Microarray and microfluidic methodology for genotyping cytokine gene polymorphisms

Microarray and Microfluidic Methodology for Genotyping Cytokine Gene Polymorphisms Youvraj R. Sohni, James R. Cerhan, and Dennis O’Kane ABSTRACT: Cytokine genetic polymorphisms are the subject of disease-association studies that require largescale human genotyping. Polymerase chain reaction based custom microarrays and microfluidics systems were used to develop genotyping assays for following cytokine polymorphisms: tumor necrosis factor-␣ G-308A, interleukin-4 (IL-4) C-589T, interferon-␥ (CA)n repeats, IL-1RN 86-bp variable number of tandem repeats (VNTR), and CCR5 32-bp indel. For G-308A, 70.9% of DNA samples assayed were homozygous for wild type, 25.5% were heterozygous, and none were homozygous for variant allele. For C-589T, 35.5% of DNA samples were homozygous for wild type, 38% were heterozygous, and 22% were homozygous for variant. For IL-1RN VNTR, 71% of DNA samples were homozygous and the remainder were heterozygous. For CCR5, 96.4% of amplicons were homozygous for wild type, and 3.6% were heterozygous containing deletion. For IFN-␥ (CA)n repeats, 35.6% had

INTRODUCTION Cytokines may be produced by multiple diverse cell types and regulate magnitude and nature of immune responses. A hallmark of the immune system is its redundancy due to pleiotropism exhibited by cytokines with multiple effects on different target cells [1]. Genetic polymorphisms in cytokines affect gene transcription and cause interindividual variations in cytokine production, thus influencing the outcome of infectious diseases, cancers, and autoimmune diseases (see Table 1). Technologic advancements fueled by the genomic revolution have resulted in identification of increasing numbers of DNA sequence variations resulting in increasing number of disease-association studies as understanding

From the Mayo Clinic Cancer Center Microarray Shared Resource (Y.R.S., D.O.), Department of Health Sciences Research (J.R.C.), and Departments of Laboratory Medicine and Pathology (D.O.), Mayo Clinic, Rochester, MN. Address reprint requests to: Dr. Youvraj R. Sohni, MCCC Microarray Shared Resource, Guggenheim 10-11B, Mayo Clinic, 200 First St SW, Rochester MN 55905; Tel: (507) 284-4850; Fax: (507) 266-0824; E-mail: [email protected]. Received May 6, 2003; revised July 3, 2003; accepted July 8, 2003. Human Immunology 64, 990 –997 (2003) © American Society for Histocompatibility and Immunogenetics, 2003 Published by Elsevier Inc.

2,2 alleles, 42.2% had 2,3 alleles, and 11% had 3,3 alleles with alleles 1 through 5 corresponding to 11 through 15 repeats, respectively. There was good concordance between the results we obtained and current “gold-standard” methodologies for analyzing single nucleotide polymorphisms and size polymorphisms. Electronic DNA concentration with high stringency predisposes microarray technology to hybridization fidelity and accuracy, and microfluidics systems outperform conventional methodologies for size polymorphisms. Comprehensive genotyping can be achieved for clinical epidemiologic studies on cytokine gene polymorphisms using this approach. Human Immunology 64, 990 –997 (2003). © American Society for Histocompatibility and Immunogenetics, 2003. Published by Elsevier Inc. KEYWORDS: cytokines; genotyping; microarrays; microfluidics; polymorphisms

the genetic basis of disease is expected to transform diagnosis, treatment, and prevention. Association studies with a large sample size, in which cases of disease are compared with matched controls from the same population, are likely to have a greater chance of detecting small effects. Also, as opposed to pedigree analysis, association studies with genes containing common variants that have a small effect on disease risk have the analytical power to decrease sample size [2, 3]. These clinical epidemiologic studies would benefit from genotyping aimed at detecting all genetic variations, which requires application and use of several technologies. We have previously used polymerase chain reaction (PCR) based microarray and microfluidic chip-based technologies to sequentially probe single nucleotide polymorphisms (SNPs) in long amplicons [4] and to analyze size polymorphisms, respectively [5]. In this article we describe methodology to analyze genetic polymorphisms in cytokines. The methodology we describe is useful for high throughput genetic analyses of several different types of cytokine genetic variations with accuracy, reproducibility, and sensitivity. Table 1 lists the cytokine polymorphisms 0198-8859/03/$–see front matter doi:10.1016/S0198-8859(03)00174-5

Microarray and Microfluidic Cytokine Genotyping

991

TABLE 1 List of polymorphisms analyzed in study Gene location

Polymorphism

Assay system

Functional data

TNF␣ 6p21.3

G-308A (promoter)

Microarray chip

A allele associated with higher TNF␣ levels

IL-4 5q23-3I

C-589T (promoter)

Microarray chip

T allele associated with higher IL-4 activity and IgE levels

IL-1RN 2q14.2 86-bp VNTR (intron 2) Microfluidic chip Allele 2 associated with higher IL-1b levels in vitro

IFN-␥ 12q24.1 CA repeat (intron 1) CCR5

32-bp deletion

Microfluidic/ *2 allele associated with microarray higher IFN-␥ levels chip Microfluidic chip Truncated protein is nonfunctional

Disease association *A allele associated with NHL and CLL, sepsis, cerebral malaria, mucocutaneous leishmaniasis, and scarring trachoma, among others. *T allele associated with elevated IgE levels in asthmatic families and is a risk factor for development of atopy, asthma, and rhinitis in infants. *2 allele associated with lupus erythematosus, ulcerative colitis, alopecia areata, diabetic nephropathy, and chronic hypochlorhydria in response to H. pylori infection, and increased risk of gastric cancer. *2 allele associated with allograft fibrosis. Disease progression appears slower in CCR5 deletion heterozygotes than in individuals homozygous for the normal CCR5 gene.

Abbreviations: IFN ⫽ interferon; IL ⫽ interluekin; TNF ⫽ tumor necrosis factor; VNTR ⫽ variable number of tandem repeats.

analyzed in the current study and summarizes their functions and disease association. Custom microarrays were used to analyze tumor necrosis factor-␣ (TNF␣) G-308A, and interleukin-4 (IL-4) C-589T SNPs. Microfluidic chip-based methodology was used to genotype IL-1RN 86-bp variable number of tandem repeats (VNTR) and CCR5 32-bp deletion polymorphism. A combination of microfluidic and customizable microarrays were used to genotype the interferon-␥ (IFN-␥) (CA)n microsatellite polymorphism.

MATERIALS AND METHODS Polymerase Chain Reaction The PCR amplifications were optimized and performed for all loci analyzed in this study. The list of primers (IDT, Coralville, IA, USA) and PCR conditions are illustrated in Table 2. DNA samples were obtained from DNA Polymorphism Discovery Resource of Coriell Cell Repositories, which has cell lines and DNA from 450 anonymous,

TABLE 2 Primers and PCR conditions Polymorphism

Primers (5⬘-3⬘) sense/antisense

PCR mixture

25 ␮l AmpliTaq Gold Master Mix, 1 ␮M primers, DNA template, and water to 50 ␮l IL-4b Biotin-GCCTCACCTGATACGACCTG 25 ␮l AmpliTaq Gold Master C-589T TGTCCGAATTTGTTGTAATGC Mix, l ␮M primers, DNA template, and water to 50 ␮l IL-1RNc CCCCTCAGCAACACTCC 12.5 ␮l AmpliTaq Gold Master 86-bp GGTCAGAAGGGCAGAGA Mix, l ␮M Primers, DNA VNTR template, and water to 25 ␮l CCR5d GTCTCTCCCAGGAATCATCTTTACCAGATCTC 12.5 ␮l AmpliTaq Gold Master 32-bp TTAGGATTCCCGAGTAGCAGATGACCATGACA Mix, l ␮M Primers, DNA deletion template, and water to 25 ␮l IFN-␥e Biotin-GCTGTCATAATAATATTCAGAC 25 ␮l AmpliTaq Gold Master CA repeats CGAGCTTTAAAAGATAGTTCC Mix, l ␮M primers, DNA template, and water to 50 ␮l TNF␣a G-308A

a

Biotin-AAGGAAACAGACCACAGACCTG GGATACCCCTCACACTCCCC

Thermal cycling 95 °C for 10 min; 30 cycles of 94 °C, 60 °C, and 72 °C for 1 min each; and 72 °C for 7 min 95 °C for 10 min; 30 cycles of 95 °C, 56 °C, and 72 °C for 1 min each; and 72 °C for 5 min 95 °C for 10 min; 35 cycles of 94 °C, 60 °C, and 72 °C for 30 sec each; and 72 °C for 7 min 95 °C for 10 min; 35 cycles of 94 °C, 60 °C, and 72 °C for 30 sec each, and 72 °C for 7 min 95 °C for 10 min; 35 cycles of 95 °C, 56 °C, and 72 °C for 1 min each; and 72 °C for 7 min

Mira et al. [16]; b Primers designed in-house using Oligo 6.61 software and GenBank sequence; c El-Omar et al [12]; d Lu et al. [38]; e Pravica et al. [13]. Abbreviations: IFN ⫽ interferon; IL ⫽ interleukin; PCR ⫽ polymerase chain reaction; TNF ⫽ tumor necrosis factor.

992

unrelated individuals [6]. As supplied each DNA sample has a concentration between 300 and 400 ng/␮l, and these were diluted to a concentration of 20 ng/␮l for PCR. No information is included with respect to disease state or ethnicity, and no inferences on population genetics should be made from allele frequencies identified. Microarray Genotyping of TNF␣ G-308A and IL-4 C-589T SNP SNPs were genotyped using the NanoChip platform (Nanogen, San Diego, CA, USA). The principle of this methodology and its application has been previously described [7–10]. Briefly, a biotinylated PCR primer was used to allow for attachment of amplicon to streptavidin in the agarose permeation layer of the microarray. The choice of primer to be biotinylated was dictated by favorable base stacking interactions between anchor stabilizer oligonucleotide and reporter probe oligonucleotide, which together form a sequence complimentary to amplicon region containing the SNP. Cy3- and Cy5labeled wild type and variant reporter probes, respectively, were designed with SNP as 3⬘terminal base and Tm within 2 °C of each other. Following PCR, biotinylated amplicons were desalted using Millipore Multiscreen PCR plates (Billerica, MA, USA) and transferred to a Nunc V-bottom plate and resuspended in 50 mmol/L L-histidine buffer to a final concentration ranging between 5 and 40 nmol/L in a total volume of 60 ␮l. Amplicons and L-histidine buffer blanks were electronically addressed by software-driven mapping to user-designated sites. Chip array was then passively treated with 0.1-M NaOH to denature amplicons and for loading the hybridization mixture for probing. Hybridization mixture consisted of 250 nmol/L stabilizer oligonucleotide and 500 nmol/L reporter probe oligonucleotides in a high salt buffer. Microarray was imaged using separate lasers for both Cy3 and Cy5, and temperature was used to discriminate between matched and mismatched reporters. Known heterozygotes verified by dye-terminator bidirectional sequencing performed on ABI 377 DNA sequencers (Applied Biosystems, Foster City, CA, USA), were used to normalize hybridization efficiency between dye-labeled reporters. Genotypes were designated based on biallelic fluorescence intensity ratios. A biallelic fluorescence intensity ratio of ⱕ 1:3 was deemed heterozygous and a ratio of ⱖ 1:5 deemed homozygous. There was no genotype designation made for fluorescence intensity ratios in between 1:3 and 1:5. Homozygous wild type alleles hybridized with Cy3 labeled reporter probe, whereas homozygous variant alleles hybridized only with Cy5 labeled reporter probe. Each heterozygous complex hybridized with both labeled probes for each allele pair tested.

Y. R. Sohni et al.

Microfluidic Chip Analysis of IL-1RN VNTR and CCR5 Indels For size polymorphisms, PCR products were analyzed using DNA 500 (or 1000) LabChip kit on an Agilent 2100 Bioanalyzer (Agilent Technologies, Wilmington, DE, USA) as described previously [5]. Chips were prepared according to instructions provided by the manufacturer. Typically, 25 ␮l of dye concentrate was added into a DNA gel matrix vial, vortexed and the gel-dye mix microcentrifuged through a spin filter at 2240g for 15 minutes. Gel-dye mix, 5 ␮l of markers per sample well, 1 ␮l of biosizing ladder, and 1 ␮l of each test sample were loaded per kit protocol. During the assay electrokinetic sample application to the chip permits mass-based separation of DNA fragments in microchannels filled with sieving polymer and fluorescent dye detection using fluorescence imaging. Fragments are sized using a sizing ladder consisting of a mixture of DNA fragments with preset concentrations and sizes and standard curve of migration time against DNA size is plotted from the DNA sizing ladder by linear interpolation. Upper and lower markers bracketing the DNA sizing range serve as internal standards aligning the ladder well with individual sample wells. DNA fragment sizes were determined for each well from the calibration curve in conjunction with markers. Genotypes are designated based on fragment sizes obtained at the end of the run. Genotyping of IFN-␥ CA Dinucleotide Repeat Polymorphism Polymerase chain reaction products were initially analyzed using Agilent 2100 Bioanalyzer for sizing and initial low-resolution analysis. However, based on the approximate size obtained (⫾ 5 bp) the precise number of repeats can be accurately genotyped using microarray analysis. The advantage of this approach is that it is not necessary to probe for all possible number of repeats for a given microsatellite polymorphism. In other studies (unpublished data) we have used this approach for the highly polymorphic heme oxygenase-1 promoter CA repeats, which vary in number between 15 and 40. Instead of probing each sample for all possible repeats, which would be cumbersome, one can use the probes that correlate with the approximate size obtained. We used the NanoChip array for high-resolution accurate analysis of the microsatellite. The molecular strategy behind this hybridization-based methodology has previously been described [11]. Briefly, target DNA directs juxtaposition of 3⬘-terminal base of capture oligonucleotide and 5⬘-terminal base of reporter resulting in base stacking between capture and reporter that stabilizes DNA hybridization. A mismatch is indicated between spatial association of capture and reporter oli-

Microarray and Microfluidic Cytokine Genotyping

993

TABLE 3 List of oligonucleotide probes used for analyzing SNPs and microsatellite Polymorphism

Stabilizer oligonucleotide (5⬘-3⬘)

Reporter probes (5⬘-3⬘)

TNF␣ G-308A

CATGCCCCTCAAAACCTATTGCCTCCATTTCTTTTGGGGA

IL-4 C-589T

AGAATAACAGGCAGACTCTCCTACCCCAGCACTGGGG

IFN-␥ (CA)n repeats

(TG)nGAGATTTGATTTTGTGTTGTAAGAATAAAATACA where n ⫽ number of repeats from 11 through 15

Cy3-ACCCCGTCC C (wild type) Cy5-AAC CCC GTC CT (variant) GACAATGTTCTCC-Cy3 (wild type) AACAATGTTCTCC-Cy5 (variant) Cy3-ATGTGCGAGTG

Abbreviations: IFN ⫽ interferon; IL ⫽ interleukin; SNP ⫽ single nucleotide polymorphism; TNF ⫽ tumor necrosis factor.

chip array for chemically denaturing the amplicons to ensure the single-strandedness of the amplicon prior to stabilizer address. Finally a capture map addressed the stabilizer oligonucleotides to user-specified array site for probing each target amplicon. Each stabilizer was used at a concentration of 250 nmol/L. At the end of amplicon and stabilizer hybridization, Cy3-labeled reporter was manually added to array to permit passive hybridization with target. The reporter concentration was 500 nmol/L in a final volume of 60 ␮l. Thermal discrimination was performed in the instrument reader over a range of temperatures in 2 °C increments to obtain the optimal temperature for reporter discrimination. Chip array was imaged with green laser to determine the test sites containing fluorescent reporter.

gonucleotide termini, which will contain a gap or an overlap if the number of corresponding repeats is fewer or greater, respectively. Hybridization complex will be formed with contiguous capture and reporter termini in case of a perfect match. Differential stability of reporter can be ascertained under chemical and thermal stringency conditions and the number of repeats determined. Capture stabilizer oligonucleotides were designed for each specific allelic locus with one stabilizer for each repeat allele. Length of stabilizer depended on the number of repeat sequences to be probed. Length of stabilizer was n-1 repeat units plus flanking sequence where n is the number of CA repeats in the amplicon. Reporter oligonucleotides were designed to have bases that are complimentary to the unique sequence of the locus plus one repeat unit on the 5⬘-end of the oligonucleotide. Reporter was labeled with fluorophore Cy3 on its 3⬘end. Together the stabilizer and reporter oligonucleotides were complimentary to the biotinylated amplicon strand. Stabilizer and reporter oligonucleotide sequences used are presented in Table 3. Biotinylated desalted PCR products were combined with 100 mM L-histidine (Sigma Chemical Co., St. Louis, MO, USA) buffer in 96-well Nunc v-bottom plate (Nunc, Rockilde, Denmark) for a final volume of 60 ␮l. Amplicons were electronically addressed by softwaredriven mapping to individual array sites. For background control, blanks containing 60 ␮l of 50 mM L-histidine were simultaneously addressed to the chip array. A second user-defined loading protocol was mapped to passively address 0.1-M NaOH for 5 minutes to the

Verification of Analyses For SNPs, amplifications were performed on ten random DNA samples and bidirectionally sequenced using dyeterminator chemistry on an ABI 377 DNA sequencer. For size polymorphisms, 6-FAM-labeled PCR products (0.75 ␮l) and GeneScan 500 (Applied Biosystems) internal lane size standards were analyzed using GeneScan 3.0. RESULTS The microarray SNP genotyping results are summarized in Table 4. For TNF␣ G-308A, 39 of 55 (70.9%) DNA samples assayed were homozygous for wild type (GG), 14 of 55 (25.5%) were heterozygous (GA), and none were

TABLE 4 Summary of SNP genotyping results (number ⫽ 55) Gene a

IL-4 C-589Tb

TNF␣ G-308A Alleles Frequency a

GG 39 (70.9%)

GA 14 (25.5%)

AA 0

CC 19 (35.5%)

CT 21 (38.2%)

Two samples could not be designated genotype due to biallelic fluorescence intensity ratios falling between 1:3 and 1:5. Three samples could not be designated genotype due to biallelic fluorescence intensity ratios falling between 1:3 and 1:5. Abbreviation: SNP ⫽ single nucleotide polymorphism.

b

TT 12 (22%)

994

Y. R. Sohni et al.

TABLE 5 Summary of VNTR, indel, and microsatellite genotyping results (number ⫽ 55) IL-1RN 86-bp VNTRa allele distribution

IFN-␥ CA repeatsb allele distribution (n ⫽ 45)

CCR5 32-bp deletion allele distribution

1,1

1,2

2,2

2,4

CCR5

CCR5, ⌬32

⌬32

2,2

2,3

3,3

36 (65.5%)

14 (25.5%)

3 (5.5%)

1 (1.8%)

53 (96.4%)

2 (3.6%)

0

16 (35.6%)

19 (42.2%)

5 (11.1%)

a

One sample PCR failed. Five samples were designated no calls. Abbreviations: IFN ⫽ interferon; IL ⫽ interleukin; PCR ⫽ polymerase chain reaction; VNTR ⫽ variable number of tandem repeats.

b

homozygous for variant (AA). For IL-4 C-589T, 19 of 55 (35.5%) DNA samples were homozygous for wild type (CC), 21 of 55 (38%) were heterozygous (CT), and 12 (22%) were homozygous for variant. All amplicons were genotyped in a single hybridization run on one microarray following PCR. No genotypes could be assigned for two amplicons for G-308A and three amplicons for C-589T, respectively, because the biallelic fluorescence intensity ratio was between 1:3 and 1:5. The microfluidic chip-based genotyping for size polymorphisms are summarized in Table 5. For IL-1RN 86-bp VNTR, a single PCR was performed for each of the 55 DNA samples followed by microfluidic chipbased analysis. Alleles were sized and coded by criteria established by El-Omar et al. [12]. Following analyses, 36 of 55 (65.5%) amplicons assayed were homozygous with four repeats (1, 1 alleles), 14 of 55 (25.5%) were heterozygous with four and two repeats (1,2 alleles), 3 of 55 (5.5%) were homozygous with two repeats (2,2 alleles), and 1 of 55 amplicons (1.8%) assayed was heterozygous with two and three repeats (2, 4 alleles). For CCR5, 53 of 55 (96.4%) amplicons were homozygous for the wild type and 2 of 55 (3.6%) amplicons had the heterozygous genotype (CCR5, ⌬32). There were no samples with the homozygous genotype (⌬32, ⌬32) for the 32-bp deletion. For all the 55 DNA samples, a single PCR followed by microfluidic analysis was performed in one run. The coding criteria established by Pravica et al. [13] were allele designation for IFN-␥ (CA)n repeats. Alleles 1 through 5 correspond to 11 through 15 CA repeats. Of the 45 DNA samples assayed, 16 (35.6%) carried 2,2 alleles, 19 (42.2%) had 2,3 alleles, 5 (11%) had 3,3 alleles, and 5 samples could not be designated a genotype. Alleles 2 and 3 correspond to 12 and 13 dinucleotide repeats, respectively. All DNA samples were genotyped following a single PCR and single microarray analysis. We have confirmed our results with the existent “gold-standard” sequencing and GeneScan analyses for size polymorphisms. Ten samples sequenced for SNPs and five samples analyzed for size polymorphisms exhibited complete concordance.

DISCUSSION The genetic loci analyzed in current study are functional and common polymorphisms involved in immune function and regulation. TNF␣ is a proinflammatory cytokine that provides a rapid form of host defense against infection and is principal mediator of response to gramnegative bacteria. The TNF␣ polymorphism has direct effect on gene regulation and may be responsible for high TNF␣ phenotype, which simulates deleterious effects, induced by bacterial endotoxins [14]. The *A allele is associated as a susceptibility factor in non-Hodgkin’s lymphoma, chronic lymphocytic leukemia [15], and septic shock [16]. In the current study, no homozygous variants (AA) were found though 25.5% of DNA samples were heterozygous. Due to its genetic diversity and the fact that it acts against a variety of pathogens, TNF␣ has been the subject of association studies, among others, on cerebral malaria [17], mucocutaneous leishmaniasis [18], and scarring trachoma [19]. IL-4 is required for IgE production and plays a critical role in eosinophil-mediated inflammatory reactions and is a growth and differentiating factor for T cells [1]. The C-589T polymorphism has been associated with elevated IgE levels in asthmatic families [20 –22] and is risk factor for development of atopy, asthma, and rhinitis in infants. [23]. In the current study, 38% of DNA samples were heterozygous and 22% were homozygous for variant allele. IL-1RN is a naturally occurring competitive inhibitor of IL-1, with which it is structurally homologous. Though biologically inactive, IL-1RN may serve to regulate IL-1 action [1]. IL-1RN polymorphism is a VNTR of 86-bp sequence that has 5 alleles between 2 and 6 repeats. The 4 repeat and 2 repeat alleles are the most common and in the current study, 32.8% of DNA samples assayed carried *2 allele in heterozygous conditions that are associated with chronic inflammatory diseases, such as lupus erythematosus [24], ulcerative colitis [25], and alopecia areata [26]. In addition, *2 is also associated with diabetic nephropathy [27] and IL-1RN/IL-1RN2 genotype is associated with susceptibility to chronic hy-

Microarray and Microfluidic Cytokine Genotyping

pochlorhydria in response to Helicobacter pylori infection, and increased risk of gastric cancer [12]. The chemokine receptor CCR5 is expressed on T cells, monocytes, and tissue macrophages, among other cell types. Individuals that do not express this receptor are highly resistant to HIV infection [1]. A 32-bp deletion allele was identified in CCR5 gene and the frequency of CCR5 deletion heterozygotes was significantly elevated in groups of individuals that had survived HIV-1 infection for more than 10 years. There were no samples homozygous for the deletion polymorphisms though 3.6% of DNA samples assayed carried the deletion under heterozygous conditions. Samson et al. [29] report that the 32-bp deletion results in a frameshift generating a nonfunctional receptor. The frequency of the 32-bp deletion allele in the CCR5 gene was significantly elevated in groups of individuals that survived HIV-1 infection for more than 10 years. Disease progression appears to be slower in CCR5 deletion heterozygotes than in individuals homozygous for the normal CCR5 gene [28, 38]. IFN-␥ is a potent activator of mononuclear phagocytes and promotes T lymphocyte differentiation and IgG subclass switching. It also acts on neutrophils, stimulates natural killer cellular cytolytic activity, and promotes macrophage-rich inflammatory and IgE-dependent reactions [1]. Pravica et al. [13] identified the presence of variable length dinucleotide (CA) repeat polymorphism in the IFN-␥ gene. In current study, 77.8% of samples assayed were heterozygous for allele 2, which has significant correlation with in vitro IFN-␥ production. The same allele 2 was associated with allograft fibrosis [30]. Due to the critical and decisive role cytokines play in the regulation of immune system and their genetic diversity that affect transcription leading to interindividual phenotypic variations, they are the subject of such scrutiny for studying disease association as discussed above. Such genetic analyses are essential in clinical epidemiologic studies in which large numbers of individuals need to be genotyped for many markers and such studies have increasingly motivated efforts toward superior marker detection and genotyping technology. Therefore, it is important to develop reliable methodologies to analyze these mutations for diagnosis and treatment. Conventional technologies, such as gel electrophoresis, restriction fragment length polymorphism (RFLP), passive probe hybridization, and DNA sequencing tend to be labor-intensive and time consuming. In case of RFLP, only in 50% of cases is there an altered restriction site polymorphism due to SNP variants, which is a serious limitation, and in some instances the assay is difficult to reproduce [31]. Microarray and microfluidic chip-based technologies are powerful tools for analyses of genetic polymorphisms.

995

In the foreseeable future these technologies are expected to be integrated into a routine clinical laboratory setting for diagnostic and therapeutic purposes as part of medical practice. Such technologic advancement fueled by the genomic revolution is imperative for development of new clinical laboratory tests as part of DNA-based molecular diagnostics [32]. Custom microarrays are useful for high-throughput genotyping particularly in clinical epidemiologic studies. The microarray technology that we used for genotyping is based upon the principle that efficiency of probe binding depends upon target/probe similarity because most stable duplexes are formed by fully complimentary sequences. Differential stability between perfect and imperfect duplexes is used for mutation detection and abetted by thermal and chemical stringency for maximal mutation detection. Under optimized assay conditions, the one-base mismatch sufficiently destabilizes hybridization to prevent the allelic probe from annealing to the target sequence [33]. A perfect duplex remains stable whereas a mismatch is unstable with a single mismatch decreasing stability by 3 to 10 °C [34]. Microarray methodology permits high throughput, costeffective molecular diagnostic mutation analysis [35]. In case of microsatellite genotyping assay, it is difficult to design assays with synchronous hybridization as stable hybridization can occur without alignment of unique flanking sequences. However, in our methodology rapid electronic DNA concentration accompanied with high stringency conditions predispose to fidelity and accuracy of target hybridization [11]. The microfluidic system that we used for genotyping IL1-RN and CCR5 polymorphisms outperform conventional gel-electrophoresis-based genotyping technologies. It is at least 25-fold more sensitive than conventional ethidium bromide stained gel-based methodology. This is very critical because selective amplification of heterozygous alleles with more efficient amplification of shorter alleles may lead to false genotype assignments. PCR parameters, such as Mg⫹ concentration, denaturation temperature, and cycle number influence amplification reproducibility. Needless to say, post-PCR technologies cannot be more accurate than the PCR itself and false genotyping results will be inconspicuous unless there are large deviations from the Hardy-Weinberg equilibrium [36]. In addition, the microfluidics method is timely and generates comparable data from samples prepared under a variety of conditions. This technology improves assay throughput as well as improves data quality while reducing human errors. Another advantage of this methodology is that unlabeled PCR products can be directly loaded and analyzed. [5, 37]. Large-scale human genotyping requires technologies with minimal number of steps, high accuracy, and ability

996

to automate at a reasonable cost. The methodologies that we described were developed easily and rapidly based on sequence information. The assays are robust and compared well with current gold-standard methodologies. Although we analyzed SNPs, VNTRs, and microsatellites across five different cytokine genes, it demonstrates that a combined used of these methodologies can generate genotyping data that is fairly comprehensive. We are aware that these technologies are useful for highthroughput mutation detection but not mutation scanning. However, we foresee that in the not too distant future, these technologies can be integrated in the clinical laboratory as part of routine medical practice as the robustness of these assays withstands variations in DNA source and quality without compromising genotyping quality. ACKNOWLEDGMENTS

This work was supported by Mayo Foundation and NCI grant RO1 CA92153.

REFERENCES 1. Abbas AK, Lichtman AH, Pober JS: Cellular and Molecular Immunology. Third Edition. Philadelphia: WB Saunders, 1997. 2. Collins FS, Guyer MS, Chakravarti A: Variations on a theme: cataloguing human DNA sequence variation. Science 278:1580, 1997. 3. Gray IC, Campbell DA, Spurr NK: Single nucleotide polymorphisms as tools in human genetics. Human Mol Genet 9:2403, 2000. 4. Sohni YR, Dukek D, Taylor W, Ricart E, Sandborn W, O’Kane D: Active electronic arrays for genotyping of NAT2 polymorphisms. Clin Chem 47:1922, 2001. 5. Sohni YR, Burke JP, Dyck PJ, O’Kane D: Microfluidic chip-based method for genotyping microsatellites, VNTRs and insertion/deletion polymorphisms. Clin Biochem 36:35, 2003. 6. Collins FS, Brooks LD, Chakravarti A: A DNA polymorphism discovery resource for research on human genetic variation. Genome Res 8:1229, 1998. 7. Gilles PN, Wu DJ, Foster CB, Dillon PJ, Chanock SJ: Single nucleotide polymorphic discrimination by an electric dot blot assay on semiconductor microchips. Nature Biotech 17:365, 1999. 8. Sosnowski RG, Tu E, Butler WF, O’Connell JP, Heller MJ: Rapid determination of single base mismatch mutations in DNA hybrids by direct electric field control. Proc Natl Acad Sci USA 94:1119, 1997. 9. Edman CF, Raymond DE, Wu DJ, Tu E, Sosnowski RG, Butler WF, Nerenberg M, Heller MJ: Electric field directed nucleic acid hybridization on microchips. Nucleic Acids Res 25:4907, 1997.

Y. R. Sohni et al.

10. Gurtner C, Tu E, Jamshidi N, Haigis RW, Onofrey TJ, Edman CF, Sosnowski R, Wallace B, Heller MJ: Microelectronic array devices and techniques for electric field enhanced DNA hybridization in low-conductance buffers. Electrophoresis 23:1543, 2002. 11. Radtkey R, Feng L, Muralhidhar M, Duhon M, Canter D, DiPierro D, Fellon S, Tu E, McElfresh K, Nerenberg M, Sosnoswski R: Rapid, high fidelity analysis of simple sequence repeats on an electronically active DNA microchip. Nucleic Acids Res 28:e17, 2000. 12. El-Omar EM, Carrington M, Chow W-H, McColl KEL, Bream JH, Young HA, Herrera J, Lissowska J, Yuan C-C, Rothman N, Lanyon G, Martin M, Fraumeni Jr JF, Rabkin CS: Interleukin-1 polymorphisms associated with increased risk of gastric cancer. Nature 404:398, 2000. 13. Pravica V, Asderakis A, Perrey C, Hajeer A, Sinnott PJ, Hutchinson IV: In vitro production of IFN-␥ correlates with CA repeat polymorphism in the human IFN-␥ gene. Eur J Immunogenet 26:1, 1999. 14. Wilson AG, Symons JA, McDowell TL, McDevitt HO, Duff GW: Effects of a polymorphism in the human tumor necrosis factor ␣ promoter on transcriptional activation. Proc Natl Acad Sci USA 94:3195, 1997. 15. Chouchane L, Ahmed SB, Baccouche S, Remadi S: Polymorphism in the tumor necrosis factor-␣ promoter region and in the heat shock protein 70 genes associated with malignant tumors. Cancer 80:1489, 1997. 16. Mira J-P, Cariou A, Grall F, Delcleux C, Losser M-R, Heshmati F, Cheval C, Monchi M, Teboul J-L, Riche F, Leleu G, Arbibe L, Mignon A, Delpech M, Dhainaut J-F: Association of TNF2, a TNF-␣ promoter polymorphism, with septic shock susceptibility and mortality. JAMA 282:561, 1999. 17. McGuire W, Hill AV, Allsopp CE, Greenwood BM, Kwiatkowski D: Variation of the TNF␣ promoter region associated with susceptibility to cerebral malaria. Nature 371:508, 1994. 18. Cabrera M, Shaw MA, Sharples C, Williams H, Castes M, Convit J, Blackwell JM: Polymorphism in tumor necrosis factor genes associated with susceptibility to mucocutaneous leishmaniasis. J Exp Med 182:1259, 1995. 19. Conway DJ, Holland MJ, Bailey RL, Campbell AE, Mahdi OS, Jennings R, Mbena E, Mabey DC: Scarring trachoma is associated with polymorphism in the tumor necrosis factor alpha gene promoter and with elevated TNF-a levels in tear fluid. Infect Immun 65:1003, 1997. 20. Marsh DG, Neely JD, Ghosh B, Freidhoff E, EhrlichKautzky E, Krishnaswamy G, Beaty TH: Linkage analysis of IL-4 and other chromosomes 5q31.1 markers and total serum immunoglobulin E concentration. Science 264:1152, 1994. 21. Meyers DA, Postma DS, Panhuysen CI, Xu J, Amelung PJ, Levitt RC, Bleecker ER: Evidence for a locus regulating total serum IgE levels mapping to chromosome 5. Genomics 23:464, 1994.

Microarray and Microfluidic Cytokine Genotyping

22. Rosenwasser LJ, Klemm DJ, Dresback JK, Inamura H, Mascali JJ, Klinnert M, Borish L: Promoter polymorphisms in the chromosome 5 gene cluster in asthma and atopy. Clin Exp Allergy 25(Suppl):74, 1995. 23. Zhu S, Chan-Yeung M, Becker AB, Dimich-Ward H, Ferguson AC, Manfreda J, Watson WTA, Pare PD, Sandford AJ: Polymorphisms of the IL-4, TNFa, and FcERIB genes and the risk of allergic disorders in at-risk infants. Am J Resp Crit Care Med 161:1655, 2000. 24. Blakemore AIF, Tarlow JK, Cork MJ, Gordon C, Emery P, Duff GW: Interleukin-1 receptor antagonist gene polymorphism as a disease severity factor in systemic lupus erythematosus. Arthritis Rheum 37:1380, 1994. 25. Mansfield JC, Holden H, Tarlow JK, Di Giovine FS, McDowell TL, Wilson AG, Holdsworth CD, Duff GW: Novel genetic association between ulcerative colitis and the antiinflammatory cytokine interleukin-1 receptor antagonist. Gastroenterology 106:637, 1994. 26. Tarlow JK, Clay FE, Cork MJ, Blakemore AIF, McDonagh AJG, Messenger AG, Duff GW: Severity of alopecia areata is associated with a polymorphism in the interleukin-1 receptor antagonist gene. J Invest Dermatol 103: 387, 1994. 27. Blakemore AIF, Cox A, Gonzalez A-M, Maskill JK, Hughes ME, Wilson RM, Ward JD, Duff GW: Interleukin receptor antagonist allele (IL1RN*2) associated with nephropathy in diabetes mellitus. Hum Genet 97:369, 1996. 28. Dean M, Carrington M, Winkler C, Huttley GA, Smith MW, Allikmets R, Goedert JJ, Buchbinder SP, Vittinghoff E, Gomperts E, Donfield S, Vlahov D, Kaslow R, Saah A, Rinaldo C, Detels R, Hemophilia Growth and Development Study, Multicenter AIDS Cohort Study, Multicenter Hemphilia Cohort Study, San Francisco City Cohort Study, ALIVE Study, O’Brien SJ: Genetic restriction of HIV-1 infection and progression to AIDS by a deletion allele of the CKR5 structural gene. Science 273: 1856, 1996. 29. Samson M, Libert F, Doranz BJ, Rucker J, Liesnard C, Farber C-M, Saragosti S, Lapoumerouilie C, Cognaux J, Forceille C, Muyldermans G, Verhofstede C, Burtonboy G, Georges M, Imai T, Rana S, Yi Yanji Smyth RJ, Collman RG, Doms RW, Vassart G, Parmentier M: Re-

997

30.

31.

32. 33.

34.

35.

36.

37.

38.

sistance to HIV-1 infection in caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene. Nature 382:722, 1996. Awad M, Pravica V, Perrey C, El Gamel A, Yonan N, Sinnott PJ, Hutchinson IV: CA repeat allele polymorphism in the first intron of the human interferon-␥ gene is associated with lung allograft fibrosis. Hum Immunol 60:343, 1999. Meenagh A, Williams F, Ross OA, Patterson C, Gorodezky C, Hammond M, Leheny WA, Middleton D: Frequency of cytokine production polymorphisms in populations from Western Europe, Africa, Asia, the Middle East and South America. Hum Immunol 63:1055, 2002. Weinshilboum RM: The genomic revolution and medicine. Mayo Clinic Proc 77:745, 2002. Kwok P-Y: Methods for genotyping single nucleotide polymorphisms. Annu Rev Genomics Hum Genet 2:235, 2001. Dong F, Allawi HT, Anderson T, Neri BP, Lyamichev VI: Secondary structure prediction and structure-specific sequence analysis of single-stranded DNA. Nucleic Acids Res 29:3248, 2001. Santacroce R, Ratti A, Caroli F, Foglieni B, Ferraris A, Cremonesi L, Margaglione M, Seri M, Ravazzolo R, Restagno G, Dallapiccola B, Rappaport E, Pollak ES, Surrey S, Ferrari M, Fortina P: Analysis of clinically relevant single-nucleotide polymorphisms by use of microelectronic array technology. Clin Chem 48:2124, 2002. Kaiser R, Tremblay P-B, Roots I, Brockmoller J: Validity of PCR with emphasis on variable number of tandem repeat analysis. Clin Biochem 35:49, 2002. McCaman MT, Murakami P, Pungor E, Jr, Hahnenberger KM, Hancock WS: Analysis of recombinant adenoviruses using an integrated microfluidic chip-based system. Anal Biochem 291:262, 2001. Lu Y, Nerurkar VR, Dashwood W-M, Woodward CL, Ablan S, Shikuma CM, Grandinetti A, Chang H, Nguyen HT, Wu Z, Yamamura Y, Boto WB, Merriweather A, Kurata T, Detels R, Yanaghira R: Genotype and allele frequency of a 32-bp deletion mutation in the CCR5 gene in various ethnic groups: absence of mutation among Asians and Pacific Islanders. Int J Infect Dis 3:186, 1999.