ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 29, No. 1 Copyright © 1999, Institute for Clinical Science, Inc.

The Molecular Pathology Laboratory of the 21st Century*

FREDERICK L. KIECHLE, M.D., Ph.D., XINBO ZHANG, M.D., Ph.D., and TADEUSZ MALINSKI, Ph.D.f

Department o f Clinical Pathology, William Beaumont Hospital, Royal Oak, MI 48073 and fDepartment of Chemistry, Oakland University, Rochester, M I 48309

ABSTRACT

Human cells contain deoxyribonucleic acid in mitochondria and nuclei. Human diseases may be caused by mutations in mitochondrial DNA, nuclear DNA or both. The volume of work performed in the diagnostic molecular pathology laboratory will continue to grow as more disease-related mutations are discovered. Many factors will influence the diagnostic molecular pathology laboratory in the 21st century, such as future clinical laboratory organization, amplification methods, specimen integrity, ethical guidelines and opportunities to expand service. In the evaluation of a patient suspected of a mitochondrial DNA mutation, care must be exercised in the selection of a primer for amplification and of the specimen to be examined for the mutation. The uneven distribution of normal and abnormal mitochondrial DNA within the various tissues (heteroplasmy) may result in a normal mitochondrial DNA sequence if the wrong tissue is examined. The presence of mitochondrial-like sequences (pseu­ dogenes) within nuclear DNA may result in amplification of nuclear genes if generic primers are used to duplicate a mitochondrial DNA gene. Diabetes mellitus is a heterogeneous disease with mutations occurring in a variety of proteins leading to either prereceptor, receptor or postreceptor defects. In this example, the diagnostic molecular pathology laboratory may be asked to define the specific genotype a specific patient with this common phenotype may possess.

Introduction cells.1,2 Human diseases may be caused by mutations in mitochondrial and/or nuclear Deoxyribonucleic acid (DNA) is located in DNA.1“3 There is growing interest in provid­ the nuclei and mitochondria of mammalian ing diagnostic assays based on identifying spe­ cific DNA sequences.1^4 The Human Genome * Address correspondence and reprint requests to: Project has focused attention on the associa­ Frederick L. Kiechle, M.D., Ph.D., Department of Clini­ tion of mutations in nuclear DNA (nDNA) cal Pathology, William Beaumont Hospital, 3601 West 13 with human diseases.5 Mitochondrial DNA Mile Road, Royal Oak, MI 48073-6769, Telephone (248) 551-8020, FAX (248) 551-3694, e-mail fkiechle@ (mtDNA) has been completely sequenced beaumont.edu. and each strand possesses 16,569 base pairs.1,3 59 0091-7370/99/0100-0059 $00.00 © Institute for Clinical Science, Inc. 60 KIECHLE, ZHANG, AND MALINSKI

The field of diagnostic molecular pathology Probe Laboratoiy in the Department of Clini­ is expanding rapidly.6 As new genes are cal Pathology at William Beaumont Hospital sequenced, the search for a disease-related was opened in November, 1991. The total mutation or rearrangement of the gene rapidly number of billable procedures has increased follows.5,7 Applications of diagnostic molecu­ each year (figure 1). A royalty agreement for lar pathology include genetics,7 inborn errors clinical use of polymerase chain reaction of metabolism,1,7-9 forensic pathology,10 hema- (PCR) was signed with Roche in late 1992.16 topathology,11 coagulation,12 neoplasia,4 In 1993, PCR assays were introduced and now virology,13 microbiology,13,14 histocompatibil­ exceed the number of Southern blot proce­ ity15 and mtDNA disorders.1,3 dures performed in the laboratory (figure 1). This article will review the workload of a In 1996, ligand chain reaction (LCR) assays for diagnostic molecular pathology laboratory Chlamydia trachomatis and Neisseria gonnor- from 1991 to 1997; consider factors which will hoea were introduced. These changes reflect shape the diagnostic molecular pathology in the benefits of nucleic acid amplification tests, the 21st century, such as future clinical labo­ including faster turnaround time, decreased ratory organization, amplification methods, labor intensity and increased efficiency. Today, specimen integrity, ethical guidelines and the laboratory offers four Southern blot assays expansion of service opportunities; review spe­ (B/T cell, her, bcl-2 gene rearrangements, cific issues related to defining mtDNA disor­ fragile X syndrome detection), five PCR assays ders; use diabetes mellitus as a model to illus­ (paternity by histocompatibility loci, Factor V, trate how diagnostic molecular pathology may hereditary hemochromatosis, 5,10 methylene- aid in classification and treatment. tetrahydrofolate reductase gene, angiotensin converting enzyme I gene), and two reverse D ia g n o s t ic M o l e c u l a r P a t h o l o g y T est s transcriptase PCR procedures (qualitative and a t W il l ia m B e a u m o n t H o sp it a l quantitative hepatitis C virus detection). There has been a rapid increase in test volume from Under the direction of D. Crisan, M.D., 0.04 percent (84 assays) of total laboratory test Ph.D., and D.H. Farkas, Ph.D., the Molecular volume in January, 1994, to 0.35 percent

10000

9000 1

8000 1

7000 1

6000 1

5000 1 4000 3000 2000 1000

0 - ■ — ' rrZ^: - 1991 1992 1993 1994 1995 1996 1997 F i g u r e 1. Annual volume of tests based on Southern blot (— ♦ —), polymerase chain reaction (—■—), ligase chain reaction (—A —) and total tests performed from 1991 to 1994 at William Beaumont Hospital. MOLECULAR PATHOLOGY LABORATORY OF THE 21st CENTURY 61

(1,093 assays) in July, 1997. The laboratory no rithm in the hematology analytical unit defin­ longer operates with a deficit. To achieve the ing when a slide should be prepared by the prediction of 5 percent of total laboratory test slide maker/stainer.24 Laboratories which are volume in the early 21st century,17 a total of labor intensive, like transfusion medicine, his­ 15,500 molecular pathology procedures per tocompatibility, microbiology, virology, diag­ month would need to be performed. Factors nostic molecular pathology and flow cytom­ which will influence the growth in diagnostic etry, will remain in separate locations or may molecular pathology will be described. be sent out to reference laboratories. If the core laboratory is located one or more miles

C l in ic a l L a b o r a to r y o f t h e F u t u r e from the hospital(s) it serves, the hospital(s) will need to have an immediate response labo­ Benge, et al,18 reviewed the costs of deliv­ ratory to handle time-critical specimens. This ering clinical chemistry services at Vanderbilt rapid response lab will contain equipment University (Nashville, TN) over ten years. Five which duplicates some of the services offered primary conclusions were derived: 1) increases in the core lab, like electrolytes, glucose, etc. in direct costs were entirely attributed to salary The automated line in the core laboratory may increases; 2) there was no change in inflation- be fast enough to handle STAT specimens, but corrected direct costs per test; 3) the profit it may be located too far away to provide margin was directly influenced by the hospi­ timely specimen delivery. One method to tal’s accounting method for calculating indirect reduce the test volume in the immediate costs; 4) the workload performed by each response lab is to offer those assays that employee increased with time; and 5) labor improve patient outcome as a part of the point- costs were approximately 50 percent of the of-care testing program. laboratory’s operating budget. As the number of capitated commercial contracts for labora­ A s s e s s m e n t o f T e c h n o l o g y tory services increases, cost reduction efforts in purchased supplies, consumables and capi­ New technology must be assessed. For tal equipment will reach a maximum. Atten­ example, bioelectronic chips provide a method tion will then turn to staff reduction. This for lysing bacteria on one chip and hybridiza­ reduction in labor costs may be achieved by tion for identification on a second chip.25 This consolidation of laboratories within a hospital technology comes currently with a high or two or more hospitals, outsourcing specific price.26 However, in the future, photolitho­ laboratory services, or introduction of labora­ graphed 1 cm2 chips may provide the possibil­ tory automation in isolated work cells or in a ity for DNA mutation analysis and infectious core laboratory model.19-23 microbe identification at the bedside. The Therefore, the future hospital laboratory time required to identify a viral or microbio­ will consist of a core laboratory featuring total logical organism will be significantly reduced laboratory automation performing non-time by molecular detection. Chip technology has dependent, high-volume tests in hematology, been used for continuous-flow PCR with a coagulation, clinical chemistry and immuno- total reaction time for 20 cycles of 90 sec­ chemistry. The automated line will also con­ onds27 and a variety of other applications.28,29 tain a single module or separate units capable Amplification methods for increasing the of preanalytical processing like centrifugation, number of gene or transcript copies are often aliquotting, decapping and recapping.22 Adja­ used to increase the sensitivity of molecular cent to this total laboratory automation unit biologic assays. Sensitivity may be increased will be manual or semi-automated laboratory further by using specific gene probes in functions which support the core lab. For hybridization assays of the amplified nucleic example, manual differential counts will be acid targets.30 Careful evaluation of each assay performed from slides generated by the algo­ is required. For example, the thermal stable 62 KIECHLE, ZHANG, AND MALINSKI

DNA polymerase may be inhibited during by RNases. There is limited data available on PCR by heparin,31 heme from lysed red the stability of nucleic acids in medical speci­ cells,32 urine,33 vitreous or aqueous fluid34 and mens. The Reference Guide for Diagnostic RNA.35 Polyamines36 interfere with or form- Molecular Pathology/Flow Cytometry: Fascicle amide37 promotes the specificity of this PCR- V II43 and other references describe issues based amplification. PCR is influenced by related to nucleic acid stability in a variety of other factors, including annealing tempera­ medical specimen types.44 46 For example, for ture, magnesium concentration and primer Southern blot-based assays, table I outlines selection. A mutation in a target molecule’s DNA stability when tissue is stored at a variety primer binding site may lead to no generation of conventional temperatures. of amplicons (DNA) during PCR.38 Careful Blood collected for inborn error of metabo­ primer selection improves PCR sensitivity lism screening on Guthrie cards contains DNA and specificity.39,40 which is stable for three years if stored from Alternate methods for target or signal ampli­ 4-25°C.47 The whole blood spots contain PCR fication have been reported including self­ inhibitors47 which may be removed by water sustained sequence replication, strand dis­ wash or commercial DNA purification kit.48 placement activation, repair chain reaction, The purified DNA may be used for the detec­ ligase chain reaction, Q-beta replicase system, tion of genetic diseases by PCR including cys­ ligation activated transcription and nucleic tic fibrosis.47,48 acid sequence-based amplification.41,42 Each RNA-based tests like RT-PCR-based hepa­ of these alternatives to PCR are being evalu­ titis C virus or HIV I detection may be col­ ated for potential clinical applications. The lected in a physician’s office or clinic some availability of standardized clinical applications distance from the laboratory. Since RNA is of these amplification methods will provide a quite labile, the RNA target must be processed wider selection of kits for diagnostic molecular rapidly and the RNA removed from contami­ pathology detection challenges. nating RNases released from neutrophils in the peripheral blood sample 43,45 The need for

S p e c im e n I n t e g r it y rapid specimen processing will require the col­ lection site to process the specimen. Aso, et DNA is a very stable macromolecular, while al,49 report that mononuclear cells collected by RNA species are likely to be rapidly degraded conventional density gradient methods may be

TABLE I

Southern Blot-based Assays - Tissue

NCCLS46 PPSH - CAP43

Room Temperature (25°C) NR, freeze or fix immediately NR, 5 min max; snap freeze with liquid N2 optimal Refrigerator (2-8°C) Transport on wet ice; cold NR, 1 hour max (30 phosphate-buffered saline (2-6°C) days placenta44) 24 hours - slices (< 0.5 cm) Freezer (-20°C) At least 2 weeks Freezer (-70°C) At least 2 years Fixed (25°C) Minced in ethanol > 1 year

NR = Not recommended. MOLECULAR PATHOLOGY LABORATORY OF THE 21“ CENTURY 63 filtered through glass fiber membranes. The patent.56 A similar situation occurs for the RNA on the dried membranes was stable at medical use of PCR for diagnostic use held by room temperature for one week. This method Roche (Nutley, NJ).16 The use of PCR to gen­ provides a specimen that is easy to transport to erate a billable test result requires royalty pay­ the Diagnostic Molecular Pathology Labora­ ment to Roche. tory. Improved methods for the collection of stable nucleic acid preparations from medical E x pa n s io n o f S e r v ic e O pportunities specimens, especially for RNA species, are still needed. Their availability will aid in the acqui­ The service provided by the hospital-based sition of good quality specimens for RT-PCR Diagnostic Molecular Pathology Laboratory is or other RNA-based clinical assays. limited primarily to diagnostic screening or monitoring of disease processes.6 However, in

E t h ic a l G u id e l in e s the future, this laboratory will be asked to assist in evaluating and monitoring a variety The resolution of ethical and procedural of treatment strategies like bone marrow issues will facilitate the timely acceptance and transplantation, somatic gene therapy and anti­ introduction of new diagnostic molecular pro­ sense oligonucleotides.6 One example may be cedures. The discovery of a candidate gene for the selection of specific chemotherapeutic hereditary hemochromatosis and two disease- agents that induce the greatest amount of associated mutations has been reported.50 An apoptosis in a patient’s leukemic cells.57 Che­ expert panel reviewed the ethical and health motherapeutic agents exert their antitumor policy implications of genetic testing for this effects by inducing programmed cell death or disorder.51 They concluded that population- apoptosis. Kravtsov, et al,57 found that purified based screening not be implemented at the leukemic cells from patients with the same present time secondary to uncertainties diagnosis demonstrated significant variability related to disease prevalence, penetrance and in chemosensitivities among three chemo­ care of asymptomatic people with a hemochro­ therapeutic agents tested for the ability to matosis gene mutation. These guidelines will induce apoptosis. be reviewed in one year. Other policy state­ We have used a model cell line, BC3H-1 ments and guidelines have been published for myocytes, known to undergo apoptosis in the specific diseases and their detection meth­ presence of the antitumor fluorescence dye, ods52,53 or specific molecular biological meth­ Hoechst 33342.58-60 The short-lived radical ods and their current limitations.54 This area nitric oxide (NO) has recently emerged as a requires continuous attention and updating as novel inhibitor or inducer of apoptosis.61 NO new information becomes available. can induce apoptosis in different types of An additional impediment to progress in the cells.62,63 To determine if Hoechst 33342- development of the diagnostic molecular induced apoptosis may be enhanced or pathology laboratory rests in the current prac­ reduced by NO, we treated BC3H-1 myocytes tice of patenting sequences in DNA.55 If the with 7.5 |xg/rnL Hoechst 33342 and increasing medical use of a specific gene and its mutated concentrations of the NO donor, sodium nitro- forms are patented for evaluation of patients or prusside (SNP), from 1-10 mmol/L (figure 2). relatives for a specific disease, then a royalty Incubation of BC3H-1 myocytes for six hours would need to be paid to the patent holder. A with different concentrations of SNP plus 7.5 variety of companies have applied for a patent |xg/mL Hoechst 33342 significantly enhance for the hereditary hemochromatosis gene.56 BC3H-1 myocyte death when compared to the One company (Progenitor, Menlo Park, CA) SNP-treated group or 7.5 (xg/rnL Hoechst has assigned exclusive rights to SmithKline 33342-treated group, suggesting that NO and Beecham Clinical Laboratories (Collegeville, Hoechst 33342 interact synergistically in PA) to perform laboratory testing under their inducing cell death (figure 2). This synergy is 64 KIECHLE, ZHANG, AND MALINSKI

CM 0. 0. CM Q. a. a. CM «sf a. CM a. CO z z CO Z z z z CO z z " c CO (0 CO CS) CO in CO n o CO CO 2 2 CO s CO «= co o z JL 2 2 in 2 X s EE E E I E E E T— 1- + in in + U) in + o o + CM oi

F i g u r e 2. Effect of sodium nitroprusside (SNP) and Hoechst 33342 (H33342) in inducing cell death. BC3H-1 myocytes cultured in MEME/10 percent fetal bovine serum were treated for 6 hours with different concentrations of SNP from 1-10 mmoI/L in the presence of 7.5 fig /m L Hoechst 33342. Each bar indicates the mean + standard deviation from three separate experiments. *, p < 0 .0 2 ; **, p < 0 .0 0 1 . illustrated by the more marked nuclear mor­ occurred in the BC3H-1 myocytes treated with phologic changes (chromatin condensation, both SNP and Hoechst 33342 (figure 4D), half moon configurations) induced in BC3H-1 suggesting synergistic action. These data dem­ cells by 10 mmol/L SNP plus 7.5 |j,g/mL onstrate how techniques to evaluate apoptosis Hoechst 33342 (figure 3D) compared to SNP using cells, such as the patient’s leukemic cells, (figure 3C) or Hoechst 33342 (figure 3B) in tissue culture may be used to evaluate effec­ treatment alone. To confirm further that tive combinations of chemotherapeutic agents Hoechst 33342 and SNP treatment induced to treat specific malignant processes. cell death via apoptosis in the BC3H-1 myo­ cytes, genomic DNA was extracted from Evaluation of mtDNA Disorders BC3H-1 myocytes after treatment with 10 mmol/L SNP, 7.5 |j,g/mL Hoechst 33342 or Although mtDNA disorders have a low the combination of both and analyzed by 1.5% prevalence among specific populations,64 agarose gel electrophoresis (figure 4). No there are an increasing number of disease- DNA fragmentation ladder of multiples of 180 associated mutations reported and specific pit­ base pairs was observed in control or Hoechst falls which may lead to an underestimation of 33342-treated cells (figure 4A, B). However, mtDNA disease. Most human cells contain both 10 mmol/L SNP and SNP plus Hoechst- between two to ten copies of mtDNA per induced DNA fragmentation in the character­ mitochondrion.65 The genes encoded in istic ladder of multiples of 180 base pairs mtDNA include two ribosomal ribonucleic observed in apoptotic cells (figure 4C, D). The acids (rRNA), 22 transfer ribonucleic acids greatest degree of DNA fragmentation (tRNA) and 13 polypeptides utilized to MOLECULAR PATHOLOGY LABORATORY OF THE 21st CENTURY 65 assemble proteins required for oxidative phos­ of establishing an association between mtDNA phorylation.3 Other proteins required for point mutations and disease. The most variable mitochondrial assembly and mtDNA replica­ region of the mtDNA is the 1,122 base pair tion66 are encoded in nDNA, are synthesized noncoding control region which contains two in the cytosol and imported into the mitochon­ hypervariable regions.83 The development of dria by a specific process.67 Mitochondrial dis­ an integrated and comprehensive human orders may result from mutations in mtDNA, mtDNA database is funded by the Eu Biotech nDNA or both.3'65'68,69 The first mtDNA dis­ Program and is called MitBASE.84 Polymor­ order was described at the molecular level by phisms are stored in the sequence variant sec­ Wallace, et al.70 in 1988. They reported a spe­ tion of the database which can be found on the cific point mutation in the mtDNA gene for internet at http://www.ebi.ac.uk/htbin/ nicotinamide adenine dinucleotide (reduced Mitbase/mitbase.pl. form) dehydrogenase which is associated with

Leber’s hereditary optic neuropathy. Since A mplification o f m t DNA that time, numerous other mtDNA mutations have been described which are associated with If the frequency of a mtDNA mutation is specific diseases. These mutations include very low (0.001 to 0.1% of total mtDNA), the base substitutions in protein subunit genes, mtDNA gene with the mutated sequence may base substitutions in tRNA or rRNA, deletions need to be amplified to simplify detection.74-76 in mtDNA, duplications of mtDNA or deple­ For example, polymerase chain reaction tion of mtDNA3’65’68’69’71-79 (table II). These amplification is required to detect mtDNA mutations may lead to altered mitochondrial deletions which are reported to increase with function, including impaired respiratory func­ age. However, Southern blot analysis is tion80-81 and decreased mitochondrial mem­ adequate for detection of the higher frequency brane potential.81 of mtDNA deletions (20 to 50% of total There are several potential pitfalls to avoid mtDNA) observed in Pearson syndrome and during the molecular biologic evaluation of a in muscle but not other tissues in Kearns- patient with a mtDNA disorder: polymor­ Sayre syndrome.69-85 phisms of mtDNA, primer selection for ampli­ Usually, mtDNA regions are amplified in fication of mtDNA genes and the importance preparations of total cellular DNA containing of heteroplasmy (normal and abnormal both nDNA and mtDNA.69-74 During evolu­ mtDNA in the same mitochondria or tissue) tion, mtDNA fragments have been inserted on the selection of the specimen to be ana­ into nDNA. This process has been so extensive lyzed and on the molecular biologic method to that nDNA sequences homologous to the be used. majority of the human mtDNA genome have been reported.86 Mitochondrial sequences or

P olymorphism o f mtDNA pseudogenes may be present in nDNA as single or multiple copies.86“88 In three cases At least 130 normal silent polymorphic of MELAS (A3243G mutation), the varia­ variations have been discovered in human tions between the respective level of normal mtDNA using restriction enzyme analysis, and mutated mtDNA species due to the DNA sequence analysis, denaturing gradient co-amplification of nuclear-embedded mtDNA gel electrophoresis and polymerase chain reac­ sequences was never greater than 10 per­ tion combined with single-strand conforma­ cent.89 Therefore, nuclear-encoded mitochon­ tion analysis.69-82-83 These polymorphic vari­ drial pseudogenes did not make a major con­ ants may result in amino acid substitution in tribution in the determination of the level of protein which does not affect the protein’s heteroplasmy in these patients. Primers function. The presence of these sequence designed to amplify nDNA may generate polymorphisms does increase the complexity mtDNA or synthesize single-stranded mtDNA 6 6 KIECHLE, ZHANG, AND MALINSKI

F igure 3. Characterization of nuclear morphological changes in BC3H-1 myocytes induced by 10 mmol/L SNP, 7.5 |xg/mL Hoechst 33342 or the combination of 10 mmol/L SNP plus 7.5 p,g/mL Hoechst 33342. BC3H-1 myocytes initially cultured in MEME/10 percent fetal bovine serum were incubated with SNP, Hoechst 33342 or SNP plus Hoechst 33342 for 6 hours. Magnification = 200x. A) Untreated BC3H-1 myocytes. B) 7.5 |xg/mL Hoechst 33342. C) 10 mmol/L SNP. D) 7.5 jxg/inl, Hoechst 33342 plus 10 mmol/L SNP. by mismatch priming.90 Primers designed to ing of nDNA primers to mtDNA or binding of amplify mtDNA may generate the pseudogene mtDNA primers to nDNA. Lists of primer and adjacent nDNA sequences.91 These prob­ sequences not found in mtDNA for nDNA lems should be suspected if polymerase chain primer construction90 and frequent sequences reaction amplification produces more than one in mtDNA for mtDNA primer assembly91 band or different bands, nucleotide sequence have been reported. is unusual and/or unexpected deletions/ insertions or frameshifts are observed.88 H eteroplasmy Therefore, primer selection is critical when Normal and mutated mtDNA are inherited using crude total cellular DNA to avoid bind­ maternally. Diseases of mitochondrial dysfunc­ MOLECULAR PATHOLOGY LABORATORY OF THE 21st CENTURY 67

F i g u r e 3. Continued.

tion with Mendelian inheritance pattern sug­ for detecting the mutation (see amplification gest a mutation in a nDNA gene required for above) and which specimens should be mitochondrial function.65 All children receive selected for investigation.3’65,69,72,73'92 the maternal mtDNA but only daughters pass This situation is illustrated by a mtDNA dis­ it on to their progeny. order called mtDNA depletion, which is char­ If a tissue possesses only normal or abnor­ acterized by very low levels of normal mtDNA mal mtDNA, it is said to be homoplastic for in affected tissues. This disorder is usually one or the other type of mtDNA. There are detected within the first two years of life and very few examples where a homoplastic distri­ may occur with an autosomal recessive inheri­ bution of abnormal mtDNA has been reported tance pattern or sporadically in two clinical in mtDNA diseases. More frequently, there is forms.69,71-73,93 The first clinical presentation a mixture of normal and abnormal mtDNA per is a fatal congenital form or early-onset form mitochondrion and/or per tissue stud- with liver failure.72,73,93 In these patients, the ied.3-65’68'69 This condition is called hetero- liver has the greatest reduction in mtDNA as plasmy. The frequency of abnormal mtDNA in measured as a ratio to a nDNA gene. The liver a specific tissue dictates the method selected may contain 7 percent mtDNA compared to 6 8 KIECHLE, ZHANG, AND MALINSKI

reduce the possibility of making multiple cop­ ies of a nDNA encoded pseudogene with mtDNA sequence homology. Also, the degree of heteroplasmy among tissues must be con­ sidered prior to selecting specimens for inves­ tigation of mtDNA mutations. Therapy for mtDNA disorders is primarily supportive,65 however, gene therapy using transfected mito­ chondria98 may provide hope for the future.

Diabetes Mellitus: Mutations in mtDNA and/or nDNA F i g u r e 4. Agarose gel electrophoresis of genomic DNA from BC3H-1 myocytes treated with 7.5 p,g/mL Diabetes mellitus is a group of metabolic Hoechst 33342 or 10 mM SNP or 7.5 p,g/mL Hoechst plus diseases characterized by hyperglycemia 10 mM SNP for 6 hours. Genomic DNA from cells was extracted and electrophoresed. A) Control; B) 7.5 (ig/mL resulting from defects in insulin secretion, Hoechst 33342; C) 10 mM SNP; D) 7.5 |xg/mL Hoechst insulin action or both.99-101 Mutations have 33342 + 10 mM SNP. been reported attributing to diabetes mellitus at all three major sites of insulin action (table control tissue, the muscle 50 percent and kid­ IV). Errors may occur prior to insulin binding ney, brain and heart have normal mtDNA con­ to its membrane bound receptor (prereceptor tent.73 The second late-onset form presents defects), at the structure or function of the with weakness, hypotonia or developmental insulin receptor (receptor defects) or at the delay in the first two years of life.71 Muscle transmembrane signaling mechanism initiated biopsies of these patients yield 2 to 34 percent after insulin bind to its receptor (postreceptor of mtDNA content compared to controls.71 defects).99-101 Table IV demonstrates that the Other tissues possess a minor reduction or etiology of diabetes mellitus is very hetero­ normal number of mtDNA copies. Therefore, geneous. A brief description of mutations it is important to investigate enzyme deficien­ noted in table IV will follow and illustrates the cies and mtDNA depletion in the appropri­ power of diagnostic molecular pathology in ate tissue.73,93 A reduction in the nuclear defining the specific genotypic cause of a com­ encoded gene product, transcription factor A, mon phenotype. required for mtDNA replication (table III) has been reported as a potential explanation for G e n e t ic D e f e c t s A l t e r in g the reduction in DNA observed in these b -c e l l F u n c t io n patients.94 However, a causal explanation is still lacking and this finding may be a second­ Insulin secretion from the beta cell involves ary phenomenon.95 several steps. First, glucose travels into the A reversible reduction in mtDNA in skeletal beta cell using a specific glucose transporter muscle has been reported after long-term Zid­ (GLUT 2) in the plasma membrane. Glucose ovudine (Azidothymidine) therapy.96 Azido- is then phosphorylated to glucose-6-phosphate thymidine triphosphate inhibits mtDNA poly­ by glucokinase. This reaction is followed by a merase gamma,97 thus reducing the mtDNA complex sequence of biochemical reactions content in skeletal muscle.96 (table III) which results in increased intracellular ATP produced by mitochondria. ATP-dependent S um m ary potassium channels are closed and voltage- activated calcium channels open to increase Care should be exercised in selection of intracellular calcium which triggers the secre­ primers used for amplifying mtDNA genes to tion of insulin.102 MOLECULAR PATHOLOGY LABORATORY OF THE 21sl CENTURY 69

TABLE II

Examples of Mitochondrial DNA Disorders

Mutation Type Associated Disorders

Depletion >Fatal congenital form with liver failure, autosomal recessive or sporadic. >lnfantile-onset myopathy, autosomal recessive or sporadic. >Zidovudine (Azidothymidine) therapy. Deletion (with or without duplication) >Pearson syndrome. >Kearns-Sayre syndrome. >Noninsulin-dependent diabetes meliitus with deafness. >Wolfram syndrome. >ttypokalemic periodic paralysis, inherited idiopathic dilated cardiomyopathy. >Aging Base substitute in protein subunit genes Mitochondrial ATPase 6: >Mild retinitis pigmentosa and migranes. >Neurogenic muscle weakness, ataxia and retinitis pigmentosa. >Leigh disease. Variable mitochondrial proteins: >Leber’s hereditary optic neuropathy. Base substitution in mt tRNA tRNAL«u >MELAS/PEO/diabetes with hearing loss. >MELAS/multisystem disease. >MERRF/PEO/diabetes. >MELAS. >Cardiomyopathy. >Cardiomyopathy/myopathy. tRNA11®: >Encephalocardiomyopathy. >Cardiomyopathy. >PEO. tRNA^": >Myopathy. tRNALys; >MERRF. >MERRF/MELAS. tRNAG|y; >Chronic intestinal pseudo-obstruction with myopathy and ophthalmopleigia. tRNAPro: >Prominal myopathy. Base substitution in mt >Maternally inherited aminoglycoside-induced deafness. rRNA >Familial deafness.

MT = Mitochondrial. tRNA » Transer ribonucleic acid. rRNA = Ribosomal ribonucleic acid. ATPase = Adenosine triphosphatase. MELAS = Mitochondrial encephalomyopathy, lactic acidosis, and strokelike episodes. MERRF = Myoclonic epilepsy and ragged red fiber disease. PEO = Progressive external ophthalmoplegia. 70 KIECHLE, ZHANG, AND MALINSKI

TABLE III Mutations in the glucokinase gene may rep­ resent the most common cause of type 2 dia­ Mitochondrial DNA Depletion - Mechanism betes identified to date. Mutations have been identified in a variety of racial and ethnic mtDNA requires: groups. In France, 57 percent (20/35) families 1. DNA polymerase gamma (inhibited by AZT) studied with maturity-onset diabetes of the 2. Short RNA primer from transcription of one young type 2 have a mutation in the glucoki­ mtDNA strand which requires: nase gene. By extrapolation, a mutation in this >RNA polymerase gene may be present in 6 percent of patients >mt transcription factor A (decreased in with type 2 diabetes in France.105 autosomal recessive forms) GLUT 2 mt = mitochondrial. AZT = Azidothymidine. The glucose transporter (GLUT 2) has been RNA = ribonucleic acid. implicated, along with glucokinase, in glucose regulation of insulin secretion in the pancre­ Autosomal recessive hyperinsulinemic hypo­ atic (3-cell. Mutations in the GLUT 2 gene glycemia of infancy is associated with a muta­ leading to a decrease in glucose transport tion in (3-cell ATP-sensitive K+ channel which activity contribute to the development of ges­ is inactive.103 Therefore, the 3-cell mem­ tational diabetes mellitus in a patient with a branes are always depolarized, leaving the volt­ valine 197 to isoleucine mutation.106 Three age-dependent Ca2+ channels open, which different homozygous GLUT 2 mutations results in a continuous release of insulin.103 have been reported in patients with the Fan- coni-Bickel syndrome.107 In each case, the G l u c o k in a s e (MODY2) mutated GLUT 2 molecule is predicted to lack Glucokinase catalyzes the phosphorylation function as a glucose transporter.107,108 The of glucose and represents the rate limiting Fanconi-Bickel syndrome is characterized by step in glycolysis. The glucokinase gene is hepatorenal glycogen accumulation, Fanconi expressed only in insulin-secreting pancreatic nephropathy and impaired utilization of glu­ fi-cells and in hepatocytes. The enzyme regu­ cose and galatose. lates insulin secretion in fâ-cells by sensing the concentration of glucose and regulates hepatic MODY1, MODY3 AND MODY4 glucose disposal by a similar mechanism.104 The glucokinase gene consists of 12 exons. The activation of transcription of a variety Mutations have been detected using genomic of genes known to be important in glucose DNA from peripheral white blood cells and metabolism in liver, pancreatic islets and PCR primers chosen to amplify each of the 12 other tissues are dysfunctional in patients exons of the glucokinase gene and adjacent with MODY1, MODY3 and MODY4. flanking sequences, like exon/intron junctions MODY1 is associated with mutations in or single-strand conformational polymorphism hepatic nuclear factor-4a109,110 located on analysis with sequencing of bands with abnor­ chromosome 20q. It is known to activate tran­ mal mobility. Forty-two glucokinase mutations scription of the hepatic nuclear factor-la have been detected.105 The majority are mis- (mutated in MODY3).111-113 Only 30 percent sence mutations which lead to decreased glu­ of patients with MODY1 became insulin- cokinase activity. Diabetic patients with these requiring, which may lead to the development mutations have one normal and one abnormal of associated microvascular complications.109 gene. The severity of their disease may be MODY3 is associated with mutations in at related to the degree of dysfunction exhibited least three functional domains of the hepatic by the mutant glucokinase. nuclear factor-la gene located on chromo­ MOLECULAR PATHOLOGY LABORATORY OF THE 21s' CENTURY 71

TABLE IV

Some Mutations Associated with Diabetes Mellitus

Prereceptor Defects: Genetic defects altering Glucokinase (MODY2) f3-cell function Glucose transporter 2 (Fanconi-Beckel syndrome) Hepatic nuclear factor-4a (MODY1) Hepatic nuclear factor-1 a (MODY3) Insulin promoter factor-1 (MODY4) Mitochondrial DNA (maternally inherited diabetes mellitus)

Receptor Defects Insulin receptor mutations

Postreceptor Defects Insulin receptor substrate-1 mutations

MODY = Maturity-onset diabetes of the young. some 12q.112 This transcription factor directly This process would be dysfunctional, resulting regulates the expression of glucose metabolism in lower levels of mitochondrial ATP produc­ genes, like the insulin gene, the glucokinase tion and, therefore, less insulin secretion.113 gene, the pyruvate kinase gene, and others.112 Maternally inherited type 1 and type 2 diabe­ These patients exhibit a severe form of diabe­ tes mellitus and deafness have been associated tes requiring insulin and is associated with with mutations with tRNALeu(UUR>; base pair microvascular complications.11 1-113 3243 (A G)116-117 or 3264 (T C)118 and in The transcription factor called insulin pro­ tRNALys; base pair 8296 (A —» G).119 Diabetics motor factor-1 (IPF-1) is required for pancre­ with a tRNALeu(UUR) mutation are approxi­ atic development and insulin gene transcrip­ mated to be 1.5 percent of total patients and tion in response to glucose.114 Homozygous tRNALys mutation at 1.0 percent.119 inheritance of IPF-1 mutation leads to pancre­ atic agenesis and heterozygous carriers of the Structurally A b n o r m a l I n s u l in mutant IPF-1 allele develop autosomal domi­ o r P r o in s u l in nant early onset type 2 diabetes mellitus (MODY4). The mutant IPF-1 also inhibits the Defects in the insulin gene have been transcription of insulin and other (5-cell spe­ reported to be inherited as an autosomal domi­ cific genes in these heterozygotes.114 nant phenotype with patients presenting with either hyperproinsulinemia or hyperinsu- M itochondrial DNA linemia.120 The patients present with charac­ teristic clinical and laboratory findings are out­ The inhibition of mitochondrial oxidative lined in table V. phosphorylation in pancreatic beta cells Insulin was the first protein to be purified, impairs insulin secretion.115 A mutation in a crystallized and sequenced. The insulin gene mtDNA gene for specific tRNA species may located on the short arm of chromosome 11, lead to abnormal truncated protein or kineti- was the first gene to be cloned, and human cally dysfunctional secondary to modification insulin was the first pharmacologic product of the enzyme’s tertiary structure. The oxida­ manufactured with the aid of recombinant- tive phosphorylation sequence requires some DNA technology. proteins produced in the mitochondria from The insulin gene is found in all nucleated the transcription and translation of mtDNA. cells but only expressed in beta cells of the 72 KIECHLE, ZHANG, AND MALINSKI

TABLE V for more than 80 percent of circulating “insu­ lin” immunoreactivity (normal <20 percent). Abnormal Insulin Syndrome The glucose tolerance of affected subjects deteriorates with age. Autosomal dominant phenotype. Hyperinsulinemia or hyperproinsulinemia. Mutations in the insulin gene were initially Hyperglycemia/diabetes in some cases. detected in genomic DNA extracted from leu­ Abnormal and normal insulin by HPLC. kocytes by Southern blot analysis using the Increased insulimC-peptide ratio. 32P-labeled insulin gene. Screening for struc­ Increased abnormal insulin halg—life. turally abnormal insulin or proinsulin using No increase in insulin-antagonists. restriction endonucleases demonstrated a fre­ No antibodies against insulin or insulin quency of mutant insulin (B24 or B25) of 4.69 receptor. per 1,000 patients with NIDDM.124 Normal response to exogenous insulin. Decreased response to endogenous insulin.

H u m a n I n s u l in R e c e p t o r M u t a t io n s

The human insulin receptor (HIR) is a islets of Langerhans in the pancreas. The insu­ transmembrane glycoprotein composed of two lin gene is composed of three exons and two a subunits of 135 kilodaltons (kD) each and introns. A single polypeptide precursor, pre­ two P subunits of 95 kD each held together by proinsulin, is translated from the mRNA tran­ inter- and intrasubunit disulfide bonds. There scribed from the insulin gene. Preproinsulin is are approximately 100 to 250,000 cell surface converted to proinsulin which contains 86 receptors per mammalian cell. Insulin- amino acids. Following the loss of four amino responsive tissues, like adipose, muscle and acids, equimolar amounts of C-peptide and liver cells, contain a higher number of recep­ insulin (20 amino acid A-chain; 31 amino acid tors. Two insulin receptor variants (HIR-A, B-chain connected by two disulfide bridges) HIR-B) have been described125 126 that differ are produced. Substrate recognition by the in the C-terminal end of the a subunit. HIR-A protease or convertase that changes proinsulin lacks 12 amino acids at position 716 in the to insulin appears to be complex.121 COOH-terminus of the extracellular a sub­ Defects in the insulin gene have been unit. The HIR-B variant is increased in the reported to be inherited as an autosomal domi­ skeletal muscle from patients with non-insulin nant phenotype, with patients presenting with dependent diabetes mellitus compared with either hyperproinsulinemia or hyperinsu­ control. Insulin binding and endocytotic rates linemia.122’123 Three specific gene mutations differ for the two HIR variants. It is not known result in reduced biological activity for the whether the two variants transmit different mutant insulin (Insulin Chicago, PheB25Leu; intracellular signals after insulin binding. Insulin Los Angeles, PheB24Ser; Insulin At least 50 HIR mutations have been iden­ Wakayama ValA3Leu). The patients present tified, including point mutations and dele­ with abnormal insulin syndrome characteris­ tions,127 leading to a variety of consequences tics. The B24, B25 and A3 amino acids repre­ (table VI). For example, one receptor variant sent contact sites between insulin and its contains a silent mutation at amino acid 1046 receptor. Hyperproinsulinemia may be the and a point mutation (TGG —> TCG) at posi­ consequence of a mutation in the connecting tion 1188 converting tryptophan to serine. This sequence between the C-peptide and A-pep- latter mutation occurs within the tyrosine tide (ArgC65His and ArgC65Leu) or in the kinase domain and may explain the impaired B-chain (HisBlOAsp), resulting in a proinsulin beta subunit activity observed in this patient species that is not completely converted to with severe insulin resistance, acanthosis nigri­ insulin. Proinsulin or its intermediates account cans and polycystic ovary syndrome. NIDDM MOLECULAR PATHOLOGY LABORATORY OF THE 21sl CENTURY 73

TABLE VI muscle cells.131 Other signal transduction pathways result in the activation of serine/ Consequences of Human Insulin threonine kinases. Receptor Devects Efforts to identify IRS-1 mutants in patients with severe insulin resistance have uncovered Reduce HIR tyrosine kinase activity. Decreased receptor autophosphorylation. two mutants that demonstrate dysfunctional Abnormal substrate levels. signal transduction, a postreceptor de­ Reduced insulin binding. fect.132133 One mutant IRS-1 failed to bind Increase HIR degradation. normally to PI 3'-kinase p85 and the second Impaired transport of HIR to the cell demonstrated a decreased mitogenic response membrane. secondary to reduced binding to the HIR.133 Processing errors in proreceptor to the a and Additional studies to uncover new IRS-1 muta­ P subunits. tions continue.134 Decreased HIR mRNA content and HIR number. S u m m a ry HIR = Human insulin receptor. Dissecting insulin action into potential defects at the prereceptor, receptor or post­ is associated with a missense mutation receptor level will demonstrate subsets of (Thr831 —> Ala) in the HIR of three of 51 patients with heterogeneous pathogenetic NIDDM Japanese patients screened.128 pathways and potential new therapeutic meth­ Patients with HIR mutations seldom present ods. Diagnostic molecular pathology labora­ with common form of NIDDM but rather with tory support will be required for the identifi­ syndromes of severe insulin resistance like cation of mutations in specific genes associated Type A insulin resistance, leprechunism or with diabetes mellitus. Robson-Mendenhall syndrome. Acknowledgment IRS-1 a n d T y pe 2 D ia b e t e s M e l l it u s This work was partially support by a grant from the William Beaumont Hospital Research Institute. Thanks Insulin binds to the extracellular a-subunit are extended to Pat Schmidt for typing this manuscript. of the HIR129 and activates an intrinsic tyro­ sine kinase within the intracellular P-subunit R eferences of HIR. This enzyme autophosphorylates spe­ cific tyrosine residues within the p-subunit. 1. Kiechie FL, Quattrociocchi-Longe TM. The role of Insulin receptor substrate-1 (IRS-1) and IRS-2 the molecular probe laboratory in the 2 1 st century. Lab Med 1992;23:758-763. are peptide substrates phosphoiylated by the 2. Bodmer W. W here will genome analysis lead us forty activated insulin receptor tyrosine kinase. Both years on? Ann NY Acad Sei 1995;758:414-426. IRS-1 and IRS-2 undergo phosphorylation at 3. Graeber MB, Müller U. Recent developments in the molecular genetics of mitochondrial disorders. J multiple tyrosine residues.130 The phosphory- Neurol Sei 1998;153:251-263. lated IRS-1/-2 then binds with proteins that 4. Plummer SJ, Casey G. Are we any closer to genetic contain a src-homology-2 (SH2) domain. IRS- testing for common malignancies? Nature Med 1996;2:156-158. 1/-2 act as adaptor or docking molecules which 5. Green ED, Waterston RH. The human genome link HIR to various cellular reactions regulated project. Prospects and implications for clinical medi­ by insulin. For example, IRS-1/-2 bind to the cine. JAMA 1991;266:1966-1975. 6 . Kiechie FL. Diagnostic molecular pathology in the p85 subunit of phosphatidylinositol 3'-kmase twenty-first century. Clin Lab Med 1996;16:213- and activate it, resulting in insulin-stimulated 222. GLUT4 translocation. IRS-2 is dephosphoiy- 7. McKusick VA. Current trends in mapping human genes. FASEB J 1991;5:12-20. lated much more rapidly and activates PI 3 '-ki­ 8 . Kiechie FL. Molecular biology of porphyrias. Lab nase more transiently than IRS-1 in skeletal Med 1993;24:648-655. 74 KIECHLE, ZHANG, AND MALINSKI

9. Kiechle FL, Holmes RD. Molecular biology of leu­ 29. Marshall A, Hodgson J. DNA chips: an array of pos­ kodystrophies. Lab Med 1994;25:658-663. sibilities. Nature Biotechnol 1998;16:27-31. 10. Weedn VW, Rogers GS, Henry BE. DNA testing in 30. Block W. A biochemical perspective of the polymer­ the forensic laboratory. Lab Med 1998;29:484-489. ase chain reaction. Biochemistry 1991;30:2735-2747. 11. Cline MJ. The molecular basis of leukemia. N Engl 31. Satsangi J, Jewell DP, Welsh K, Bunce M, Bell J. J Med 1994;330:328-336. Effect of heparin on polymerase chain reaction. Lan­ 12. Martinelli I, Sacchi E, Landi G, Taioli E, Duca F, cet 1994;343:1509-1510. Mannucci PM. High risk of cerebral-vein thrombosis 32. Mercier B, Gaucher C, Feugeas O, Mazurier C. in carriers of a prothrombin-gene mutation and in Direct PCR from whole blood, without DNA extrac­ users of oral contraceptives. N Engl J Med 1998;338: tion. Nucleic Acids Res 1990;18:5908. 1793-1797. 33. Khan G, Kangro HO, Coates PJ, Heath RB. Inhibi­ 13. Alanen A. Polymerase chain reaction in the detection tory effects of urine on the polymerase chain reaction of microbes in amniotic fluid. Ann Med 1998;30: for cytomegalovirus DNA. J Clin Pathol 1991;44: 288-295. 360-365. 14. Della-Latta P, Whittier S. Comprehensive evaluation 34. Wiedbrauk DL, Werner JC, Drevon AM. Inhibition of performance, laboratory application, and clinical of PCR by aqueous and vitreous fluids. J Clin Micro usefulness of two direct amplification technologies 1995;33:2643-2646. for the detection of Mycobacterium tuberculosis 35. Pikaart MJ, Villeponteau B. Suppression of PCR complex. Am J Clin Pathol 1998;110:301-310. amplification by high levels of RNA. Bio Techniques 15. Begovich AB, Erlich HA. HLA typing for bone mar­ 1993;14:24-25. row transplantation. New polymerase chain reaction- 36. Ahokas H, Erklala MJ. Interference of PCR ampli­ based methods. JAMA 1995;273:586-591. fication by the polyamines, spermine and spermi­ 16. Anderson C. Roche cuts controversial PCR fees, dine. PCR Meth Applic 1993;3:65-68. testing limits. Nature 1992;355:379. 37. Sarkar G, Kapelner S, Sommer SS. Formamide can 17. Kiechle FL. Residency in pathology: the William dramatically improve the specificity of PCR. Nucleic Beaumont Hospital perspective. Am J Clin Pathol Acids Res 1990;18:7465. 1993;100(Suppl 1):S29-S30. 38. Fujimura FK, Northrup H, Beaudet AL, O’Brien 18. Benge H, Csako G, Pari FF. A 10-year analysis of WE. Genotyping error with the polymerase chain “revenues”, costs, staffing and workload in an aca­ reaction. N Engl J Med 1990;322:61. demic medical center clinical chemistry laboratory. Clin Chem 1993;39:1780-1787. 39. He Q, Maijamala M, Soini H, Mertsola J, Viljanen MK. Primers are decisive for sensitivity of PCR. Bio 19. Fattal GA, Tanasijeric MJ, Winkelman JW. Evolu­ Techniques 1994;17:82-86. tion beyond the shared services model of consoli­ dated hospital clinical laboratories. Arch Pathol Lab 40. Weighardt F, Biamonti G, Riva S. A simple proce­ Med 1995;119:701-705. dure for enhancing PCR specificity. PCR Meth 20. Steiner JW, Root JM, Michel RL. The Transforma­ Applic 1993;3:77-80. tion o f Hospital Laboratories: Why Regionalization, 41. Compton J. Nucleic acid sequence-based amplifica­ Consolidation, and Reengineering Will Lead Labo­ tion. Nature 1991;350:91-92. ratories into the 21st Century. Hospital Technology 42. Gershon D. DNA diagnostic tools for the 21st cen­ Series (AHA012040). Chicago, IL: American Hospi­ tury. Nature Med 1995;1:102-103. tal Association, 1995;14(15):1—33. 43. Kiechle FL, Chambers LM, Cox RS, et al. Reference 21. Boyd JC, Felder RA, Savory J. Robotics and the Guide for Diagnostic Molecular Pathology/Flow changing face of the clinical laboratory. Clin Chem Cytometry: Fascicle VII. College of American 1996;42:1901-1910. Pathologists, 1996. 22. Felder RA. Automation of preanalytical processing 44. Farkas DH, Drevon AM, Kiechle FL, DiCarlo RG, and mobile robotics. IN: Kost GJ (ed). Handbook of Heath EM, Crisan D. Specimen stability for DNA- Clinical Automation, Robotics and Optimization. based diagnostic testing. Diag Mol Pathol 1996;5: New York: John Wiley and Sons, Inc., 1996:541-554. 227-235. 23. Markin RS. Recent trends in clinical laboratory auto­ 45. Farkas DH, Kaul KL, Wiedbrauk DL, Kiechle FL. mation. Clin Lab Manag Rev 1998;12:176-180. Specimen collection and storage for diagnostic 24. Roethe M, Wingfield S, Barranco P, Charache S. pathology investigation. Arch Pathol Lab Med 1996; Robotics in the hematology laboratory. An evaluation 120:591-596. of the productivity of the Sysmex HS-330. Am J Clin 46. Immunoglobulin and T-Cell Receptor Gene Rear­ Pathol 1995;103:154-158. rangement Assays: Approved Guidelines. Wayne, 25. Cheng J, Sheldon EL, Wu L, et al. Preparation and PA: National Committee for Clinical Laboratory hybridization analysis of DNA/RNA from E. coli on Standards, 1995:NCCLS Document MM2-A. microfabricated bioelectronic chips. Nature Biotech- 47. Makowski GS, Aslanzedeh J, Hopfer SM. In situ nol 1998;16:541-546. PCR amplification of Guthrie card DNA to detect 26. Robertson D. Chips advance but cost constraints cystic fibrosis mutations. Clin Chem 1995;41:471- remain. Nature Biotechnol 1998;16:509. 479. 27. Kopp MU, de Mello AJ, Manz A. Chemical amplifi­ 48. Makowski GS, Davis EL, Nadeau F, Hopfer SM. cation: continuous-flow PCR on a chip. Science Polymerase chain reaction amplification of Guthrie 1998;280:1046-1048. card deoxyribonucleic acid: extraction of nucleic acid 28. Ramsay G. DNA chips: State-of-the-art. Nature Bio­ from filter matrices. Ann Clin Lab Sci 1998;28:254- technol 1998;16:40-44. 259. MOLECULAR PATHOLOGY LABORATORY OF THE 21st CENTURY 75

49. Aso Y, Kawamura M, Hamaguchi Y, Shioiri T, Mit- 6 8 . Suomalainen A. Mitochondrial DNA and disease. suhashi M. Rapid, stable ambient storage of leuko­ Ann Med 1998;29:235-246. cyte RNA from whole blood. Clin Chem 1998;44: 69. Kiechle FL, Kaul KL, Farkas DH. Mitochondrial 1782-1783. disorders. Methods and specimen selection for diag­ 50. Pietrangel A. Hemochromatosis 1998: is one gene nostic molecular pathology. Arch Pathol Lab Med enough? J Hepatol 1998;29:502-509. 1996;120:597-603. 51. Burke W, Thompson E, Khourg MJ, et al. Heredi­ 70. Wallace DC, Singh G, Lott MT, et al. Mitochondrial tary hemochromatosis. Gene discovery and its impli­ DNA mutation associated with Leber’s hereditary cations for population-based screening. JAMA 1998; optic neuropathy. Science 1988;242:1427-1430. 280:172-178. 71. Macmullan CJ, Shoubridge EA. Mitochondrial DNA 52. Guidelines for the molecular genetics predictive test depletion: prevalence in a pediatric population in Huntington’s disease. Neurology 1994;44:1533- referred for neurologic evaluation. Pediatr Neurol 1536. 1996;14:203-210. 53. Park V, Howard-Peebles, P, Sherman P, Taylor A. 72. Bakker HD, Vanden Bogert C, Schottle HR, Zwart Wulfsberg E. Policy statement: American College of R, Wijberg FA, Spelbrenk JN. Fatal neonatal liver Medical Genetics. Fragile X syndrome: diagnostic failure and depletion of mitochondrial DNA in three and carrier testing. Am J Med Genet 1994;53:380- children in one family. J Inherit Metab Dis 1996;19: 381. 112-114. 54. American College of Medical Genetics. Prenatal 73. Maaswinkel-Mooij PD, Vanden Bogert C, Schölte interphase fluorescence in situ hybridization (FISH) HR, Onkenhout W, Brederoo P, Poorthius BJHM. policy statement. Am J Hum Genet 1997;53:526- Depletion of mitochondrial DNA in the liver of a 527. patient with lactic acidemia and hypoketotic hypo­ 55. Doll JJ. The patenting of DNA. Science 1998;280: glycemia. J Pediatr 1996;128:679-683. 689-690. 74. Zhang C, Peters LE, Linnane AW, Nagley P. Com­ 56. Auxter S. Patent issues cloud future of hereditary parison of different quantitative PCR procedures in hemochromatosis screening. Clin Lab News the analysis of the 4977-bp deletion in human mito­ 1997;23(12): 14. chondrial DNA. Biochem Biophys Res Commun 1996;223:450-455. 57. Kratsov VD, Greer JP, Whitlock JA, Koury MJ. Use 75. Marin-Garcia J, Goldenthal MJ, Ananthakrishraa R, of the microculture kinetic assay of apoptosis to et al. Specific mitochondrial DNA deletions in idio­ determine chemosensitivities of leukemias. Blood pathic dilated cardiomyopathy. Cardiovascular Res 1998;92:968-980. 1996;31:306-313. 58. Zhang X, Kiechle FL. Hoechst 33342-induced apop­ 76. Wei Y-H, Kao S-H, Lee H-C. Simultaneous increase tosis in BC3H-1 myocytes. Ann Clin Lab Sei 1997; of mitochondrial DNA deletions and lipid peroxida­ 27:260-275. tion in human aging. Ann NY Acad Sei 1996;786:24- 59. Zhang X, Kiechle FL. Mechanism of Hoechst 33342- 43. induced apoptosis in BC3H-1 myocytes. Ann Clin 77. Verma A, Moraes CT, Sherbert RT, Bradley WG. A Lab Sei 1998;28:104-114. MERRF/PEO overlap syndrome associated with the 60. Zhang X, Kiechle FL. Hoechst 33342 induces apop­ mitochondrial DNA 3243 mutation. Neurology 1996; tosis and alters TATA box binding protein/DNA 46:1334-1336. complexes in nuclei from BC3H-1 myocytes. Bio- 78. Silvestri G, Servidei S, Rana M, et al. A novel mito­ chem Biophys Res Commun 1998;248:18-21. chondrial DNA point mutation in the tRNA11“ gene is 61. Nicotera P, Brune B, Bagetta G. Nitric oxide: associated with progressive external ophthalmople­ inducer or suppressor of apoptosis? TIPS 1997;18: gia. Biochem Biophys Res Commun 1996;220:623- 189-190. 627. 62. Filep JG, Lapierre C, Lachance S, Chan JSD. Nitric 79. Maassen JA, Kadowaki T. Maternally inherited dia­ oxide co-operates with hydrogen peroxide in induc­ betes and deafness: a new diabetes subtype. Diabe- ing DNA fragmentation and cell lysis in murine lym­ tologia 1996;39:375-382. phoma cells. Biochem J 1997;321:897-901. 80. Sobreira C, Davidson M, King MP, Miranda AF. 63. Tabuchi A, Oh E, Taoka A, Sakurai H, Tsuchiya T, Dihydrorhodamine 123 identifies impaired mito­ Tsuda M. Rapid attenuation of AP-1 transcriptional chondrial respiratory chain function in cultured cells factors associated with nitric oxide (NO)-mediate harboring mitochondrial DNA mutations. J Histo- neuronal cell death. J Biol Chem 1996;271:31061- chem Cytochem 1996;44:571-579. 31067. 81. James AM, Wei Y-H, Pang C-Y, Murphy MP. 64. Lamont PJ, Surtees R, Woodward CE, Leonard JV, Altered mitochondrial function in fibroblasts con­ Wood NW, Harding AE. Clinical and laboratory taining MELAS or MERRF mitochondrial muta­ findings in referrals for mitochondrial DNA analysis. tions. Biochem J 1996;318:401-407. Arch Dis Child 1998;79:22-27. 82. Hanekamp JS, Thilly WG, Chaudhry MA. Screening 65. DiMauro S, Wallace DC (eds). Mitochondrial DNA for human mitochondrial DNA polymorphisms with in Human Pathology. New York: Raven Press; 1993. denaturing gradient gel electrophoresis. Hum Genet 6 6 . Ropp PA, Copeland WC. Cloning and characteriza­ 1996;98:243-245. tion of human mitochondrial DNA polymerase, 83. Alonso A, Martin P, Albarran C, Garcia O, Sancho DNA polymerase y. Genomics 1996;36:449-458. M. Rapid detection of sequence polymorphisms in 67. Mori M, Terada K. Mitochondrial protein import in the human mitochondrial DNA control region by animals. Biochim Biophys Acta 1998;1403:12-27. polymerase chain reaction and single-strand confor­ 76 KIECHLE, ZHANG, AND MALINSKI

mation analysis if mutation detection enhancement 100. Gerbitz K-D, van den Ouweland JMW, Maassen JA, gels. Electrophoresis 1996;17:1299-1301. Jaksch M. Mitochondrial diabetes mellitus: a review. 84. Attimonelli M, Calö D, Cooper JM, et al. The mit- Biochim Biophys Acta 1995;1271:253-260. BASE human data set structure. Nucl Acids Res 101. Report of the expert committee on the diagnosis and 1998;26:116-119. classification of diabetes mellitus. Diabetes Care 85. Fromenty B, Manfredi G, Sadlock J, Zhang L, King 1997;20:1183-1197. MP, Schon EA. Efficient and specific amplification 102. Regazzi R, Verchere CB, Halban PA, Polonsky KS. of identified partial duplications of human mitochon­ Insulin production: from gene to granule. Diabeto- drial DNA by long PCR. Biochim Biophys Acta 1996; logia 1997;40:B33-B38. 1308:222-230. 103. Wolfsdorf JI. Hyperinsulinemic hypoglycemia of infancy. J Pediatr 1998;132:1-3. 8 6 . Hu G, Thilly WG. Multi-copy nuclear pseudogenes of mitochondrial DNA reveal recent acute genetic 104. Matschinsky FM, Glaser B, Magnuson MA. Pancre­ changes in the human genome. Curr Genet 1995;28: atic ß-cell glucokinase. Closing the gap between 410-114. theoretical concepts and experimental realities. Dia­ betes 1998;47:307-315. 87. Avctander P. Comparison of a mitochondrial gene 105. Velho G, Blanche H, Vaxillaire M, et al. Identifica­ and a corresponding nuclear pseudogene. Proc R tion of 14 new glucokinase mutations and description Soc Lond B 1995;262:13-19. of the clinical profile of 42 MODY-2 families. Dia- 8 8 . Zhang D-X, Hewitt GM. Nuclear integrations: chal­ betologia 1997;40:217-224. lenges for mitochondrial DNA markers. Trends Ecol 106. Mueckler M, Kruse M, Strube M, et al. A mutation Evol 1996;11:247-251. in the Glut 2 glucose transporter gene of a diabetic 89. Parfait B, Rustin P, Munnich A, Rötig A. Coampli­ patient abolishes transport activity. J Biol Chem fication of nuclear pseudogenes and assessment of 1994;269:17765-17767. heteroplasmy of mitochondrial DNA mutations. Bio- 107. Santer R, Schneppenheim R, Dombrowski A, Götze chem Biophys Res commun 1998;247:57-59. H, Steinman B, Schaub J. Mutations in GLUT2, the 90. Zullo S, Kennedy JL, Gelenter J, et al. Eliminating gene for the liver-type glucose transporter, in mitochondrial DNA competition for nuclear DNA patients with Fanconi-Bickel syndrome. Nature primers. PCR Methods Appl 1993;3:39-45. Genet 1997;17:324-326. 91. DeSalle R, William AK, George M. Isolation and 108. Santer R, Schneppenheim R, Dombrowski A, Götze characterization of animal mitochondrial DNA. H, Steinman B, Schaub J. Fanconi-Bickel syn- Methods Enzymol 1993;224:176-204. drome-a congenital defect of the liver-type faeilita- 92. Gailermann N, Retzlaff S, Wang Y-L, et al. A het- tive glucose transporter. J Inher Metab Dis 1998;21: eroplasmic point mutation of mitochondrial 191-194. tRNALeu

117. Silvestri G, Bertini E, Servidei S, et al. Maternally 126. Flier JS. Lilly lecture: syndromes of insulin resis­ inherited cardiomyopathy: a new phenotype associ­ tance. From patient to gene and back again. Diabe­ ated with the A to G at nt 3243 of mitochondrial tes 1992;41:1207-1209. DNA (MELAS mutation). Muscle Nerve 1997;20: 127. Velho G, Froguel PH. Genetic determinants of non- 221-225. insulin-dependent diabetes mellitus: strategies and 118. Suzuki Y, Hosokawa K, Suzuki S, et al. Diabetes asso­ recent results. Diabetes & Metabolism (Paris) 1997; ciated with a novel 3264 mitochondrial tRNALeu