both papers liberally use standard industrial quality methods, including the concepts of special and common causes of error and the use of benchmarking across industries to provide perspective on error rates. The diffusion of such ideas and tools is gratifying evidence that quality management is coming of age in medicine, not only in the US but around the world. Second, the papers demonstrate the applicability of modern quality theories and tools to the job of improving medical processes. The relevance of these theories and tools will strike some observers as self-evident. When the jargon of quality management is stripped away, all that Witte et al. and Plebani and Carraro have done is to apply the scientific method to understanding the causes and (inferentially) the solutions of flaws in a clinical process. They developed methods of error detection, counted the errors, formulated and tested hypotheses about the sources of error, and reported the results. What could be less remarkable in a discipline (medicine) that prides itself on its roots in biomedical science? The point, however, is not that the methods themselves are innovative (they are not), but that they have thus far been so erratically applied in the daily work of medical practice. Medicine has done a remarkable job of using science to improve the technologies available for diagnosing and curing illness. The healthcare system has been much less diligent about applying the scientific method to improving the performance of the increasingly complex processes in which those technologies are embedded: the processes, for example, of deciding which laboratory tests to order, collecting the specimens, moving them to the laboratory, organizing the information that results, communicating it to physicians, and acting on the results. If a system is no better than its weakest link, then many of our most exciting new diagnostic and treatment methods have been hobbled by failures in the human systems they depend on. For decades, other sectors of society have been perfecting approaches to reducing errors that result from human and system failures. Out of insularity or defensiveness, health professionals and institutions have shown little interest in these approaches. Now, at long last, this situation is changing. The result is likely to be healthcare that is not only better but cheaper, and much more satisfying to practice. 1 Witte DL, VanNess SA, Angstadt DS, Pennell BJ. Errors, mistakes, blunders, outliers, or unacceptable results: how many?. Clin Chem 1997 ; 43 : 1952 -1956. Crossref Search ADS PubMed 2 Plebani M, Carraro P. Mistakes in a stat laboratory: types and frequency. Clin Chem 1997 ; 43 : 1348 -1351. Crossref Search ADS PubMed © 1997 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Counterpoint To (measure apo)B or not to (measure apo)B: a critique of modern medical decision-makingSniderman, Allan, D
doi: 10.1093/clinchem/43.8.1310pmid: N/A
Abstract The measurement of apo B provides critical information that is complementary to that provided by the plasma and lipoprotein lipids for the assessment of coronary risk and the choice of appropriate pharmacological therapy. Why then is this measurement not in more widespread clinical use? I suggest two explanations. First, against the evidence, there is a lingering perception that problems persist in its measurement in routine clinical practice. Far from this being the case, however, the measurement of apo B has met every reasonable standard of laboratory precision and reliability to allow its widespread introduction in clinical laboratories. The second impediment is that the introduction of new tests has become subject to the authority of consensus conferences, a new approach to medical decision-making. The number of such conferences is increasing astronomically, and their reports are major determinants of clinical practice and allocation of resources. Notwithstanding the benefits they have brought, here I argue that, just as with any other scientific method, the merits of this new method of decision-making need to be examined critically; for if we do not, a process that was established to introduce change may, in fact, retard it or destroy it altogether. In response to the above article by McNamara et al., who wrote in response to the editorial by Cianflone and myself, published in Clinical Chemistry last year [1], it appears that the word “standardization” means something different to them than it does to us. Words do matter, and they may be slippery creatures indeed. To illustrate this, let me first list various meanings that another important word, hypercholesterolemia, has taken on and then return to the word under discussion: standardization. In addition, I believe that a much more important question should be addressed; namely, if the measurement of apo B is as important as we have stated, why is it not in more widespread use? Trying to answer this question may shed light on some of the most important influences on decision-making in modern medicine, influences that have largely escaped common view and criticism, and it is on this issue I will particularly focus. I am a cardiologist, and coronary artery disease dominates my discipline. The initial clinical study by my colleagues and I involving apo B was published in 1980 (2). We reported, as did Avogaro et al. before us (3), that many patients with coronary disease had apo B concentrations that were disproportionately increased compared with the serum concentrations of total or LDL-cholesterol. But attention then—and since—as McNamara et al. make clear, has centered almost exclusively on total and LDL-cholesterol, not on apo B. Unfortunately, the fact of the matter is that most patients with coronary disease do not have markedly high cholesterol concentrations. Indeed, for the most part, their concentrations of total and LDL-cholesterol are indistinguishable from those in individuals who do not have coronary disease (4)(5)(6)(7). In my opinion, that is the reason why the “normal” limits for cholesterol have decreased successively, first from the 95th to the 90th percentile (8), thence to the 75th [9], and after that all the way to the 50th percentile in my country, Canada (10). Some very reputable people say that virtually everyone in Western Society is hypercholesterolemic in comparison with societies in which vascular disease is rare—and, of course, in an important way, they are right. But where does that leave those of us who try to diagnose and treat individuals from Western societies? By definition, hypercholesterolemia used to be the exception in coronary patients, but by changing the definition, it became the rule—the difficulty being, of course, that as sensitivity increased, specificity plummeted. Changing the definition created two true, but contradictory, realities. The first is that for large groups, risk undoubtedly increases as cholesterol concentrations mount (11). By contrast, for individuals, because of the enormous overlap of values, unless total or LDL-cholesterol is markedly increased, little prognostic information of value is obtained. The clinical difficulty that this situation creates should have meant that new tests developed to yield additional information as to risk would be greeted with enthusiasm. That was what I thought would be the case with apo B. I was wrong. Our editorial (1) dealt with the merits of measuring apo B. The above paper by McNamara et al. deals with whether the measurements of plasma triglyceride, HDL-cholesterol, and LDL-cholesterol are standardized. Although I recognize and respect the considerable efforts made by many to improve the accuracy and precision of the measurements of the plasma and lipoprotein lipids, I differ from McNamara et al. on the meaning of “standardization.” For me, as a clinician, it means that the test measures accurately what it purports to measure and that if any laboratory uses any approved manufacturer’s reagents in an approved fashion, that laboratory will get the same answer as any other laboratory that does the same. By that definition, the determinations of both cholesterol and apo B have been “standardized.” But nothing in life or medicine is that simple. In the case of cholesterol, a Definitive Method exists to measure it; apo B values, on the other hand, are assigned by comparison with a primary preparation whose mass has been determined by amino acid composition. In the hierarchy of analytical precision, the measurement of cholesterol would seem ahead of apo B. Indeed, it is, but not as far as one might initially assume, given that the cholesterol Reference Methods against which the manufacturer’s products are calibrated have a bias when compared with the Definitive Method. That is to say, the everyday “reference methods” for neither cholesterol nor apo B are Definitive Methods. Perhaps another word would help us here: “harmonization,” by which is meant the process that ensures that different manufacturers’ products give the same answers in the everyday world; measurements of both cholesterol and apo B meet that standard. But harmonization has a weak sound to it; standardization has a stronger, more scientific ring to it, even though its definition in the Concise Oxford Dictionary, “obtain by analysis specific value of (solution etc.) for purposes of comparison,” comes very close to harmonization. Whatever word we choose—standardization or harmonization—must not obscure the fact that both cholesterol and apo B can be accurately and precisely measured in clinical laboratories. However, that is not yet the case for HDL- and LDL-cholesterol and triglyceride. To be sure, primary Reference Methods (not Definitive Methods) for HDL-cholesterol and triglyceride have been developed by the CDC, but they have not yet been fully implemented, even by the CRMLN laboratories. In my view, considering the measurements of HDL-cholesterol and triglyceride as being in the same category as cholesterol and apo B measurements would demand an elasticity in the word “standardization” that the Mad Hatter might approve of, but I do not. Moreover, just how many laboratories in the world are approved by use of this standard of comparison, and how are we clinicians to know which is which as we review their reports? My understanding is that even in North America only a relatively small number of the total are “approved.” And even for those that are, compliance in the US, for example, is certified only at 6-month intervals. Indeed, it was because the measurements of apoproteins had been truly standardized (12)(13)(14)(15) that we felt the studies of Contois et al. (16)(17) were particularly important. Unfortunately, recognition of this achievement is not as widespread as it should be, with the view persisting even amongst some eminent authorities that reliable measurement of apoproteins can still be achieved only in research laboratories. Even McNamara et al., while acknowledging—indeed, even applauding—the standardization of apoprotein measurement at the beginning of their article, seem to question it at the end, evidence of just how deeply old doubts are lodged and just how difficult they are to dispel. Meanwhile, while I continue to disagree with McNamara et al. on the issue of standardization, my views are similar to theirs on the too frequent inaccuracies involved in determination of LDL-cholesterol. To overcome this problem, they appear to suggest that direct measurement of LDL-cholesterol should rapidly be adopted. On this score, I believe that although the methods proposed to date are promising, they have not been fully evaluated and, in any case, they are far from being uniformly standardized. Now let me use this opportunity to address the even more important issue: Why is measurement of apo B not routine? That was the major thrust of our editorial [1]: Measurement of apo B provides critical information, complementary to that obtained from measuring the lipoprotein lipids, information that is essential for accurate clinical diagnosis and the right choice of therapy. The normolipidemic patient may have an increased number of LDL particles but if apo B is not measured, that will be missed. In normolipidemic individuals with an increased apo B, the risk of cardiovascular disease is increased to the same extent as in persons with type II hyperlipoproteinemia (18). The same is true for the hypertriglyceridemic patient with an increased apo B (18). Moreover, how can therapy be intelligently selected for the hypertriglyceridemic patient if apo B is not known (19)? Indeed, is any therapy required for hypertriglyceridemic patients if apo B is not increased (19)? Packard and colleagues (20) reported that the commonest dyslipoproteinemia associated with coronary disease is mild hypertriglyceridemia, low HDL-cholesterol, and increased numbers of small, dense LDL particles. Genest et al. showed that familial hypoalphalipoproteinemia can also be associated with an increased apo B concentration (21). Surely these diagnostic advantages should not be restricted to research laboratories, now that apo B assays have been standardized. Assume for a moment that measurement of apo B does add important information to the characterization of vascular risk and the choice of therapy. Why then is it available in so few laboratories? I suggest it is because decision-making in medicine has changed radically in the past few years. There is little difficulty generating a lobby for a test where payment is assured, but it is virtually impossible to obtain support for a test for which payment is not already in place. I am not arguing that anyone can, or should, do tests for nothing. But what criteria are being used to judge whether a test will be paid for and who are applying the criteria? Much of our decision-making has been turned over to consensus conferences, typically small groups of individuals who consider questions of the moment and then issue guidelines on how these should be managed. The consensus or guidelines are promulgated as the formal views of whatever group chose the members of the panel. The acronym of the meeting becomes the author and the authority of the report, eclipsing the identities of the individuals who actually constructed it. Much good has occurred by this route, but the limitations of the process have evaded scrutiny. In a word—dare I say it—the process is unstandardized. Indeed, no approach I am aware of has been accepted so broadly with so little analysis of its contents and methods. From 1992 to January 1997, the Medline Database and the Health Star database list 913 consensus conferences, 141 from the NIH alone. As for guidelines, 3286 are indexed as publication types. I have not reviewed each citation, but unquestionably a number will be retrieved both as a guideline and as a consensus statement. As well, some have been published more than once. But even halving the numbers still leaves a remarkable growth industry. Moreover, the conclusions of these conferences can carry enormous consequence, both for the practice of medicine and for the economics of the practice of medicine. In some areas, many millions of lives may be affected and the allocation of many millions of dollars determined by a process not subject to peer review. We, the readers, are usually told little or nothing about the total base of information considered, how it was analyzed, how often the members met, and for how long. The strength of evidence is often weighted, but how were these decisions reached? In general, no minority reports are presented. What then does the word “consensus” mean? How often is what comes out determined by who went in? How often does the need to achieve “consensus” mean that such gatherings become meetings of the like-minded? Regarding lipid analyses, I believe it is fair to say that the NCEP guidelines (9)(22) drive the process around the world. Each cycle has begun with an NCEP report, after which other groups and countries respond. The reports differ, in some ways importantly, but overall they are also very similar. With respect to the last NCEP conference (22), I would be the first to point out that, at that time, important information about apo B was missing. Most critically, the measurement of apoproteins had not been standardized; moreover, clear-cut results from a large prospective study favoring apo B determinations were not available. Both of those requirements have now been met (12)(13)(14)(15)(23). But does that mean that nothing will change until this group meets again? And what if they say apo B measurements are not clinically useful? Given the authority the process has acquired, does this mean any change is put on hold indefinitely? Delay is not neutral. The strategies that lower LDL-cholesterol concentrations lower apo B. The strategies that lower LDL-cholesterol and apo B save lives and reduce the need for bypass surgery and angioplasty (24)(25)(26)(27). Because small, dense LDL particles contain less cholesterol than normal LDL particles do, apo B is a better marker of LDL particle number than is LDL-cholesterol (28). Using apo B, therefore, does not change the LDL argument; it merely extends it. Bluntly put, by not measuring apo B, we are not treating a large number of people who could otherwise have been helped. What about cost? No one can deny the thrust of governments everywhere to control medical costs. Who believes governments’ first priority is now quality of care rather than cost of care? In the US, the situation is even more complex with, for the first time, healthcare being organized into large economic units for which the ultimate goal is profit. Given this reality, how likely is it that new tests and therapies will be introduced as early as possible? In this case, however, without developing the argument here, I believe measurement of apo B will save money as well as lives. As already noted, the key problem with the cholesterol algorithm is that unless values are markedly increased, cholesterol is a poor marker of individual risk. Many patients must be treated when only a few will benefit. Apo B now, and other markers soon (29)(30), will markedly improve our ability to recognize the truly high-risk individuals so that fewer need be better treated and overall cost will be less. In this Counterpoint, I have ranged far from the narrow issue of standardization that prompted it. I have addressed issues, not individuals, but I am aware of the risks that criticism holds for the critic. Misinterpretation of motive can divert attention from the message to the messenger. Even so, I must express concern about the inherent risks in what seems to me to be a powerful, unseen, and dangerous trend to excessive concentration of decision-making in modern medicine. I am also concerned about the dangers of excessive commercialization of the practice of medicine. I am concerned that we are at risk of losing sight of who we are and why we do what we do. If that occurs, we will lose the essence of what we are, a profession committed to the best care of each life entrusted to us. Mike Rosenbloom Laboratory for Cardiovascular Disease, Royal Victoria Hospital, 687 Pine Ave. West, Montreal, Quebec, H3A 1A1. Fax 514-982-0686; e-mail [email protected]. 1 Sniderman AD, Cianflone K. Measurement of apolipoproteins: time to improve the diagnosis and treatment of atherogenic dyslipoproteinemias [Editorial]. Clin Chem 1996 ; 42 : 489 -491. Crossref Search ADS PubMed 2 Sniderman AD, Shapiro S, Marpole DG, Skinner BF, Teng B, Kwiterovich PO, Jr. Association of coronary atherosclerosis with hyperapobetalipoproteinemia [increased protein but normal cholesterol levels in human plasma low-density (B) lipoproteins]. Proc Natl Acad Sci U S A 1980 ; 77 : 604 -608. Crossref Search ADS PubMed 3 Avogaro P, Bittolo Bon G, Cazzolato G, Quinci GB. Are apolipoproteins better discriminators than lipids for atherosclerosis? Lancet 1979;i:901–3.. 4 Kannel WB, Castelli WP, Gordon T, McNamara PM. Serum cholesterol, lipoprotein, and the risk of coronary heart disease. The Framingham Study. Ann Intern Med 1974 ; 74 : 1 -12. 5 Gotto AM, Gorry GA, Thompson JR, Cole JS, Trost R, Yeshurun D, DeBakey ME. Relationship between plasma lipid concentration and coronary artery disease in 496 patients. Circulation 1977 ; 56 : 875 -883. Crossref Search ADS PubMed 6 Holmes DR, Jr, Elveback LR, Frye RL, Kottke BA, Ellefson RD. Association of risk factor variables and coronary artery disease documented with angiography. Circulation 1981 ; 63 : 293 -299. Crossref Search ADS PubMed 7 Goldstein JL, Hazzard WR, Shrott HG, Bierman EL, Motulsky AG. Hyperlipidemia in coronary disease. I. Lipid levels in five survivors of myocardial infarction. J Clin Invest 1973 ; 52 : 1533 -1543. Crossref Search ADS PubMed 8 Consensus Conference. Lowering blood cholesterol to prevent heart disease. JAMA 1985;253:2080–6.. 9 National Cholesterol Education Program Expert Panel. Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults. Arch Intern Med 1988;148:36–9.. 10 Canadian Consensus Conference on Cholesterol: final report. The Canadian Consensus Conference in the prevention of heart and vascular disease by altering serum cholesterol and lipoprotein risk factors. Can Med Assoc J 1988;139(Suppl):1–8.. 11 Stamler J, Wentworth D, Neuton JD. Is the relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? Findings in 356,222 primary screenees of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA 1986 ; 256 : 2823 -2828. Crossref Search ADS PubMed 12 Marcovina SM, Albers JJ, Dati F, Ledue TB, Ritchie RF. International Federation of Clinical Chemistry standardization project for measurements of apolipoproteins A-I and B. Clin Chem 1991 ; 37 : 1676 -1682. Crossref Search ADS PubMed 13 Albers JJ, Marcovina SM, Kennedy H. International Federation of Clinical Chemistry standardization project for measurements of apolipoproteins A-I and B. II. Evaluation and selection of candidate reference materials. Clin Chem 1992 ; 38 : 658 -662. Crossref Search ADS PubMed 14 Marcovina SM, Albers JJ, Henderson LO, Hannon WH. International Federation of Clinical Chemistry standardization project for measurements of apolipoproteins A-I and B. III. Comparability of apolipoprotein A-I values by use of international reference material. Clin Chem 1993 ; 39 : 773 -781. Crossref Search ADS PubMed 15 Marcovina SM, Albers JJ, Kennedy H, Mei JV, Henderson LO, Hannon WH. International Federation of Clinical Chemistry standardization project for measurements of apolipoproteins A-I and B. IV. Comparability of apolipoprotein B values by use of international reference material. Clin Chem 1994 ; 40 : 586 -592. Crossref Search ADS PubMed 16 Contois JH, McNamara JR, Lammi-Keefe CJ, Wilson PWF, Massov T, Schaefer EJ. Reference intervals for plasma apolipoprotein B determined with a standardized commercial immunoturbidimetric assay: results from the Framingham Offspring Study. Clin Chem 1996 ; 42 : 507 -514. Crossref Search ADS PubMed 17 Contois JH, McNamara JR, Lammi-Keefe CJ, Wilson PWF, Massov T, Schaefer EJ. Reference intervals for plasma apolipoprotein A-I determined with a standardized commercial immunoturbidimetric assay: results from the Framingham Offspring Study. Clin Chem 1996 ; 42 : 515 -523. Crossref Search ADS PubMed 18 Lamarche B, Després JP, Moorjani S, Cantin B, Dagenais GR, Lupien PJ. Prevalence of dyslipidemic phenotypes in ischemic heart disease (prospective results from the Quebec Cardiovascular Study). Am J Cardiol 1995 ; 75 : 1189 -1195. Crossref Search ADS PubMed 19 Vega GL, Grundy SM. Primary hypertriglyceridemia with borderline high cholesterol and elevated apolipoprotein B concentrations. Comparison of gemfibrozil vs lovastatin therapy. JAMA 1990 ; 264 : 2759 -2763. Crossref Search ADS PubMed 20 Griffin BA, Freeman DJ, Tait GW, Thomson J, Caslake MJ, Packard CJ, Shepherd J. Role of plasma triglyceride in the regulation of plasma low density lipoprotein (LDL) subfractions: relative contribution of small, dense LDL to coronary heart disease risk. Atherosclerosis 1994 ; 106 : 241 -253. Crossref Search ADS PubMed 21 Genest J, Jr, Bard JM, Fruchart JC, Ordovas JM, Schaefer EJ. Familial hypoalphalipoproteinemia in premature coronary artery disease. Arterioscler Thromb 1993 ; 13 : 1728 -1737. Crossref Search ADS PubMed 22 National Cholesterol Education Program Expert Panel. Second Report of the Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). Circulation 1994;89:1329–445.. 23 Lamarche B, Moorjani S, Lupien PJ, Cantin B, Bernard PM, Dagenais GR, Després JP. Apolipoprotein A-1, B levels and the risk of ischemic heart disease during a five-year follow-up of men in the Québec Cardiovascular Study. Circulation 1996 ; 94 : 273 -278. Crossref Search ADS PubMed 24 Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–9.. 25 Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996 ; 335 : 1001 -1009. Crossref Search ADS PubMed 26 Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW, et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 1995 ; 333 : 1301 -1307. Crossref Search ADS PubMed 27 The Post Coronary Artery Bypass Graft Trial Investigators. The effect of aggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation on obstructive changes in saphenous-vein coronary-artery bypass grafts. N Engl J Med 1997;336:153–62.. 28 Teng B, Thompson GR, Sniderman AD, Forte TM, Krauss RM, Kwiterovich PO, Jr. Composition and distribution of low-density lipoprotein fractions in hyperapobetalipoproteinemia, normolipidemia and familial hypercholesterolemia. Proc Natl Acad Sci U S A 1983 ; 80 : 6662 -6666. Crossref Search ADS PubMed 29 Després JP, Lamarche B, Mauriège P, Cantin B, Dagenais GR, Moorjani S, Lupien PH. Hyperinsulinemia as an independent risk factor for ischemic heart disease. N Engl J Med 1996 ; 334 : 252 -257. 30 Lamarche B, Tchernof A, Moorjani S, Cantin B, Dagenais GR, Lupien PJ, Després JP. Small, dense low-density lipoprotein particles as a predictor of the risk of ischemic heart disease in men. Prospective results from the Québec Cardiovascular Study. Circulation 1997 ; 95 : 69 -75. Crossref Search ADS PubMed © 1997 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Extraction of glyceric and glycolic acids from urine with tetrahydrofuran: utility in detection of primary hyperoxaluriaDietzen, Dennis, J;Wilhite, Timothy, R;Kenagy, David, N;Milliner, Dawn, S;Smith, Carl, H;Landt,, Michael
doi: 10.1093/clinchem/43.8.1315pmid: N/A
Abstract Primary hyperoxaluria (PH) is an autosomal recessive metabolic abnormality characterized by excessive oxalate excretion leading to nephrocalcinosis and progressive renal dysfunction. Type I primary hyperoxaluria (PH I) results from a deficiency of alanine:glyoxylate aminotransferase, whereas type II disease has been traced to a deficiency of d-glycerate dehydrogenase. The two syndromes are often distinguished on the basis of organic acids that are coexcreted with oxalate: glycolate and l-glycerate in type I and type II disease, respectively. Routine organic acid analysis with diethyl ether extraction followed by gas chromatographic analysis failed to detect normal and increased concentrations of these diagnostic metabolites. Subsequent extraction of urine with tetrahydrofuran (THF), however, extracted 75% of added glycerate, 42% of added glycolate, and 75% of added ethylphosphonic acid (internal calibrator). THF extraction was analytically sensitive enough to allow determination of normal excretion of glycolate (14–72 μg/mg creatinine) and glycerate (0–5 years, 12–177 μg/mg creatinine and >5 years, 19–115 μg/mg creatinine). Four of five patients with PH I and both patients with type II disease were correctly identified. Thus, THF extraction is a convenient adjunct to routine organic acid analysis and facilitates the detection of PH. indexing terms: inborn errors of metabolism, oxalates, renal calculi, laboratory diagnosis, gas chromatography Oxalate excreted in urine arises from metabolism of amino acids and carbohydrates in liver cytosol and peroxisomes (1)(2). Oxalate is ordinarily a minor end product of these pathways. In peroxisomes, glyoxylate is derived by oxidation of glycine and glycolate, and then is largely transaminated to glycine through the action of alanine:glyoxylate aminotransferase (AGT, E.C. 2.6.1.44), but can alternatively be oxidized to oxalate via l-2-hydroxyacid oxidase (E.C. 1.1.3.1).7 In the cytosol, hydroxypyruvate derived from glucose and fructose is primarily reduced to l-glycerate or d-glycerate by lactate dehydrogenase (E.C. 1.1.1.27) or d-glycerate dehydrogenase (GDH, E.C. 1.1.1.29), respectively. Under normal circumstances a small amount of hydroxypyruvate is also oxidized to oxalate through pathways that remain undefined. Primary hyperoxaluria (PH) results from inherited deficiency of AGT or GDH, which ordinarily metabolize glyoxylate and hydroxypyruvate to less toxic products. The net effect of these deficiencies is an increased commitment of intermediates to oxalate, resulting in increased urinary excretion of oxalate, supersaturation of urine with calcium oxalate, and, in advanced disease, systemic deposition of calcium oxalate. Formation of renal stones and the resulting renal dysfunction are the presenting features of the disease (3)(4). This disorder is distinct from enteric hyperoxaluria, which results from hyperabsorption of oxalate from the digestive tract (5). Type I primary hyperoxaluria (PH I) is an autosomal recessive deficiency of AGT with increased excretion of both oxalate and glycolate in urine (6)(7)(8)(9)(10). Type II primary hyperoxaluria (PH II) results from a deficiency of GDH, leading to coaccumulation of oxalate and l-glycerate in urine (11)(12). European literature has reported that 1–2% of childhood cases of end-stage renal disease may be attributable to some form of PH (13)(14). PH I is diagnosed more commonly, has an earlier onset, and a poorer prognosis than PH II (11)(15)(16). Some cases of PH I are responsive to pyridoxine (an AGT cofactor), high fluid intake, phosphate treatment, or citrate administration, yet heroic measures such as kidney and (or) liver transplantation are commonly required after irreversible kidney damage (2)(17)(18)(19)(20). Patients with PH II have a much more benign prognosis, though some do progress to end-stage renal disease (12). Thus, detection of PH and the distinction between PH I and PH II is necessary for patients who might benefit most from aggressive intervention. Currently this is accomplished by determination of AGT activity in liver biopsy specimens (6)(7) and measurement of glycolate and glycerate excretion. Liver biopsy allows direct assessment of enzymatic deficiency, but has several disadvantages: patient discomfort, the risk of an invasive procedure, cost, long turnaround time, and (currently) no source for GDH determination. In addition, individuals homozygous for AGT deficiency sometimes have enzyme activities indistinguishable from heterozygotes who are clinically normal (21)(22). Reliable detection of l-glycerate and -glycolate in urine may obviate the need for biopsy in many cases. Specific methods for glycolate and glycerate determination have been developed (23)(24), but most laboratories, even specialized ones aimed at detecting genetic metabolic disease, do not maintain specific testing protocols for glycolate and l-glycerate. These compounds are only occasionally detected in routine organic acid analysis by gas chromatography–mass spectrometry (GC-MS) because their polar character makes them difficult to extract from urine with diethyl ether or ethyl acetate, the common solvents used in routine organic acid analysis (25). With current methods, normal concentrations of glycolate and glycerate are rarely detected; extreme increases in excretion may be required for detection. This could partly explain the observation by Danpure that 30% of PH I patients with marked hyperoxaluria did not display increased glycolate excretion (26). We reasoned that a procedure that increased the efficiency of glycolate/glycerate extraction would provide more accurate normal ranges and potentially improve the sensitivity of organic acid analysis to detect and distinguish between PH I and PH II earlier in the course of the disease. Rimoldi et al. improved the extraction of polar compounds such as citric, hydroxybutyric, and orotic acids from urine by using tetrahydrofuran (THF) (27). Here, we describe the utility of THF extraction in aiding the diagnosis of PH. Materials and Methods Materials. Bis(trimethylsilyl)trifluoroacetamide (BSTFA) and oxalic, hippuric, succinic, and d,l-lactic acids were obtained from Sigma (St. Louis, MO). Ethylphosphonic acid, glycolic acid, and THF were purchased from Aldrich (Milwaukee, WI). d,l-Glyceric acid was obtained from ICN (Cleveland, OH). Specimens. Random urine specimens were adjusted to pH 2 to ensure complete recovery of oxalate and maintained at −20 °C. A preliminary experiment showed that glycerate and glycolate were stable in acidified urine for up to 3 months. Oxalate salts precipitated in as little as 2 weeks of storage even at pH 2, and thus oxalate was not measured on specimens stored for longer periods. Normal ranges for glycerate, glycolate, and oxalate (normalized to creatinine) were established with 65 specimens from children and adults without evidence of liver or kidney disease. Specimens from children <6 months of age were obtained from outwardly healthy children visiting the outpatient clinic at St. Louis Children’s Hospital. One glycolate and seven oxalate values were statistically excluded from the normal range by the outlier analysis of Reed et al. (28); oxalate:creatinine ratios are known to be nonnormally distributed in children (9). This study was conducted in accord with a protocol approved by the human studies committees of Washington University and the Mayo Clinic. Analytical. A volume of urine containing 500 μg of creatinine was diluted to 5 mL. Ethylphosphonic acid (250 μg) was added as internal calibrator. The urine was saturated with NaCl and acidified to pH 1 with HCl before extraction. Extraction was performed three times with 5 mL of diethyl ether and then five times with 5 mL of THF. THF and ether extracts were pooled separately and concentrated to dryness under nitrogen. Residual water was removed by reconstituting the residue with 200 μL of benzene to form an azeotropic mixture and again reducing it to dryness under nitrogen. Residues were then derivatized by incubation in pyridine:BSTFA (1:1 by vol) for 15 min at 60 °C. Calibrators were dissolved in pyridine and derivatized with an equal volume of BSTFA before use. Derivatives were analyzed on a Varian 3700 gas chromatograph (Varian Instrument Group, Palo Alto, CA) equipped with a DB-1 column (0.53 mm i.d.; PJ Colbert Assoc., St. Louis, MO) by using a temperature program of 7 min at 80 °C followed by a rise to 260 °C at 6 °C per minute. The injector and detector temperatures were both 250 °C. Compounds were detected by flame ionization and identity of the peaks was confirmed with a Finnigan ITD mass spectrometer (Finnigan MAT, San Jose, CA). Quantification of oxalate, glycolate, and glycerate was based on the detector response to a known amount of each compound and corrected for recovery of the internal calibrator. Glycolic, oxalic, and ethylphosphonic acid calibrators were prepared by dissolving highly pure material in pyridine and derivatizing immediately before use. Glyceric acid was supplied as a syrup with a significant water content and was lyophilized before dissolving in pyridine. Creatinine was determined on the Vitros 700 XR (Johnson and Johnson, Rochester, NY). Results Extraction of glycerate and glycolate. An aliquot of urine containing 500 μg of creatinine and supplemented with 250 μg each of succinate (C4H6O4), hippurate (C9H9O3), oxalate (C2H4O4), glycerate (C3H7O4), and glycolate (C2H5O3) was saturated with NaCl, acidified, and extracted three times with 5.0 mL of ether (standard protocol), then multiple times with 5.0 mL of THF. The first ether and THF extracts were concentrated, derivatized, and analyzed by GC. The more hydrophobic molecules, hippurate and succinate, were extracted effectively by ether while the very polar C-2 and C-3 acids were not extracted from the urine until THF was used (Fig. 1 ). Lactate, with intermediate hydrophobicity, was partially extracted with ether but predominantly recovered in THF extracts. Several other polar compounds (phosphate, urea, and citrate) were also efficiently extracted with THF. Ethylphosphonic acid (C2H7PO3) was used as an internal calibrator because of its absence in human urine and polar character similar to glycerate and glycolate. It was extracted exclusively with THF. Optimization of THF extraction. The goal of the protocol was to extract as much of each diagnostic compound as possible and to match the recovery of each to the internal calibrator. To determine the optimal number of extractions, 250 μg each of oxalate, glycerate, glycolate, and ethylphosphonate were added to an aliquot of normal urine (containing 500 μg creatinine) that was extracted three times with ether and then repeatedly extracted with THF. When individual THF extracts were analyzed, oxalate was recovered predominantly in two extractions, whereas five extractions were required to recover a comparable amount of glycerate, glycolate, and ethylphosphonate (Table 1 ). In subsequent studies, pooling of five successive extracts resulted in recovery of 43% ± 13%, 76% ± 7%, 43% ± 6%, and 71% ± 9% of oxalate, glycerate, glycolate, and ethylphosphonate, respectively (mean ± SD, n = 8). Correction of values based on the recovery of the internal calibrator, therefore, slightly underestimates the amount of glycolate and oxalate present. Extraction of glycolate and glycerate was linear up to 1000 μg/mg creatinine (Sy|x = 18 μg/mg creatinine, r = 0.9883, and Sy|x = 47 μg/mg creatinine, r = 0.9683, respectively). A precision study consisting of six runs over 6 weeks was performed with a normal urine pool containing mean glycolate and glycerate concentrations of 28.5 and 80.6 μg/mg creatinine, respectively. At these concentrations, where imprecision is likely to be high, CVs were 17% and 28.8% for glycolate and glycerate determination, respectively. Peaks corresponding to 5 μg/mg creatinine were readily detectable above the baseline signal. Analysis of patient specimens. Normal ranges were determined from random urine specimens obtained from healthy children of laboratory employees and adults with no history of renal disease and are reported as the actual range of observed values. Specimens from infants <6 months of age were obtained from outwardly healthy children visiting the outpatient clinic at St. Louis Children’s Hospital. Glycerate excretion in this healthy population was dependent on age (Fig. 2 ). Children <5 years of age (n = 19) displayed higher excretion rates of glycerate (12–177 vs 19–115 μg/mg creatinine) than older children and adults (n = 39). Normal glycolate excretion was 14–72 μg/mg creatinine (n = 64). No gender differences were apparent. The new extraction protocol was applied to specimens from 16 PH patients seen at the Mayo Clinic Division of Nephrology to assess the ability of improved glycolate/glycerate extraction to discriminate between PH I and PH II. Patients were all 5 years of age or older and were classified by: (a) prior history of renal dysfunction, (b) hyperexcretion of oxalate/glycerate or oxalate/glycolate (determined by standard organic acid analysis at reference laboratories), (c) liver AGT activity in liver biopsy material (when performed), and (d) response to pyridoxine. Pyridoxine (a cofactor for AGT but not GDH) augments existing AGT activity in some patients, resulting in normalized oxalate excretion, and is therefore a hallmark of PH I. We examined specimens from nine individuals whose urine oxalate excretion was responsive to pyridoxine. Consistent with clinical response, glycolate and glycerate concentrations were within normal limits in this population (data not shown). Five other patients were classified as PH I, three by history of marked hyperoxaluria and glycolate hyperexcretion (DM, LF, AJ), one as the result of AGT deficiency by liver biopsy (NB), and one other (AM) on the basis of glycolate hyperexcretion and biopsy-proven AGT deficiency in an affected sibling. Four of these five patient specimens had increased glycolate (Table 2 ). The initial specimen from AM displayed high-normal excretion of glycolate (53 μg/mg creatinine), but a subsequent specimen did show high glycolate concentration (78 μg/mg creatinine). Two PH II individuals were classified by a history of oxalate/glycerate hyperexcretion at a reference laboratory. Both patients had increases in glycerate that were at least threefold above the upper limit of normal with our new method. In summary, then, four of five PH I patients unresponsive to pyridoxine and both PH II patients were detected by THF extraction. Discussion Our results show that THF extraction of urine (as an adjunct to routine organic acid analysis) significantly improves sensitivity for polar compounds such as glycolate and glycerate and allows even normal excretion to be quantified. The ability to detect normal concentrations of these compounds may improve the utility of the test in discriminating between normal and affected individuals. The efficiency of extraction of glycolate and glycerate, though much greater with THF than other standard solvents, is still incomplete, a fact that decreases precision of analysis. Nevertheless, our data indicate that precision and accuracy with THF extraction are sufficient to readily distinguish PH from normality. Performed after ether extraction, THF extraction requires no additional equipment and little additional effort in laboratories already performing organic acid analysis. The THF extraction strategy presented here compares favorably with other methods involving direct (no extraction) determination (29) or alternative extraction techniques (anion-exchange chromatography) to detect glycolate (30). Reported normal glycerate values, however, vary widely (12)(31). These differences are possibly due to the range of methods used for quantifying glycerate and their standardization. Standardization is particularly difficult with GC-MS assays. Glyceric acid is available commercially as a syrup with significant water content or as its calcium salt, which is insoluble in organic solvents and in the derivatizing agents used. Both forms must be manipulated further to provide a free acid form suitable for standardization. Differences due to standardization do not affect the ratio of increased to normal concentrations of glycerate and would, therefore, not diminish the ability of THF extraction to identify those patients with increased excretion rates of glyceric acid. Interpretation of results must be made with some qualification. First, the technique described here does not distinguish between the d and l forms of glyceric acid. Further investigation is required to confirm excretion of the l isomer, which is specific to PH II. Second, excretion of glycerate and glycolate will likely vary considerably within individuals. For instance, one sample in our study contained high-normal concentrations of glycolate; a second sample was clearly increased. Finally, progressive glomerular damage reduces filtration of glycerate and glycolate, so interpretation must always take into account the extent of residual renal function. With these considerations in mind, THF extraction of glycerate and glycolate as an adjunct to routine organic acid analysis should enable early detection and distinction of PH and avoid the need for liver biopsies in many patients. Figure 1. Open in new tabDownload slide THF enables extraction of oxalate, glycerate, and glycolate from urine. A urine specimen was supplemented with hippurate, lactate, succinate, glycolate, glycerate, oxalate, and ethylphosphonate (internal calibrator) and extracted three times with ether and then with THF. The elution profiles of the first ether extraction and the first THF extraction are shown. Oxalate, glycerate, glycolate, and ethylphosphonate were extracted from urine exclusively with THF. Figure 1. Open in new tabDownload slide THF enables extraction of oxalate, glycerate, and glycolate from urine. A urine specimen was supplemented with hippurate, lactate, succinate, glycolate, glycerate, oxalate, and ethylphosphonate (internal calibrator) and extracted three times with ether and then with THF. The elution profiles of the first ether extraction and the first THF extraction are shown. Oxalate, glycerate, glycolate, and ethylphosphonate were extracted from urine exclusively with THF. Table 1. Recovery of oxalate, glycerate, glycolate, and ethylphosphonate in five successive THF extracts. Compound . Percent extracted by THF . . . . . . 1st . 2nd . 3rd . 4th . 5th . Oxalate 48 14 2.8 0 0 Glycerate 15 18 16 14 12 Glycolate 11 10 9.2 7.4 4.7 Ethylphosphonate 26 23 14 8.6 5.0 Compound . Percent extracted by THF . . . . . . 1st . 2nd . 3rd . 4th . 5th . Oxalate 48 14 2.8 0 0 Glycerate 15 18 16 14 12 Glycolate 11 10 9.2 7.4 4.7 Ethylphosphonate 26 23 14 8.6 5.0 A representative urine specimen containing 250 μg of each of the indicated compounds was extracted five successive times with THF. Extracts were concentrated, derivatized, and analyzed separately. Table 1. Recovery of oxalate, glycerate, glycolate, and ethylphosphonate in five successive THF extracts. Compound . Percent extracted by THF . . . . . . 1st . 2nd . 3rd . 4th . 5th . Oxalate 48 14 2.8 0 0 Glycerate 15 18 16 14 12 Glycolate 11 10 9.2 7.4 4.7 Ethylphosphonate 26 23 14 8.6 5.0 Compound . Percent extracted by THF . . . . . . 1st . 2nd . 3rd . 4th . 5th . Oxalate 48 14 2.8 0 0 Glycerate 15 18 16 14 12 Glycolate 11 10 9.2 7.4 4.7 Ethylphosphonate 26 23 14 8.6 5.0 A representative urine specimen containing 250 μg of each of the indicated compounds was extracted five successive times with THF. Extracts were concentrated, derivatized, and analyzed separately. Figure 2. Open in new tabDownload slide Excretion of oxalate, glycerate, and glycolate in urine from healthy individuals. Urine specimens from healthy children and adults were subjected to the extraction protocol and analyzed. Metabolite concentration (per milligram creatinine) for each individual is plotted vs age. Figure 2. Open in new tabDownload slide Excretion of oxalate, glycerate, and glycolate in urine from healthy individuals. Urine specimens from healthy children and adults were subjected to the extraction protocol and analyzed. Metabolite concentration (per milligram creatinine) for each individual is plotted vs age. Table 2. Application of THF extraction to specimens from patients with PH I or PH II. Patient . Clinical status . Glycolate . . Glycerate . . . . μg/mg creatinine . μg/mL urine . μg/mg creatinine . μg/mL urine . NB PH I 191 55.4 31 9.1 DM PH I 165 67.7 30 12.3 LF PH I 92 4.5 <5 <0.2 AJ PH I 281 117.7 71 29.7 265 92.8 44 15.4 AM PH I 53 19.9 32 12.0 78 37.1 39 18.5 RD PH II 18 1.3 314 22.3 GD PH II 23 11.5 1359 680.9 Normals 14–72 19–115 Patient . Clinical status . Glycolate . . Glycerate . . . . μg/mg creatinine . μg/mL urine . μg/mg creatinine . μg/mL urine . NB PH I 191 55.4 31 9.1 DM PH I 165 67.7 30 12.3 LF PH I 92 4.5 <5 <0.2 AJ PH I 281 117.7 71 29.7 265 92.8 44 15.4 AM PH I 53 19.9 32 12.0 78 37.1 39 18.5 RD PH II 18 1.3 314 22.3 GD PH II 23 11.5 1359 680.9 Normals 14–72 19–115 Random urine specimens were obtained at visits to the Mayo Nephrology Clinic. All patients were 5 years of age or older. The normal range associated with this population (age >5 years) is given in the last row of the table. (To convert μg/mL glycolate and μg/mL glycerate to μmol/L, multiply by 12.99 and 9.35, respectively.) Table 2. Application of THF extraction to specimens from patients with PH I or PH II. Patient . Clinical status . Glycolate . . Glycerate . . . . μg/mg creatinine . μg/mL urine . μg/mg creatinine . μg/mL urine . NB PH I 191 55.4 31 9.1 DM PH I 165 67.7 30 12.3 LF PH I 92 4.5 <5 <0.2 AJ PH I 281 117.7 71 29.7 265 92.8 44 15.4 AM PH I 53 19.9 32 12.0 78 37.1 39 18.5 RD PH II 18 1.3 314 22.3 GD PH II 23 11.5 1359 680.9 Normals 14–72 19–115 Patient . Clinical status . Glycolate . . Glycerate . . . . μg/mg creatinine . μg/mL urine . μg/mg creatinine . μg/mL urine . NB PH I 191 55.4 31 9.1 DM PH I 165 67.7 30 12.3 LF PH I 92 4.5 <5 <0.2 AJ PH I 281 117.7 71 29.7 265 92.8 44 15.4 AM PH I 53 19.9 32 12.0 78 37.1 39 18.5 RD PH II 18 1.3 314 22.3 GD PH II 23 11.5 1359 680.9 Normals 14–72 19–115 Random urine specimens were obtained at visits to the Mayo Nephrology Clinic. All patients were 5 years of age or older. The normal range associated with this population (age >5 years) is given in the last row of the table. (To convert μg/mL glycolate and μg/mL glycerate to μmol/L, multiply by 12.99 and 9.35, respectively.) 1 Danpure CJ, Purdue PE. Primary hyperoxaluria. Scriver CR Beaudet AL Sly WS Valle D eds. The metabolic and molecular bases of inherited disease 7th ed. 1995 : 2385 -2424 McGraw-Hill New York. . 2 Baker PW, Rofe AM, Bais R. Idiopathic calcium oxalate urolithiasis and endogenous oxalate production. Crit Rev Clin Lab Sci 1996 ; 33 : 39 -82. Crossref Search ADS PubMed 3 Chesney R, Friedman A, Gilbert-Barness E. Pathological case of the month: primary oxalosis. Am J Dis Child 1992 ; 146 : 255 -256. PubMed 4 Matthews M, Stauffer M, Cameron EC, Maloney N, Sherard DJ. Bone biopsy to diagnose hyperoxaluria in patients with renal failure. Ann Intern Med 1979 ; 90 : 777 -779. Crossref Search ADS PubMed 5 Yendt ER, Cohanim M. Absorptive hyperoxaluria: a new clinical entity—successful treatment with hydrochlorothiazide. Clin Invest Med 1986 ; 9 : 44 -50. PubMed 6 Danpure CJ, Jennings PR. Peroxisomal alanine:glyoxylate aminotransferase deficiency in primary hyperoxaluria type I. FEBS Lett 1986 ; 201 : 20 -24. Crossref Search ADS PubMed 7 Danpure CJ, Jennings PR, Watts RWE. Enzymological diagnosis of primary hyperoxaluria type I by measurement of hepatic alanine:glyoxylate aminotransferase activity. Lancet 1987 ; i : 289 -291. 8 Marangella M, Petrarulo M, Vitale C, Cossedu D, Linari F. Plasma and urine glycolate assays for differentiating the hyperoxaluria syndromes. J Urol 1992 ; 148 : 986 -989. Crossref Search ADS PubMed 9 Barratt TM, Kasidas GP, Murdoch I, Rose GA. Urinary oxalate and glycolate excretion and plasma oxalate concentration. Arch Dis Child 1991 ; 66 : 501 -503. Crossref Search ADS PubMed 10 Marangella M, Petrarulo M, Bianco O, Vitale C, Finocchiaro P, Linari F. Glycolate determination detects type I primary hyperoxaluria in dialysis patients. Kidney Int 1991 ; 39 : 149 -154. Crossref Search ADS PubMed 11 Williams HE, Smith LH. l-Glyceric aciduria: a new genetic variant of primary hyperoxaluria. N Engl J Med 1968 ; 278 : 233 -239. Crossref Search ADS PubMed 12 Chlebeck PT, Milliner DS, Smith LH. Long-term prognosis in primary hyperoxaluria type II. Am J Kidney Dis 1994 ; 23 : 255 -259. Crossref Search ADS PubMed 13 Rizzoni G, Broyer M, Brunner FP, Brynger H, Challah S, Kramer P, et al. Combined report on regular dialysis and transplantation of children in Europe. Proc Eur Dial Transplant Assoc Eur Ren Assoc 1984 ; 21 : 69 . 14 Latta K, Brodehl J. Primary hyperoxaluria type I. Eur J Pediatr 1990 ; 149 : 518 -522. Crossref Search ADS PubMed 15 Seargeant LE, deGroot GW, Dilling LA, Mallory CJ, Haworth JC. Primary oxaluria type II (l-glyceric aciduria): a rare cause of nephrolithiasis in children. J Pediatr 1991 ; 118 : 912 -914. Crossref Search ADS PubMed 16 Hicks NR, Cranston DW, Charlton CA. Fifteen year followup of hyperoxaluria type II [Letter]. N Engl J Med 1983 ; 309 : 796 . 17 Leumann E, Hoppe B, Neuhaus T. Management of primary hyperoxaluria: efficacy of oral citrate administration. Pediatr Nephrol 1993 ; 7 : 207 -211. Crossref Search ADS PubMed 18 Calzavara P, Marangella M, Petrarulo M, Ballanti P, Bonucci E, Calconi G, et al. Long-term survival on renal replacement therapy for primary hyperoxaluria type I. Nephron 1993 ; 63 : 217 -221. Crossref Search ADS PubMed 19 Cochat P, Sharer K. Should liver transplantation be performed before advanced renal insufficiency in primary hyperoxaluria type I. Pediatr Nephrol 1993 ; 7 : 212 -218. Crossref Search ADS PubMed 20 Watts RWE, Danpure CJ, De Pauw L, Toussaint C, . and the European Study Group on Transplantation on Hyperoxaluria Type I. Combined liver–kidney and isolated liver transplantations for primary hyperoxaluria type I: the European experience. Nephrol Dial Transplant 1991 ; 6 : 502 -511. Crossref Search ADS PubMed 21 Danpure CJ, Jennings PR. Further studies on the activity and subcellular distribution of alanine:glyoxylate aminotransferase in the livers of patients with primary hyperoxaluria type I. Clin Sci 1988 ; 75 : 315 -322. Crossref Search ADS PubMed 22 Danpure CJ, Jennings PR, Fryer P, Purdue PE, Allsop J. Primary hyperoxaluria type I: genotypic and phenotypic heterogeneity. J Inherit Metab Dis 1994 ; 17 : 487 -499. Crossref Search ADS PubMed 23 Kasidas GP, Rose GA. A new enzymatic method for the determination of glycolate in urine and plasma. Clin Chim Acta 1979 ; 96 : 25 -36. Crossref Search ADS PubMed 24 Petrarulo M, Marangella M, Cosseddu D, Linari F. High-performance liquid chromatographic assay for l-glyceric acid in body fluids. Application in primary hyperoxaluria type 2. Clin Chim Acta 1992 ; 211 : 143 -153. Crossref Search ADS PubMed 25 Kasidas GP. Assay of oxalate and glycollate in urine. Rose GA eds. Oxalate metabolism in relation to urinary stones 1988 : 7 -26 Springer Verlag London. . 26 Danpure CJ. Molecular and clinical heterogeneity in primary hyperoxaluria type 1. Am J Kidney Dis 1991 ; 27 : 366 -369. 27 Rimoldi M, Bergomi P, Romeo A, Didonato S. A new stable-isotope dilution method for measurement of orotic acid using solvent-extracted urine. J Inherit Metab Dis 1994 ; 17 : 243 -244. Crossref Search ADS PubMed 28 Reed AH, Henry RJ, Mason WB. Influence of statistical method used on the resulting estimate of normal range. Clin Chem 1971 ; 17 : 275 -284. Crossref Search ADS PubMed 29 Morganstern BZ, Milliner DS, Murphy ME, Simmons PS, Moyer TP, Wilson DM, Smith LH. Urinary oxalate and glycolate excretion patterns in the first year of life: a longitudinal study. J Pediatr 1993 ; 123 : 248 -251. Crossref Search ADS PubMed 30 Chalmers RA, Watts RWE. The quantitative extraction and gas–liquid chromatographic determination of organic acids in urine. Analyst 1972 ; 97 : 958 -967. Crossref Search ADS PubMed 31 Hoffmann G, Aramaki S, Blum-Hoffmann E, Nyhan WL, Sweetman L. Quantitative analysis for organic acids in biological samples: batch isolation followed by gas- chromatographic–mass spectrometric analysis. Clin Chem 1989 ; 35 : 587 -595. Crossref Search ADS PubMed © 1997 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Apolipoprotein E genotyping by capillary electrophoretic analysis of restriction fragments, De Bellis, Gianluca;Salani,, Giuliana;Panigone,, Silvia;Betti,, Ferruccio;Invernizzi,, Luigia;Luzzana,, Massimo
doi: 10.1093/clinchem/43.8.1321pmid: N/A
Abstract We present the genotyping of apolipoprotein (apo) E by means of restriction fragment analysis of amplified genomic DNA by high-performance capillary electrophoresis and a replaceable non-gel-sieving matrix. This procedure streamlines the genotyping of apo E in large-scale population studies because of the automation and speed of capillary electrophoresis. indexing terms: genotype determination, Alzheimer disease, high-performance capillary electrophoresis Apolipoprotein (apo) E is an important constituent of several plasma lipoproteins, mainly VLDL and HDL and chylomycrons.1 It is involved in the redistribution of lipids in the liver and is implicated in growth and repair of injured neurons in the nervous system (1). Apo E has been associated with the risk of developing cardiovascular diseases and in familial type III hyperlipoproteinemia (2). More recently a strong association between apo E and Alzheimer disease has been demonstrated (3). Human apo E exists in three main isoforms (E2, E3, and E4) related to two polymorphic sites at codons 112 and 158 of the gene located on chromosome 19 (Table 1 ). These isoforms arise from three alleles (ε2, ε3, and ε4 respectively) combined in six different genotypes. The E4 isoform has been associated with Alzheimer disease as a major risk factor. In particular in late-onset Alzheimer subjects with known familial occurrence, the ε4 allele frequency is much higher than in age-matched individuals, whereas ε2 is much lower (4). Furthermore the age of onset of the disease is related to the ε4 allele dose (3). Over 40 independent studies discussing apo E/Alzheimer association in different populations have been published so far with comparable results (5). Alzheimer disease has a severe impact in the elderly population worldwide. Large studies to follow the elderly population with respect to this pathology are currently under way, as are clinical trials to test drugs with benefical effects on affected individuals. This makes the typing of apo E isoforms very important in Alzheimer studies. An interesting recent review focused on apo E relevance in laboratory medicine and reported the complete range of analytical techniques devoted to the apo E polymorphism determination (6). Phenotyping is usually performed by means of isoelectrofocusing techniques, but these present several drawbacks in their application to this specific analytical problem. Apo E genotyping has been developed to avoid such problems. Several recent papers dealt with the analysis of the apo E genotype by means of PCR amplification from genomic DNA (6). Most of the cited studies involved restriction fragment analysis of the amplified region encompassing codons 112 and 158 of the apo E gene (7)(8). This procedure has been widely adopted, although it has some drawbacks particularly related to complex electrophoretic pattern due to partial enzymatic digestion and to the requirement of large quantities of amplified DNA (9). Modified procedures have been proposed to overcome such problems (9)(10). Here we present the identification of the apo E genotype by means of capillary electrophoresis analysis. A relatively viscous sieving media is required to achieve a good separation of the relatively small DNA fragments (48–91 bp) to be analyzed. High-performance capillary electrophoresis (HPCE) with cross-linked polyacrylamide gel offers the best performance in terms of resolution, and recently two groups (11)(12) proposed such a technique for apo E genotyping. However, such a sieving matrix cannot be replaced. Several non-gel-sieving media, prepared from polymers, have been proposed to separate DNA molecules. These have a key feature in that they are replaceable after each separation, thus performing very reproducibly from run to run. By using methyl cellulose as a replaceable sieving matrix, the digestion fragments are sized in a very short time (10 min) with a commercial automated apparatus for capillary electrophoresis with UV on-line detection. We present results demonstrating the speed, sensitivity, and reliability of this analytical procedure. Materials and Methods The preparation of the sample for the determination of the apo E genotype followed a well-estabilished procedure (7)(8) with minor modifications. Samples (178) were processed at different times. DNA extraction from whole blood was performed with a commercial kit based on chromatographic columns. This material was amplified, giving the expected DNA fragment. The amplified samples were checked for purity and quantity on a standard agarose gel (10 g/L) and digested with CfoI, an isoschizomer of HhaI. dna extraction and apo e amplification DNA was extracted and purified from 200 μL of whole blood, drawn from consenting individuals, with the blood DNA extraction kit (Qiagen, Chatsworth, CA). DNA was collected in 200 μL of water (25–150 ng/L) and stored at 4 °C. Oligonucleotides P1 and P2 (7) used for the amplification of the polymorphic region were purchased from Genset (Paris, France), diluted to 20 μmol/L, and stored at −20 °C. The DNA amplification was performed in a thermal cycler (Perkin-Elmer, Norwalk, CT) in a total volume of 100 μL. The amplification mixture contained Taq buffer diluted 1:10 as suggested by the manufacturer, deoxynucleotides (200 μmol/L each), 40 pmol of each primer, and 8 μL of purified genomic DNA. The mixture was overlaid with mineral oil and heated at 95 °C for 10 min. Taq polymerase (2.5 U; Finnzyme, Espoo, Finland) was then added and the samples were subjected to five cycles as follows: 95 °C for 1 min and 72 °C for 3 min and then to 30 cycles as follows: 95 °C for 1 min, 60 °C for 1 min, and 72 °C for 1 min. Samples were then incubated at 72 °C for 10 min and stored at 4 °C. dna digestion and purification After the amplification, 40 U of CfoI (Boehringer Mannheim, Mannheim, Germany) were added and the samples were digested overnight at 37 °C. Before digestion, the sample can be purified by means of Microspin columns (S200; Pharmacia, Uppsala, Sweden) to remove nucleotides, primers, and PCR buffer. The digested sample was then isopropanol-precipitated or ultrafiltered (we recommend this procedure) by means of MC 5000 columns (Millipore, Bedford, MA), washing two times with 400 μL of water. The resulting solution was adjusted to 20 μL with water. capillary electrophoresis analysis Analysis was performed on a BioFocus 3000 capillary electrophoresis system (Bio-Rad, Hercules, CA) by means of a standard cartridge containing a 50 cm × 50 μm coated capillary tube thermostated at 20 °C. The capillary was washed with TBE 1× (89 mmol/L Tris base, 89 mmol/L boric acid, and 1 mmol/L EDTA disodium salt) and filled with a 12 g/L solution of Methocel (Methylcellulose high density; Fluka, Buchs, Switzerland) in TBE 1×. This solution was prepared as follows: 1 g of Methocel was slowly added to water (50 mL) at 90 °C. The solution was allowed to cool at room temperature under constant stirring for 30 min and then cooled at 4 °C in a refrigerator, continuing the stirring. After 2 h it was centrifuged, filtered through a 0.45-μm filter, and stored at 4 °C. The solution was mixed with TBE 10×, and adjusted with water to a final concentration of 12 g/L in TBE 1×. Samples were electroinjected at 10 kV for 5 s. Restriction fragments were separated during a 10-min run at 15 kV and detected at 260 nm. Results The digestion of the amplified region encompassing the two polymorphic sites of interest generates a number of fragments. Fig. 1 A shows the restriction map for this region with respect to the sequence of the three most common alleles. The expected electrophoretic pattern for the six resulting genotypes is presented in Fig. 1B . Only the DNA fragments ranging from 48 to 91 bp are diagnostic for the genotyping of this locus. After some trials we selected methylcellulose (high viscosity) as the replaceable sieving media to be used in capillary electrophoresis analysis of these relatively short DNA fragments. We optimized the separation variables by means of a DNA size marker prepared by CfoI digestion of a commercial preparation of pUC18 (fragments ranging from 393 to 21 bp). This DNA ladder was analyzed by changing several variables (applied voltage, temperature, polymer concentration, sample injection conditions, capillary tube length, diameter, and coating). A standard 50 cm × 50 μm coated capillary column (Bio-Rad) was found to be appropriate for high resolution and speed of analysis. Alternatively, the analysis was performed on a 50 cm × 50 μm DB-1 coated capillary column (J&W Scientific, Folsom, CA). As shown in Fig. 2 , the diagnostic 48-, 63-, 72-, 82-, and 91-bp bands are well resolved in <10 min. All six common genotypes can be easily discriminated. The run-to-run reproducibility was tested by injecting the same sample 30 times, refilling the capillary tube each time. The CV for the retention time is 2.2%. CV for peak area is 21.7%. With an external calibrator (bromophenol blue), the CV for normalized peak area is 5.6%. The filling of the capillary takes 5 min (50 cm × 50 μm column) and therefore the reuse of the sieving matrix would be desirable to speed up the analysis. We tested this possibility by injecting the same sample several times and performing the analysis as explained without sieving matrix reneval. The first six injections gave a CV for the retention time within that reported, indicating that multiple use of the matrix is feasible. Discussion The reliability of apo E genotyping by restriction digestion and polyacrylamide gel electrophoresis (PAGE) has been recently discussed. Some important drawbacks can affect the analysis. Because of partial digestion, “ghost” bands appear in PAGE, biasing the correct genotyping. Bands having different intensities have been indicated as a possible source of problems during ethidium bromide/PAGE analysis (9). In addition, relevant quantities of amplified DNA should be loaded on the gel because of the intrinsic low sensitivity of ethidium bromide staining with small DNA fragments. Appel et al. (9) recently proposed an alternative procedure requiring radioactive labeling. In an attempt to overcome such problems we used HPCE for the analysis of the restriction fragments generated by HhaI (CfoI) digestion. This technique was recently proposed by Baba et al. (11) and Schlenck et al. (12) for apo E genotyping. Both used a cross-linked polyacrylamide gel as separation medium, thus gaining very high resolution (12). UV- (11) or laser-induced fluorescence (12) was used as detection system. The resolution we obtained with methylcellulose was lower than that gained by polyacrylamide (12), although it is sufficient to discriminate all six common apo E genotypes. However, such a sieving matrix yielded very reproducible results because of the possibility of its renewal after runs, whereas polyacrylamide capillary columns decrease their performance during their lifetime. We used UV detection, gaining good sensitivity. PCR usually gave 400 ng of amplified DNA that was precipitated, digested, desalted, and electroinjected several (10–30) times, yielding reproducible electropherograms similar to those shown in Fig. 2 . This was achieved without the need for an expensive laser apparatus for fluorescence detection, although this grants a much higher sensitivity (12). We observed a large difference comparing the signal intensity we obtained with that shown in ref. 11. This is probably due to the accurate desalting of digested samples performed before electroinjection. This allows an on-line concentration, particularly for smaller fragments. HPCE, giving a quantitative estimate of each band by UV on-line detection, helps in preventing genotyping errors due to partial digestion. In fact, peak integration allows the comparison of the relative intensities of the bands on a numerical basis. Although the peak intensity ratio can vary because of the inherent variability of electrokinetic injection, the presence of “ghost” bands, like that in Fig. 2F (the shoulder close to the 63-bp fragment), can be easily detected and correctly classified. Samples that are difficult to discriminate, because of partial digestion, can be directly redigested and analyzed because of the extremely small sample quantities required by the capillary electrophoresis analysis. Sample loading by electroinjection has proven to be very efficient, although it requires a preliminary optimization with respect to the actual concentration of the sample. The precipitation performed after digestion yields very concentrated samples that must be injected for a very short time to avoid overload and loss of resolution. Despite this fact, it is very simple to determine the genotype in very faint samples by means of longer injection time, thus overcoming the sensitivity problem. The injection of the same sample 30 times demonstrates the sensitivity of this approach and its reliability (low CV for retention time). The speed of the analysis can be further increased by shortening the capillary tube. In a 20-cm × 50-μm tube, acceptable separation was gained in <3 min (data not shown). Preparation of the sieving matrix has proven to be critical to the final result. The reported procedure must be followed carefully as explained to yield consistent results. The results achieved with this method in our laboratory, during a large-scale apo E genotype study, are extremely positive. On some occasions, in partially digested samples, genotyping by PAGE and by HPCE were in disagreement, but after further digestion, the genotyping by HPCE was confirmed. In conclusion, this procedure could be of interest to clinical laboratories involved in large-scale apo E genotype analysis (6), thanks to the high degree of automation attainable with HPCE combined with the speed, sensitivity, and reliability of the proposed analytical protocol. 1 Nonstandard abbreviations: apo, apolipoprotein; HPCE, high-performance capillary electrophoresis; TBE, Tris–boric acid–EDTA buffer; and PAGE, polyacrylamide gel electrophoresis. Table 1. Correspondence among apo E protein isoforms, amino acid and codon composition at polymorphic sites, and genotypes. Isoform . Position 112 . Position 158 . Allele . E2 Cys (TGC) Cys (TGC) ε2 E3 Cys (TGC) Arg (CGC) ε3 E4 Arg (CGC) Arg (CGC) ε4 Isoform . Position 112 . Position 158 . Allele . E2 Cys (TGC) Cys (TGC) ε2 E3 Cys (TGC) Arg (CGC) ε3 E4 Arg (CGC) Arg (CGC) ε4 Table 1. Correspondence among apo E protein isoforms, amino acid and codon composition at polymorphic sites, and genotypes. Isoform . Position 112 . Position 158 . Allele . E2 Cys (TGC) Cys (TGC) ε2 E3 Cys (TGC) Arg (CGC) ε3 E4 Arg (CGC) Arg (CGC) ε4 Isoform . Position 112 . Position 158 . Allele . E2 Cys (TGC) Cys (TGC) ε2 E3 Cys (TGC) Arg (CGC) ε3 E4 Arg (CGC) Arg (CGC) ε4 Figure 1. Open in new tabDownload slide (A) Restriction map (CfoI endonuclease) of the region encompassing the two polymorphic sites at codons 112 and 158 of the human apo E gene; (B) electrophoretic patterns expected for the six different genotypes arisng from the three most common alleles (ε2, ε3 and ε4). (A) The expected restriction site for each of the three most common alleles is represented by a vertical bar. The expected size (in bp) for the resulting fragments >30 bp is reported. Figure 1. Open in new tabDownload slide (A) Restriction map (CfoI endonuclease) of the region encompassing the two polymorphic sites at codons 112 and 158 of the human apo E gene; (B) electrophoretic patterns expected for the six different genotypes arisng from the three most common alleles (ε2, ε3 and ε4). (A) The expected restriction site for each of the three most common alleles is represented by a vertical bar. The expected size (in bp) for the resulting fragments >30 bp is reported. Figure 2. Open in new tabDownload slide Capillary electrophoresis analysis of the six common genotypes [(A) ε2/ε4; (B) ε3/ε3; (C) ε4/ε4; (D) ε3/ε4; (E) ε2/ε3; (F) ε2/ε2]. The size of the diagnostic bands is reported. Fragments smaller than 48 bp (34 bp and smaller) are not resolved in these conditions and thus the corresponding peaks are not labeled. X-axis, retention time (min); y-axis, absorbance. Figure 2. Open in new tabDownload slide Capillary electrophoresis analysis of the six common genotypes [(A) ε2/ε4; (B) ε3/ε3; (C) ε4/ε4; (D) ε3/ε4; (E) ε2/ε3; (F) ε2/ε2]. The size of the diagnostic bands is reported. Fragments smaller than 48 bp (34 bp and smaller) are not resolved in these conditions and thus the corresponding peaks are not labeled. X-axis, retention time (min); y-axis, absorbance. We gratefully acknowledge Telethon (grant E213) for partial financial support. 1 Mahley RW, Innerarity TL, Rall SC, Weisgraber KH, Taylor. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science 1988 ; 240 : 622 -630. Crossref Search ADS PubMed 2 Breslow JL, McPherson J, Sanciacomo TR, Third JLHC, Tracy T, Glueck CJ. Studies of familial type III hyperlipoproteinemia using as a genetic marker the apoE phenotype E2/2. J Lipid Res 1982 ; 23 : 1224 -1235. PubMed 3 Corder EH, Saunders AM, Strittmatter WJ, Schemechel DE, Gaskell PC, Small GW, et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 1993 ; 261 : 921 -923. Crossref Search ADS PubMed 4 Corder EH, Saunders AM, Risch NJ, Strittmatter WJ, Schemechel DE, Gaskell PC, et al. Protective effect of apolipoprotein E type 2 allele decreases risk of late onset of Alzheimer’s disease. Nature 1994 ; 7 : 180 -184. 5 Strittmatter WJ, Roses AD. Apolipoprotein E and Alzheimer disease. Proc Natl Acad Sci U S A 1995 ; 92 : 4725 -4727. Crossref Search ADS PubMed 6 Siest G, Pillot T, Regis-Bailly A, Leininger-Muller B, Steinmetz J, Galteau MM, Visvikis S. Apolipoprotein E: an important gene and protein to follow in laboratory medicine [Review]. Clin Chem 1995 ; 41 : 1068 -1086. Crossref Search ADS PubMed 7 Kontula K, Aalto-Setala K, Kuusi T, Hamalainen L, Syvanen A. Apolipoprotein E polymorphism determined by restriction enzyme analysis of DNA amplified by polymerase chain reaction: convenient alternative to phenotyping by isoelectric focusing. Clin Chem 1990 ; 36 : 2087 -2092. Crossref Search ADS PubMed 8 Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with HhaI. J Lipid Res 1990 ; 31 : 545 -548. PubMed 9 Appel E, Eisenberg S, Roitelman J. Improved PCR amplification/HhaI restriction for unambiguous determination of apolipoprotein E alleles. Clin Chem 1995 ; 41 : 187 -190. Crossref Search ADS PubMed 10 Reymer WA, Gronemeyer BE, van de Burg R, Kastelein JJP. Apolipoprotein E genotyping on agarose gels [Tech Brief]. Clin Chem 1995 ; 41 : 1046 -1047. Crossref Search ADS PubMed 11 Baba Y, Tomisaki R, Sumita C, Tsuhako M, Miki T, Ogihara T. High resolution separation of PCR product and gene diagnosis by capillary electrophoresis. Biomed Chromatogr 1994 ; 8 : 291 -293. Crossref Search ADS PubMed 12 Schlenck A, Visvikis S, O’Kane M, Siest G. High resolution separation of PCR product and gene diagnosis by capillary electrophoresis. Biomed Chromatogr 1996 ; 10 : 48 -50. Crossref Search ADS PubMed © 1997 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
A homogeneous method for genotyping with fluorescence polarizationGibson, Neil, J;Gillard, Helen, L;Whitcombe,, David;Ferrie, Richard, M;Newton, Clive, R;Little,, Stephen
doi: 10.1093/clinchem/43.8.1336pmid: N/A
Abstract We combined the amplification refractory mutation system (ARMS™) and fluorescence polarization (FP) to give a homogeneous genomic DNA genotype analysis method. Oligonucleotide probes labeled with the fluorescein dyes fluorescein isothiocyanate and 5-([4,6-dichlorotriazin-2-yl]amino)fluorescein and the rhodamine dye 6-carboxyrhodamine were included in amplification mixes and were annealed to PCR products after amplification. Hybridization was accompanied by an increase in the FP of the probe. We demonstrated homogeneous genotyping by analyzing human DNA samples for ΔF508 mutation status of the cystic fibrosis transmembrane conductance regulator gene. The genotypes determined with the method described herein were in full agreement with those obtained by the conventional application of ARMS. We also demonstrated the simultaneous detection of two PCR products in a single reaction. The assay method described is homogeneous and so obviates the necessity to open reaction vessels after amplification. This therefore eliminates PCR carryover contamination. indexing terms: DNA amplification, cystic fibrosis, genetic variants PCR is the cornerstone of several DNA diagnostic methods. These may be applied to the detection of microbial and viral contamination and infections, to the diagnosis of human inherited diseases, to predispositions to pathological states, and to the screening for cancers and residual disease after therapy (1)(2). During PCR, an enrichment of the DNA target by 1010–1012-fold is routinely achieved. Because the products of a reaction become the substrates in subsequent cycles of amplification with the same primer system, it is essential that there be no crossover contamination of amplicon to subsequent unamplified reaction mixes; otherwise, false-positive results will be generated. Crossover contamination prevention is usually addressed by isolating each stage of the PCR process so that setup, amplification, and detection are carried out in separate facilities with dedicated protective clothing and equipment for each area (3). Another physical precaution is the use of laminar airflows (3). Biochemical means of preventing PCR product carryover contamination include UV photoinactivation after psoralen (4)(5)(6) or isopsoralen (7)(8)(9) treatment, or more commonly, substitution of dUTP for dTTP during PCR with subsequent PCRs being pretreated with uracil DNA glycosylase (6)(9)(10)(11). An alternate way of preventing carryover contamination is to detect PCR products without opening the reaction tube after amplification. If product detection is by spectroscopic means in a sealed vessel, then the tube can be discarded after analysis, removing the risk of release of amplicon. Another advantage is that homogeneous amplicon detection avoids electrophoretic separations and should be easily adaptable to automated analysis. Fluorescence polarization (FP) detects changes in the molecular volume of a fluorophore (12) and is also capable of detecting nucleic acid hybrids in solution (A. J. Garman, Zeneca Pharmaceuticals, personal communication, (13)(14)(15)(16)).1 We show that there is a measurable increase in the FP of a fluorescently labeled oligonucleotide probe when specifically hybridized to a PCR amplicon. The amplification refractory mutation system (ARMS™) is a powerful technique for detecting mutations and polymorphisms in DNA (17) and is in part a significant improvement of the PCR. Primers that are mismatched at their 3′ termini relative to the template genomic DNA are not readily extended by, e.g., Taq DNA polymerase. Two ARMS reactions are usually performed on a DNA sample. These involve primer sets that contain a common primer in both reactions and a primer specific for one or the other allele in the separate reactions. The three possible results, depending on the genotype, are: (a) DNA that is homozygous for one allele gives a product in the reaction with the primer specific for that allele, (b) DNA from a homozygote for the other allele gives a product with the other allele-specific primer, and (c) there is product in both of the reactions when DNA from a heterozygote is tested. Here, we demonstrate that the inclusion of a fluorescently labeled probe with the ARMS primers allows the homogeneous genotyping of DNA samples. Detection of the ARMS products is achieved by measuring the change in FP of the probe bound to the amplicon. Because the sequence of the amplicon produced by each test differs only at the location of the allelic variation, one fluorescent probe can be used to detect both amplicons. To demonstrate the use of FP in the homogeneous detection of ARMS products we adapted an ARMS test for the cystic fibrosis transmembrane conductance regulator (CFTR) gene ΔF508 mutation (18). Materials and Methods Instrumentation. FP was measured with the microtiter plate reader Fluorolite FPM-2 (Jolley Research, Round Lake, IL) or a SPEX Fluoromax (SPEX Industries, Edison, NJ) fluorometer modified by the inclusion of polarizing prisms in the excitation and emission light beams. Black microtiter plates suitable for reading fluorescence were from Costar (Cambridge, MA). Reagents and solutions. PBS and 5-([4,6-dichlorotriazin-2-yl]amino)fluorescein (DTAF) were from Sigma Chemical Co., Poole, UK; Sephadex G-25, NAP-10 columns, and dNTPs from Pharmacia Biotechnology, Piscataway, NJ; Taq DNA polymerase (AmpliTaq) from P. E. Applied Biosystems, Foster City, CA; Nusieve agarose from FMC BioProducts, Rockland, ME; and ØX174 DNA/HaeIII digest molecular size marker from Life Technologies, Gaithersburg, MD. All other chemicals were from Aldrich Chemical Co., Milwaukee, WI. Oligonucleotides with the 5′ labels 6-carboxyrhodamine (ROX), fluorescein, and 6-aminohexylphosphate were from Oswel Research Products (Southampton, UK). The remaining oligonucleotides were synthesized on a P. E. Applied Biosystems 394 DNA/RNA synthesizer with P. E. Applied Biosystems reagents according to the supplier’s protocols and were purified by size exclusion filtration through NAP-10 columns. The DTAF labeling buffer was 0.10 mol/L NaCl, 0.10 mol/L sodium carbonate, pH 8.00; HPLC buffer A 0.1 mol/L ammonium chloride; HPLC buffer B 800 mL/L HPLC buffer A, 200 mL/L acetonitrile; ARMS buffer 1.2 mmol/L MgCl2, 10 mmol/L Tris-HCl, 50 mmol/L KCl, 0.1 g/L gelatin, pH 8.3; Tris–borate–EDTA electrophoresis buffer (TBE) 134 mmol/L Tris, 74.9 mmol/L boric acid, 2.55 mmol/L EDTA, 0.1 mg/L ethidium bromide; gel loading buffer 300 mL/L glycerol and 1 g/L bromphenol blue in TBE; and PBS 10 mmol/L phosphate, 2.7 mmol/L KCl, 0.137 mol/L NaCl, pH 7.4. Probe design. Probe sequences were designed with the primer analysis software Oligo 5 (National Biosciences, Plymouth, MN). Specifically, probes were to hybridize to one of the strands of the target amplicon after amplification but not during the extension phase of the PCR. If probes were hybridized during PCR extension they would be degraded by the 5′-exonuclease function of Taq DNA polymerase (19) as the enzyme extended the primer annealed to the same strand. Probe sequences were selected to have Tms between 12 and 15 °C below the temperature of the PCR extension phase. To facilitate probe hybridization, regions of the gene sequence capable of forming stable secondary structures were rejected as binding sites. An internal structure search was performed for candidate probe sequences plus 10 nucleotides up- and downstream. Only sequences that included hairpins with stems of three bp or less and where the ΔG value for loop formation was >−1.0 kcal/mol were considered. Probe labeling. The 6-aminohexylphosphate-labeled oligonucleotide probe was labeled with DTAF as follows: One milliliter of DTAF (500 mg, 940 nmol) in a mixture of acetonitrile and PBS (1:1 by vol) was added to probe 1 (Table 1 ) (21.3 nmol) in DTAF labeling buffer (500 μL). After 1 h the reaction was quenched by adding diisopropylamine (10 mg, 10 mmol) and separated on Sephadex G-25 (15 mm × 400 mm) eluted in water. The faster eluting fraction was resolved by reversed-phase HPLC [Waters μbondapak C18 (Waters Corp., Milford, MA) gradient elution profile with buffers A and B: time 0 min (100% HPLC buffer A, 0% HPLC buffer B), 2.5 min (100% A, 0% B), 7.5 min (0% A, 100% B), 13 min (0% A, 100% B), 18 min (100% A, 0% B), and 20 min (100% A, 0% B, stop)]. Several yellow-colored peaks were obtained and desalted on Sephadex G-25 as above. One peak gave a short retention time and was designated probe 2. This was confirmed to be the DTAF-labeled oligonucleotide (UV λmax 260 nm with minor peak at 495 nm). In addition, the FP of free probe 2 = 51 mP, FP of probe 2 annealed to excess probe 3 = 170 mP. DNA samples. ΔF508/ΔF508 and ΔF508/+ cell line DNA were obtained from the Coriell Institute for Medical Research (Camden, NJ), and +/+ DNA was isolated from peripheral blood leukocytes from an unaffected individual as described previously (18). Each sample was diluted to a concentration of 10 mg/L (10 ng μL−1) before use. The mutant allele was amplified with primers 1 and 2; the normal allele was amplified with primers 1 and 3. Primers 4 and 5 were included in both reactions to generate an apolipoprotein (apo) B gene amplification control amplicon. All primers (Table 2 ) were used at a concentration of 1 μmol/L; the PCR mixes included dNTPs (100 μmol/L each) and probe (30 nmol/L) in ARMS buffer (150 μL total volume). DNA (150 ng) and a mineral oil overlay were added. The samples were heated at 94 °C for 5 min; then Taq DNA polymerase (3 U) was added. Thirty-five rounds of thermal cycling were carried out (94 °C 1 min, 58 °C 1 min, 72 °C 1 min). An aliquot (10 μL) was removed after PCR for gel analysis and the remainder was examined by FP. Gel electrophoresis. Three percent gels (Nusieve agarose:agarose, 3:1) were prepared and electrophoresed in TBE buffer. ARMS reaction aliquots were mixed with gel loading buffer (10 μL) and electrophoresed against ØX174/HaeIII digest DNA size markers (500 ng). Gels were visualized by UV transillumination and photographed. FP. FP probes were annealed to the target amplicon by heating at 94 °C for 5 min, snap cooling on ice for 5 min, and equilibrating at 25 °C for 30 min. The samples were then either transferred to a black microtiter plate for automated reading on the Fluorolite FPM-2 or were transferred to quartz microcuvettes (100 μL) for manual reading on the Fluoromax. Fluorescein- and DTAF-labeled probes were excited at 495 nm and the fluorescence was detected at 525 nm; ROX-labeled probes were excited at 585 nm and the fluorescence was detected at 605 nm. Four fluorescence readings were taken to calculate FPs: IVV, reading with excitation and emission polarizers vertical; IVH, reading with excitation polarizer vertical and emission polarizer horizontal; IHV, reading with excitation polarizer horizontal and emission polarizer vertical; and, IHH, reading with excitation and emission polarizers horizontal. \[FP{=}(\mathit{I}_{\mathrm{VV}}\mathrm{{-}}\mathit{G\ {\cdot}\ I}_{\mathrm{VH}}\mathrm{)/(}\mathit{I}_{\mathrm{VV}}\mathrm{{+}}\mathit{G}\mathrm{\ {\cdot}\ }\mathit{I}_{\mathrm{VH}}\mathrm{)}\] \[\mathit{G}\mathrm{{=}}\mathit{I}_{\mathrm{HV}}\mathrm{/}\mathit{I}_{\mathrm{HH}}\] The G value is a correction factor that allows for the different light transmission characteristics of the two polarizers in the vertical and horizontal orientations. FPs are quoted as mP values. Results Genotyping by ARMS and FP. We designed a homogeneous ARMS test to analyze the ΔF508 and F508 (+) alleles of the CFTR gene (20). A panel of DNAs (+/+, +/ΔF508, and ΔF508/ΔF508) were used to generate ARMS amplicons (18) in the presence of fluorescein-labeled probe 5. Control reactions with no DNA were also run (see Fig. 1 ). An apo B-derived amplification control product was also generated with primers 4 and 5 but was not detected by FP. Gel electrophoresis of aliquots of the ARMS reactions showed amplicons of the expected size in all lanes; no additional products were seen (see Fig. 1 ). The remainder of the ARMS reactions were subjected to a heat–cool cycle to anneal probe 5 to the ARMS amplicons and the FP of the probe was determined on the SPEX Fluoromax for each of the 24 samples. The results are shown in Fig. 2 . The positive samples give a mean FP of 99 mP (n = 12, SD 2.4); the negative samples give a mean FP of 68 (n = 12, SD 2.1). By using a t-distribution test, t = 32.3 for these data sets. For 22 degrees of freedom t0.95 = 1.72 and t0.99 = 2.51, which shows that the mean FPs of the two data sets are significantly different at a 99% level of confidence. A 99% confidence interval, rounded outwards, for a future observation from the positive sample population is (91, 107; 99 ± 8), and similarly from the negative sample population is (61, 75; 68 ± 7). These intervals do not overlap and therefore demonstrate a clear separation between the two populations (see Fig. 3 ). Multiplex detection of amplicons. Simultaneous detection of several fluorophores in a mixture is possible by selecting fluorophores with discrete excitation and emission frequencies. Negative ARMS reactions are therefore distinguishable from PCR failures by detecting the amplification control in samples where no diagnostic ARMS amplicon is found. Multiplex, homogeneous detection of PCR products was demonstrated with the DTAF-labeled probe 2 for the ARMS amplicon in combination with the ROX-labeled probe 4 for the apo B amplicon. These two probes were used in three amplification mixes whereby the first contained both the ARMS and apo B products, the second only the ARMS product, and the third only the apo B product. Each experiment was in duplicate. The average FP of each probe in each amplification mixture is shown in Fig. 4 . These results show that the two FP probes can be used to detect their target amplicons in the same solution. Stability of probe–product hybrids and specificity of hybrid formation. Probe 2 was annealed to both the ARMS and apo B amplicons and the time course of the change in FP of the probe was monitored on the FPM-2 as shown in Fig. 5 . Hybridization of the probe to the apo B amplicon was not observed. These data demonstrate that the probe anneals to its target selectively where it forms a stable hybrid. Discussion The ratio of bound probe to unbound probe in a sample can be determined by FP. Maximal differences in FP values are obtained between samples that contain no bound probe and those in which all of the probe is bound to its target. Thus, FP can be used as a complementary technique to PCR to detect the generation of an amplicon. In ARMS, amplification efficiency is dramatically reduced when a primer is mismatched at the 3′ terminus relative to the genomic target. Only when primers are matched to the genomic target is a detectable amplicon generated. Detection of a mutation is therefore reduced to simply determining whether or not a sample contains an amplicon because its generation of an amplicon is dependent on the genomic DNA sequence. For FP detection the threshold concentration of product is dependent on the probe concentration. Sufficient target needs to be formed for all of the probe to be able to bind to avoid a mixture of bound and unbound probe. In such a situation there would be lower net FP than in a sample in which the probe was entirely bound. A partially bound sample is still usually distinguishable from an unbound sample but there will be a diminution in the separation of the polarizations of the positive and negative samples. For this reason we set the probe concentration below the expected concentration of any product formed. If only low concentrations of diagnostic ARMS amplicon are formed, giving a low FP, we would expect that a correspondingly small amount of amplification control would be formed. Such a sample would be identified as an amplification failure by reference to the FP of the control probe rather than as a negative sample that could lead to a misdiagnosis. FP is determined by the measurement of fluorescence intensities. The higher the probe concentration, the more accurate the measurement; lower probe concentrations favor reproducible amplicon detection. Optimum probe concentrations therefore reflect this relation between accuracy and reproducibility. We found that for ARMS reactions generating two amplicons, a probe concentration of 30 nmol/L is ideal. With primer sets at a concentration of 1 μmol/L, only 3% of primers require conversion to amplicon to produce target at 30 nmol/L. We found that in such reactions the amount of primer incorporated into amplicon is usually between 5% and 10% (data not shown). A 30 nmol/L probe concentration will therefore give a reproducible change in FP in almost all test samples. In Fig. 4 (top) the difference in FPs for the two positive samples detected by the apo B probe is probably due to a low yield of apo B amplicon. This would result in a significant proportion of probe remaining unhybridized, thereby lowering the net FP value. Thus, samples that give product at <30 nmol/L will usually be detectable, but samples in which the amplicon concentration is <10 nmol/L will be difficult to distinguish from negatives and would be identified as amplification failures to be retested. Although FP can be used quantitatively (16), it is also suited to delivering simple “yes or no” answers as is shown in Fig. 2 . Twelve of the samples contain the ARMS amplicon, 12 do not. FP clearly differentiates between the two groups with little variation about their respective mean FP. It is important to determine if there is an overlap of the data sets. We have shown that the two data sets are distinct and therefore misdiagnoses should be avoided. FP detection of PCR amplification is reliant on the hybridization of an oligonucleotide probe to one strand of a denatured PCR product. There are two competing equilibria in such a reaction. These are the binding of the probe to one of the amplicon strands and the reannealing of the two amplicon strands. The thermodynamically more stable state for the system is for the PCR product to reanneal, leaving the probe unbound. However, the rate of DNA duplex formation varies with the square root of the length of the smaller strand forming the duplex (21). The binding of the probe to one of the amplicon strands is therefore kinetically favored over the reannealing of the two amplicon strands as is shown in Fig. 5 . This Fig. also shows that the specificity of the reaction is very high. To detect a mutation by ARMS with FP, there must be three specific oligonucleotide annealing events: the binding of the two PCR primers to the genome to generate the amplicon, followed by the binding of the fluorescently labeled probe to an internal region of that amplicon. Thus an amplified region of the genome will only give rise to a positive FP result if it contains a sequence complementary to that of the probe. The chance of such a region being amplified by random priming off the genome is essentially zero. Nevertheless, confirming the specificity of the first two hybridization events in the development of an ARMS/FP assay might be considered prudent. This is conveniently done by monitoring the ARMS part of the assay with gel electrophoresis as we described. Our results show that fluorescein, DTAF, and ROX are useful for the FP detection of PCR products, and we demonstrated the multiplexed detection for two amplicons using the two latter fluorophores. Other chromophores can be used to label probes, including 6-carboxy-2′, 7′dimethoxy-4′, 5′dichlorofluorescein (JOE) and Cy-5 (data not shown). A wide range of the visible spectrum that forms the basis of a powerful multiplex detection capability is covered by these dyes. In conclusion, we have demonstrated simple, rapid, and reproducible genotyping of DNA samples using FP detection of ARMS amplicons. The detection protocol requires a postamplification heat, cool, and equilibration cycle. Reading is rapid with the microtiter plate reader FPM-2, although this instrument is not suitable for carrying out multiplex detection. The technique can be considered homogeneous in that it requires no separation or wash step, but, in the multiplex format, it must be carried out in a nonhomogeneous manner because the ARMS reactions must be transferred from one vessel to another for the FP measurement. Therefore, for homogeneous multiplex ARMS analyses with FP, a microtiter FP plate reader is needed with two novel features: (a) multiple label detection across the entire visible spectrum, and (b) the ability to carry out assays directly on samples contained in PCR amplification tubes or microplates. This would lend the system to automation and completely obviate the need for complex means of avoiding carryover contamination (3)(4)(5)(6)(7)(8)(9)(10)(11). Zeneca Diagnostics, Gadbrook Park, Northwich, Cheshire, CW9 7RA, UK. 1 Nonstandard abbreviations: FP, fluorescence polarization; ARMS, amplification refractory mutation system; CFTR, cystic fibrosis transmembrane conductance regulator; DTAF, 5-([4,6-dichlorotriazin-2-yl]amino)fluorescein; ROX, 6-carboxyrhodamine; TBE, Tris–borate–EDTA; and apo, apolipoprotein. Table 1. Probe sequences. . Target . 5′ Label . Sequence . Probe 1 ΔF508 Aminohexylphosphate GCTTAATTTTACCCTCTGAA Probe 2 ΔF508 DTAF-aminohexylphosphate As probe 1 Probe 3 ΔF508 None TTCAGAGGGTAAAATTAAGC Probe 4 Apo B ROX-aminohexylphosphate GGTCAGGGAAATCATGGAAGGAACTGGG Probe 5 ΔF508 FITC-aminoethylphosphate GCTTAATTTTACCCTCTGAA . Target . 5′ Label . Sequence . Probe 1 ΔF508 Aminohexylphosphate GCTTAATTTTACCCTCTGAA Probe 2 ΔF508 DTAF-aminohexylphosphate As probe 1 Probe 3 ΔF508 None TTCAGAGGGTAAAATTAAGC Probe 4 Apo B ROX-aminohexylphosphate GGTCAGGGAAATCATGGAAGGAACTGGG Probe 5 ΔF508 FITC-aminoethylphosphate GCTTAATTTTACCCTCTGAA FITC, fluorescein isothiocyanate. Table 1. Probe sequences. . Target . 5′ Label . Sequence . Probe 1 ΔF508 Aminohexylphosphate GCTTAATTTTACCCTCTGAA Probe 2 ΔF508 DTAF-aminohexylphosphate As probe 1 Probe 3 ΔF508 None TTCAGAGGGTAAAATTAAGC Probe 4 Apo B ROX-aminohexylphosphate GGTCAGGGAAATCATGGAAGGAACTGGG Probe 5 ΔF508 FITC-aminoethylphosphate GCTTAATTTTACCCTCTGAA . Target . 5′ Label . Sequence . Probe 1 ΔF508 Aminohexylphosphate GCTTAATTTTACCCTCTGAA Probe 2 ΔF508 DTAF-aminohexylphosphate As probe 1 Probe 3 ΔF508 None TTCAGAGGGTAAAATTAAGC Probe 4 Apo B ROX-aminohexylphosphate GGTCAGGGAAATCATGGAAGGAACTGGG Probe 5 ΔF508 FITC-aminoethylphosphate GCTTAATTTTACCCTCTGAA FITC, fluorescein isothiocyanate. Table 2. ARMS primer sequences. . Specificity . Sequence . Primer 1 ΔF508 forward CCAGACTTCACTTCTAATGATGATTATGGG Primer 2 ΔF508 mutant reverse GTATCTATATTCATCATAGGAAACACCATT Primer 3 ΔF508 normal reverse GTATCTATATTCATCATAGGAAACACCACA Primer 4 Apo B forward AGCACAGTACGAAAAACCACCTT Primer 5 Apo B reverse ACTTTTACAGGGATGGAGAACG . Specificity . Sequence . Primer 1 ΔF508 forward CCAGACTTCACTTCTAATGATGATTATGGG Primer 2 ΔF508 mutant reverse GTATCTATATTCATCATAGGAAACACCATT Primer 3 ΔF508 normal reverse GTATCTATATTCATCATAGGAAACACCACA Primer 4 Apo B forward AGCACAGTACGAAAAACCACCTT Primer 5 Apo B reverse ACTTTTACAGGGATGGAGAACG Table 2. ARMS primer sequences. . Specificity . Sequence . Primer 1 ΔF508 forward CCAGACTTCACTTCTAATGATGATTATGGG Primer 2 ΔF508 mutant reverse GTATCTATATTCATCATAGGAAACACCATT Primer 3 ΔF508 normal reverse GTATCTATATTCATCATAGGAAACACCACA Primer 4 Apo B forward AGCACAGTACGAAAAACCACCTT Primer 5 Apo B reverse ACTTTTACAGGGATGGAGAACG . Specificity . Sequence . Primer 1 ΔF508 forward CCAGACTTCACTTCTAATGATGATTATGGG Primer 2 ΔF508 mutant reverse GTATCTATATTCATCATAGGAAACACCATT Primer 3 ΔF508 normal reverse GTATCTATATTCATCATAGGAAACACCACA Primer 4 Apo B forward AGCACAGTACGAAAAACCACCTT Primer 5 Apo B reverse ACTTTTACAGGGATGGAGAACG Figure 1. Open in new tabDownload slide Ethidium bromide-stained agarose gel of ARMS reactions. Top: lanes 1–6, +/+ DNA; lanes 7–12, +/ΔF508 DNA. Bottom:lanes 13–18, ΔF508/ΔF508 DNA; lanes 19–24, no DNA controls. In both panels odd-numbered lanes represent use of the wild-type-specific ARMS primer; even-numbered lanes represent use of the ΔF508-specific ARMS primer. Sample numbers are equivalent to those analyzed by FP shown in Fig. 2 . M represents ØX174/HaeIII digest molecular size markers. Figure 1. Open in new tabDownload slide Ethidium bromide-stained agarose gel of ARMS reactions. Top: lanes 1–6, +/+ DNA; lanes 7–12, +/ΔF508 DNA. Bottom:lanes 13–18, ΔF508/ΔF508 DNA; lanes 19–24, no DNA controls. In both panels odd-numbered lanes represent use of the wild-type-specific ARMS primer; even-numbered lanes represent use of the ΔF508-specific ARMS primer. Sample numbers are equivalent to those analyzed by FP shown in Fig. 2 . M represents ØX174/HaeIII digest molecular size markers. Figure 2. Open in new tabDownload slide FP readings from ARMS reactions. Samples 1–6, +/+ DNA; samples 7–12, +/ΔF508 DNA; samples 13–18, ΔF508/ΔF508 DNA; samples 19–24, no DNA controls. Odd-numbered samples represent use of the wild-type-specific ARMS primer; even-numbered samples represent use of the ΔF508-specific ARMS primer. Sample numbers are equivalent to those shown in Fig. 1 . Figure 2. Open in new tabDownload slide FP readings from ARMS reactions. Samples 1–6, +/+ DNA; samples 7–12, +/ΔF508 DNA; samples 13–18, ΔF508/ΔF508 DNA; samples 19–24, no DNA controls. Odd-numbered samples represent use of the wild-type-specific ARMS primer; even-numbered samples represent use of the ΔF508-specific ARMS primer. Sample numbers are equivalent to those shown in Fig. 1 . Figure 3. Open in new tabDownload slide FP readings from the ARMS reactions from Fig. 2 showing the clustering of the FP values around the mean values for negative and positive reactions. Figure 3. Open in new tabDownload slide FP readings from the ARMS reactions from Fig. 2 showing the clustering of the FP values around the mean values for negative and positive reactions. Figure 4. Open in new tabDownload slide Mean values from duplicated ARMS reactions. Top: apo B amplicon detection with ROX probe plus DTAF probe for ΔF508. Bottom: ΔF508 amplicon detection with DTAF probe plus ROX probe for apo B. First column, ΔF508 and apo B primers plus ΔF508/ΔF508 DNA; second column, ΔF508 and apo B primers minus DNA; third column, ΔF508 primers alone plus ΔF508/ΔF508 DNA; fourth column, apo B primers alone plus ΔF508/ΔF508 DNA. Figure 4. Open in new tabDownload slide Mean values from duplicated ARMS reactions. Top: apo B amplicon detection with ROX probe plus DTAF probe for ΔF508. Bottom: ΔF508 amplicon detection with DTAF probe plus ROX probe for apo B. First column, ΔF508 and apo B primers plus ΔF508/ΔF508 DNA; second column, ΔF508 and apo B primers minus DNA; third column, ΔF508 primers alone plus ΔF508/ΔF508 DNA; fourth column, apo B primers alone plus ΔF508/ΔF508 DNA. Figure 5. Open in new tabDownload slide Time course showing change in FP of DTAF-labeled ΔF508 probe hybridized to ΔF508/ΔF508 amplicon (○) and to apo B amplicon (•) Figure 5. Open in new tabDownload slide Time course showing change in FP of DTAF-labeled ΔF508 probe hybridized to ΔF508/ΔF508 amplicon (○) and to apo B amplicon (•) We thank A. J. Garman for comments during the preparation of the manuscript. We thank CSP Ltd., Construction House, Grenfell Ave., Hornchurch, Essex RM12 4EH, UK, for the loan of a Fluorolite FPM-2 microtiterplate FP plate reader. The PCR process is covered by patents held by Hoffmann-La Roche. ARMS is the subject of European patent no. 0 332 435 (Zeneca Ltd.) and corresponding patents worldwide. Detection of nucleic acid sequences with FP is the subject of European patent application no. 0 382 433 (Zeneca Ltd.). 1 Mullis KB Ferré F Gibbs RA eds. The polymerase chain reaction 1994 : 323 -406 Birkhäuser Boston. . 2 Newton CR. Mutational analysis: known mutations. McPherson MJ Hames BD Taylor GR eds. PCR 2: a practical approach 1995 : 219 -253 IRL Press Oxford. . 3 Kwok S, Higuchi R. Avoiding false positives with PCR. Nature 1989 ; 339 : 237 -238. Crossref Search ADS PubMed 4 Jinno Y, Yoshiura K, Niikawa N. Use of psoralen as extinguisher of contaminated DNA in PCR. Nucleic Acids Res 1990 ; 18 : 6739 . Crossref Search ADS PubMed 5 Meier A, Persing DH, Finken M, Böttger EC. Elimination of contaminating DNA within polymerase chain reaction reagents: implications for a general approach to detection of uncultured pathogens. J Clin Microbiol 1993 ; 31 : 646 -652. Crossref Search ADS PubMed 6 Victor T, Jordaan A, du Toit R, Van Helden PD. Laboratory experience and guidlines for avoiding false positive polymerase chain reaction results. Eur J Clin Chem Clin Biochem 1993 ; 31 : 531 -535. PubMed 7 Cimino GD, Metchette KC, Tessman JW, Hearst JE, Isaacs ST. Post-PCR sterilization: a method to control carryover contamination for the polymerase chain reaction. Nucleic Acids Res 1991 ; 19 : 99 -107. Crossref Search ADS PubMed 8 Isaacs ST, Tessman JW, Metchette KC, Hearst JE, Cimino GD. Post-PCR sterilization: development and application to an HIV-1 diagnostic assay. Nucleic Acids Res 1991 ; 19 : 109 -116. Crossref Search ADS PubMed 9 Espy MJ, Smith TF, Persing DH. Dependence of polymerase chain reaction product inactivation protocols on amplicon length and sequence composition. J Clin Microbiol 1993 ; 31 : 2361 -2365. Crossref Search ADS PubMed 10 Longo MC, Berninger MS, Hartley JL. Use of uracil DNA glycosylase to control carry-over contamination in polymerase chain reactions. Gene 1990 ; 93 : 125 -128. Crossref Search ADS PubMed 11 Rys PN, Persing DH. Preventing false positives: quantitative evaluation of three protocols for inactivation of polymerase chain reaction amplification products. J Clin Microbiol 1993 ; 31 : 2356 -2360. Crossref Search ADS PubMed 12 Jolley ME. Fluorescence polarization immunoassay for the determination of therapeutic drug levels in plasma. J Anal Toxicol 1981 ; 5 : 236 -240. Crossref Search ADS PubMed 13 Murakami A, Nakaura M, Nakatsuji Y, Nagahara S, Tran-Cong Q, Makino K. Fluorescent-labelled oligonucleotide probes: detection of hybrid formation in solution by fluorescence polarisation spectroscopy. Nucleic Acids Res 1991 ; 19 : 4097 -4102. Crossref Search ADS PubMed 14 Herning T, Tamiya E, Karube I, Kobayashi S. Specific liquid hybridisation kinetics measured by fluorescence polarisation. Anal Chim Acta 1991 ; 244 : 207 -213. Crossref Search ADS 15 Devlin R, Studholme RM, Dandliker WB, Fahy E, Blumeyer K, Ghosh SS. Homogeneous detection of nucleic acids by transient-state polarized fluorescence. Clin Chem 1993 ; 39 : 1939 -1943. Crossref Search ADS PubMed 16 Walker GT, Nadeau JG, Linn CP, Devlin RF, Dandliker WB. Strand displacement amplification (SDA) and transient-state fluorescence polarization detection of Mycobacterium tuberculosis DNA. Clin Chem 1996 ; 42 : 9 -13. Crossref Search ADS PubMed 17 Newton CR, Graham A, Heptinstall LE, Powell SJ, Summers C, Kalsheker N, et al. Analysis of any point mutation in DNA. The amplification refractory mutation system (ARMS). Nucleic Acids Res 1989 ; 17 : 2503 -2515. Crossref Search ADS PubMed 18 Ferrie RM, Schwarz MJ, Robertson NH, Vaudin S, Super M, Malone G, et al. Development, multiplexing and application of ARMS tests for common mutations in the CFTR gene. Am J Hum Genet 1992 ; 51 : 251 -262. PubMed 19 Holland PM, Abramson RD, Watson R, Gelfand DH. Detection of specific polymerase chain reaction product by utilizing the 5′ to 3′ exonuclease activity of Thermus aquaticus DNA polymerase. Proc Natl Acad Sci U S A 1991 ; 88 : 7276 -7280. Crossref Search ADS PubMed 20 Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, et al. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science 1989 ; 245 : 1066 -1073. Crossref Search ADS PubMed 21 Hames BD Higgins SJ eds. Nucleic acid hybridisation: a practical approach 1985 : 47 -71 IRL Press Oxford. . © 1997 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)