Multiplex PCR using real time DNA amplification for the rapid detection and quantitation of HTLV I or II

Molecular and Cellular Probes 17 (2003) 59–68 www.elsevier.com/locate/ymcpr

Multiplex PCR using real time DNA amplification for the rapid detection and quantitation of HTLV I or II Michael C. Estes, J. Sanders Sevall* Specialty Laboratories Inc., 2211 Michigan Ave., Santa Monica, CA 90404, USA Accepted for publication 27 November 2002

Abstract A multiplex ‘real-time’ polymerase chain reaction (PCR) has been established as a general technique for the quantitation of proviral human T-lymphotrophic virus types 1 and 2 (HTLV-I/II). The technology utilizes fluorescence to measure amplification products from the tax gene of Human T-cell lymphotropic virus type 1 or the 50 long terminal repeat of Human T-cell lymphotropic virus type 2. The quantitative amplification of the standard was linear across four orders of magnitude with nearly identical amplification efficiencies for monoplex or the biplex format from 1.4 copes/assay (60 copies proviral DNA/0.5 micrograms human DNA) to 6000 copies/assay (240,000 proviral copies/0.5 micrograms human DNA). The human b-globin gene was used to normalize for human DNA input to determine the proviral DNA load. Three hundred fifty-six specimens received by Specialty Laboratories for HTLV I/II detection provided identical results in the detection of HTLV I/II proviral DNA. No additional positive specimens were identified with the biplex assay format. The coefficient of variation for the proviral DNA load was less than 30% for HTLV I or II quantitation (n ¼ 5). For spiked specimens, two groups of five separate 0.25 ml blood specimens (20 total) were spiked, respectively, with 0, 9.6, 48, 240 and 1200 copies of HTLV I or HTLV II DNA standards. The specimens were amplified with the HTLV I/II multiplex format. Twenty of twenty expected negative HTLV I or HTLV II specimens were negative (100% specificity) and 14/16 specimens spiked with 48 copies or more HTLV I were detected (87.5% sensitivity). Thirteen of sixteen HTLV II spiked specimens (.48 copies of HTLV II standard per 10 assays) were detected (81.2%). The real-time detection provides accurate and reliable results in a single amplification for both HTLV (I or II) targets with a more rapid turnaround time and a decrease in material required for results. q 2003 Elsevier Science Ltd. All rights reserved. Keywords: Real-time DNA amplification; HTLV I/II; Biplex polymerase chain reaction

Multiplex polymerase chain reaction (PCR) is a variant of PCR in which two or more target sequences can be amplified by including more than one pair of primers in the same reaction. The result can produce considerable savings of time and effort in the clinical laboratory. DNA amplification is also a valuable tool in the clinical monitoring of target sequence by enumerating the level of target at the start or a particular point in therapy. This is particularly true for viral infections [26]. A very rapid and non-labor intensive amplification technique for the accurate quantitation of a viral target is real-time amplification [24, 25]. An internal fluorescent probe provides specificity and allows the detection of amplification product that is directly * Corresponding author. Tel.: þ 1-310-828-6543; fax: þ1-310-828-6494. E-mail address: [email protected] (J.S. Sevall).

related to the copy number of the target in the amplification reaction. Human T-cell lymphoma/leukemia viruses types I and II (HTLV I and HTLV II) belong to an oncogenic genus of retroviruses which are transmitted perinatally, sexually and by blood transfusion or the sharing of needles during intravenous drug abuse [1]. The HTLV group of oncogenic viruses does not carry a known oncogene in their proviral genome and integrate randomly into the host cellular DNA. Therefore, a different strategy to induce neoplastic transformations (not yet fully defined) is used. Both viruses infect all subsets of human lymphocytes causing polyclonal, oligoclonal and monoclonal T-cell lymphocytoses with HTLV I being skewed toward the CD42 lymphocytoses and HTLV II being skewed toward the CD8þ lymphocytoses [2]. HTLV I is

0890-8508/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0890-8508(03)00002-1

60

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

the etiologic agent of CD4þ adult T-cell lymphoma/ leukemia (ATLL) which presents with varying prognoses and aggressiveness but eventually evolves into a lethal disease [1]. HTLV I has also been associated with various inflammatory diseases such as HTLV I associated myelopathy (HAM)/tropical spastic paraparesis (TSP) [3,4], HTLV I uveitis (HU) [5], infective dermatitis [6] and arthropathy [7]. HTLV II is associated with very rare cases of CD8þ neoplasia, particularly large granular lymphocytic (LGL) leukemia and cutaneous T-cell lymphoma [8,9]. The determinants of the outcome of HTLV-I, and possibly HTLV-II, associated diseases are not fully defined; however, HTLV viral load could play a major role in pathogenesis. Increased proviral load has been associated with patients with HAM/TSP or HU [10 – 12], indicating its role in the pathogenesis of these diseases. The relationship of proviral viral load and the pathogenesis of HTLV-I associated myelopathy/tropical spastic paraparesis has been confirmed using ‘real time’ quantitative PCR [13 – 17]. In the present study, a multiplexed, real-time quantitative assay is developed to detect and quantify both HTLV I and/or HTLV II [18]. The result can produce considerable savings of time and effort in the clinical laboratory.

1. Method and materials 1.1. Clinical specimens and cell lines Whole blood (ACD) (negative control DNA) was collected from 144 healthy employee donors at Specialty Laboratories with no clinical history with respect to HTLV I/II DNA infection. For standards, HTLV type II DNA or I were extracted from virus infected cells. DNA standards were prepared by serially diluting the stock of virus infected cellular DNA at 50 ng/ml. HTLV I was strain MT-2 (Advanced Biotechnologies, Columbia, MD 21046; Cat # 08-711000) and HTLV II was strain C3-44 (Mo) (Catalogue # 08-710-000). Assuming a single copy of the proviral genome is present per human genome and 3.17 £ 109 bp (1.99 £ 1012 g/mol) per human genome, 50 ng (1 ml of the stock virus infected cellular DNA) is equivalent to 15,150 viral genomes. For human DNA calibration, purified human DNA was extracted from a healthy donor and the copy number determined for the haploid human genome (3.17 £ 109 bp and 1.99 £ 1012 g/mol per human genome). One microgram human DNA was calculated to have 303,000 copies and the calibration curve was diluted to 1.9, 9.6, 48, 241, 1208, 6042 and 30,000 copies/assay. The linear regression parameters were 2 3.47 and an intercept of 38.4 for the average of ten individual runs. The amplification efficiency was 94.7%.

For clinical specimens, DNA was extracted from three hundred fifty-six remnant specimens received at Specialty Laboratories, Inc for the current HTLV I/II DNA DetectRe (Test Code 9896). The current HTLV I/II DNA test uses a DNA amplification with a non-radioactive-microtiter product detection procedure. The 356 specimens consisted of two hundred sixty-nine whole bloods, seventy-six cerebral spinal fluids (CSF), and eleven miscellaneous tissues or body fluids.

1.2. Sample preparation DNA was extracted from all specimens using a modified guanidinium thiocyanate organic extraction procedure [19]. In a 1.6 ml microcentrifuge tube, onequarter ml (0.25 ml) of the specimen was suspended in 0.8 ml of 50% v/v of phenol (buffered at pH 6.4) and guanidinium lysis buffer (6 M guanidinium thiocycnate (GuSCN), 100 mM dithiothreithol, 1 mM ethylenediaminetetraacedic acid (EDTA), 50 mM Tris(HCI) pH 8.0, 50 mg/ml glycogen, 0.5% (w/v) sodium lauryl sarcosine) and vortexed vigorously for 30 s (the specimen can be incubated at 60 8C from 15 min to 1 h if necessary to solubilized the suspension). For tissue specimens, 100– 200 mg (wet weight) was solubilized in 0.2 ml 0.1% sodium dodecylsulfate with 100 mg proteinase K at 60 8C over night prior to the guanidium-organic extraction. After solubilization, 200 ml of chloroform is added and vortex to homogeneity (30 s to 1 min at room temperature). The aqueous phase is separated from the organic phase with centrifugation and the aqueous phase collected to a sterile microfuge tube. Nucleic acids were collected by ethanol precipitation.

1.3. Primers and probes Primers were designed using the software Primer Express (Applied Biosystems, Foster City, CA, USA). The tax gene for HTLV I (Accession number: AF033819) with a forward primer HTLV I F 50 -aag act gtt tgc cca cca cc-30 , reverse primer HTLV I R 50 -ttg cca ggc tgt tag cgt g-30 and probe HTLV I Taqman probe 50 Fam-ttt cca gcc tgt tag ggc acc cg-BHQ1-30 and the long terminal repeat sequence for the HTLV II (Accession number: AF306733) HTLV II F 50 -cgc aag gac agt tca gga gg-30 ; the reverse primers HTLV II R 50 -atc ccc aag gtg agt ctc cg-30 and the Taqman HTLV II probe 50 Joectc tcg ctc cct cac cga ccc tct-BHQ1-30 . For the human globin gene (AF083884) target, the primers were B-gloF 50 -gca aga aag tgc tcg gtg c-30 with B-gloR 50 -cta ctc agt gtg gca aag gtg-30 with the probe of 50 -Fam-tag tga tgg cct ggc tca cct gga c-BHQ1-30 . The specificity of the targets were initially characterized by screening the target sequence in the nucleic acid database where there was no

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

significant cross reactions to sequences other than the primary target.

1.4. PCR amplification and HTLV I/II proviral load quantitation The PCR reaction mix (total volume of 50 ml) consisted of DNA from approximately 200,000 cells (approximately 1 mg) with 25 ml universal master mix (Applied Biosystems P/N 4304437), 12.5 pmol of the forward and reverse primer and 2.5 pmol of the Taqman probe. If the amplification is multiplexed 12.5 pmol of the forward and reverse primers for all targets (two individual targets) that are to be amplified are added and 2.5 pmol of each Taqman probe for each amplification target is added. The TaqMan Universal Master Mix is a premix of all components, except primers and probe, necessary to perform the amplification including the AmpEreasew decontamination system which is uracil-Nglycosylase [20]. For both the HTLV-I/II and the globin amplifications, after one cycle at 50 8C for 2 min and one cycle at 95 8C for 10 min, a two-step PCR procedure was used consisting of 15 s at 95 8C and 60 8C for 30 s for 40 cycles. Amplification and data acquisition were carried out using the ABI Prism 7700 Sequence Detector System (Perkin – Elmer Applied Biosystems). All standard dilutions, controls and patient samples were run in duplicate and the average value of the copy number was used to quantify HTLV I, HTLV II and globin DNA. Standard curves for HTLV I, II and globin DNA were accepted when the slopes were between 2 3.74 and 2 3.32 (corresponding to PCR efficiencies between 85 and 100% [21]). Both HTLV type I or II DNA standards were extracted from virus infected cells. The standard curve (calibration standard) is prepared by diluting the stock of virus infected cellular DNA which is obtained at 50 ng/ml. HTLV I was strain MT-2 (Advanced Biotechnologies Cat # 08-711-000) and HTLV II was strain C3-44 (Mo) (Catalogue # 08-710-000). Assuming there is one proviral genome per cellular genome, 50 ng

(1 ml of the stock virus infected cellular DNA) is equivalent to 15,150 human genomes. The stock of proviral DNA was diluted 20 ml into 0.5 ml sterile Tris – EDTA pH (8.0).

61

Dilution of the HTLV I or II stock in sterile Tris –EDTA buffer Standard

Dilution factor

Copy number/10 ml

1 2 3 4 5 6 7

TE 1:5 1:5 1:5 1:5 1:5 1:5

6000 molecules 1208 molecules 241 molecules 48 molecules 9.6 molecules 1.9 molecules 0.4 molecules

In these studies, all standard dilutions, controls and patient samples were run in duplicate. The average of the copy number was used to determine both HTLV I/II or globin copy number levels. The proviral load was calculated as the ratio of HTLV copy number/globin copy number £ 2 £ 106. The HTLV level is expressed as the number of HTLV copies/1 million blood lymphocytes. (2 £ 106 copies human genome).

2. Results 2.1. Specificity and sensitivity of the multiplexed amplification Primers and probe were initially selected from the Human T-cell Lymphotropic virus type 1 (HTLV I) tax gene (Accession number: AF033819) and Human Lymphotropic virus type 2 (HTLV 2) 50 -long terminal repeat (Accession number: AF306733) by selection of optimum amplification primers and probes for real-time amplification with the Primer Express software (Applied Biosystems, Foster City). HTLV I target sequence is: LOCUS AF485381 1038 bp DNA linear VRL 24-MAR2002 DEFINITION Human T-cell lymphotropic virus type 1 from Spain tax gene, partial ACCESSION AF485381

The HTLV II proviral target sequence is: DEFINITION Human T-cell lymphotropic virus type 2 KAA8883 50 long terminal repeat, partial sequence.

62

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

ACCESSION AF306733

Fig. 1. Amplification profile of HTLV I standard with the HTLV I (fam) probe (A) or the HTLV II (joe) probe (B). The HTLV I proviral DNA standard at 0.4, 1.9, 9.6, 48, 240, 1200 and 6000 copies per assay were amplified in a monoplex format along with 30 water blanks (no template controls, NTC) under conditions described in Section 1. Amplification was detected when the arbitrary fluorescence units (AFU) increase above the threshold fluorescence baseline. The maximum fluorescence observed for the HTLV I probe was greater than 1.0 for standard levels above 48 copies/assay and 0.2 AFU for the 1.9 copies per assay. The maximum fluorescence observed for the HTLV II standard with the HTLV I (fam) probe did not rise above 0.005 AFU.

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

A blast search of the GeneBank with each target sequence indicates that both target sequences were highly selective for their respective virus with no cross selection between the two HTLV viruses. The specificity was confirmed with 144 HTLV seronegative, whole blood specimens taken from ‘healthy adults’ with no history of HTLV II or I infection. The specimens were extracted as described in Section 1.2. Human b-globin amplification

63

confirmed amplifiable DNA was obtained with an average of 157,000 copies of the human genome present in the amplification assay (< 0.5 mg human DNA). (Assuming a genome size of 3.17 £ 109 base pairs and a molecular weight of 1.99 £ 1012 g/mol). Multiplexing HTLV I and II amplifications in the same reaction, Fig. 1(A) and (B) and Fig. 2(A) and (B) show the average (n ¼ 4) amplification profile of the HTLV proviral

Fig. 2. Amplification profile of HTLV II standard with the HTLV II (Joe) probe (A) or the HTLV I (fam) probe (B). The HTLV II proviral DNA standard at 0.4, 1.9, 9.6, 48, 240, 1200 and 6000 copies per assay were amplified in a monoplex format along with 30 water blanks (no template controls, NTC) under conditions described in Section 1. Amplification was detected when the arbitrary fluorescence units (AFU) increase above the threshold fluorescence baseline. The maximum fluorescence observed for the HTLV II probe was greater than 0.1 AFU for standard levels above 9.6 copies/assay and 0.02 AFU for the 1.9 copies per assay. The maximum fluorescence observed for the HTLV I standard with the HTLV II (joe) probe did not rise above 0.002 AFU.

64

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

DNA standard and water blanks (n ¼ 30) with its own probe, A, and the alternate viral probe, B. The individual probes do not cross react with alternate virus. Both HTLV probes are very stable across 40 thermal cycles. The standard deviation of the fluorescence over thermal cycles 3– 15 for all individual amplifications in the multiplexed format for HTLV I or HTLV II is 0.0007 and 0.0004, respectfully. To avoid a very low discrimination baseline, the threshold baseline is set at 100 times the standard deviation of the observed fluorescence between thermal cycles 3 and 15. The baseline fluorescence for threshold thermal cycle was 0.07 and 0.04 respectfully, to discriminate DNA amplification from inherent fluorescence variation. The point that the fluorescence exceeds the determined baseline is defined as the threshold thermal cycle (Ct). The threshold thermal cycles were determined for seven, five-dilutions of the HTLV provirus standard with the multiplex and monoplex format. Table 1 shows the interrun variation for the HTLV I standard curve run fifteen individual times in a multiplexed format. Table 2 tabulates the HTLV I standard amplified ten individual times in a monoplexed format where only HTLV I primers and probe are in the amplification reaction. The correlation coefficient (r 2) between the log copy number of the viral target and observed threshold thermal cycle (Ct) is 0.98 – 0.99. The slopes of the HTLV I amplification are 2 3.46 for the monoplexed format and 2 3.25 for the biplex format. The slopes of the standard curve reflect the relationship of copy number and amplification efficiency. The two slopes being within 90% of each other imply either format has the same relative amplification efficiency and multiplexing does not appear to effect affect amplification efficiency [21]. The average coefficient of variation is 4.23% for the HTLV I monoplexed assay and 3.58% for the multiplexed assay. The standard curve is very stable and reproducible across a reasonable length of time (one month) making it suitable for use as a calibrator for viral amplification. The lower level of reproducible detection is two copies of target for both monoplexed and biplexed assay. (80 copies HTLV I proviral DNA per ml specimen). Table 2A and B tabulates the HTLV II standard in both the multiplexed (A) and monoplexed (B) format. The analysis reflect the same general characteristics as seen with HTLV I. The correlation coefficient (r 2) for the linear regression analysis for both HTLV II formats is 0.98 – 0.99 and the slope is 2 3.22 and 2 3.32. Monoplexed or biplexed formats does not affect the relative efficiency of amplification. The coefficient of variation is the same for either format of 2.64 and 2.71% with the lower level again at two copies/assay (80 copies/ml of blood), The biplexed and monoplexed amplification provides the same performance of detection and quantitation with an over all reduction in total number individual reactions that must be run (decrease in over head) (Fig. 3).

Table 1 HTLV I standard with monoplex and biplex format HTLV I (copies/assay) 0.4 1.9 9.6 48.0 240 1200 6000 lnterceptb Slopec PCR eff (%)d

Mono Ct (n ¼ 10)a 39.14 37.92 34.29 31.96 30.13 27.91 25.48

SD

% CV

1.53 2.12 2.53 1.11 1.02 0.80 0.91

3.91 5.60 7.39 3.47 3.40 2.86 3.59

38.74 23.6 89.6

Biplex Ct (n ¼ 15)a 39.22 37.09 34.88 32.67 30.47 28.27 26.07

SD

%CV

0.75 2.51 1.56 1.20 0.91 0.69 0.76

3.49 3.33 3.62 2.74 1.18 2.31 1.84

38.24 23.25 103.1

Inter-run variation for the HTLV I standard. HTLV 1 standard curve was run fifteen individual times in individual runs in a multiplexed format as described in Section 1. The threshold thermal cycle (Ct) is reported for each copy number level in the assay (0,4, 1.9, 9.6, 48, 240,1200 and 6000 copies proviral DNA/assay). a Amplification parameters for HTLV I proviral DNA standard (Advanced Biotechnology Inc) as described in Section 1. The average threshold thermal cycle (Ct) was obtained from 10 individual runs in the monoplex format and 15 individual runs in the biplex format of the calibration curve. b The linear regression parameter (intercept) of the calibration curve with the log of the copy number per assay. c The linear regression parameter (slope) of the calibration curve with the log of the copy number per assay. d % amplification efficiency as (10(2 1/slope) 2 1) £ 100 [21].

Table 2 HTLV II standard with monoplex and biplex format HTLV II (copies/assay) 0.4 1.9 9.6 48.0 240 1200 6000 lnterceptb Slopec PCR eff (%)d

Mono Ct (n ¼ 10)a

SD

% CV

37.75 37.31 34.98 32.66 30.34 28.02 25.70

1.70 0.74 1.54 0.42 0.69 0.73 0.50

4.50 2.03 4.31 1.28 2.28 2.62 1.96

38.24 23.32 100.1

Biplex Ct (n ¼ 7)a 39.08 38.87 36.39 33.65 31.22 28.84 26.54

SD

%CV

1.36 1.29 1.32 0.92 0.37 0.67 0.49

3.49 3.33 3.62 2.74 1.18 2.31 1.84

38.93 23.22 104.4

Inter-run variation for HTLV I in a monoplexed format. The HTLV I standard amplified ten individual times in individual runs with a monoplexed format as described in Section 1. The threshold thermal cycle is reported for each copy level of HTLV I proviral genome (0,4, 1.9, 9.6, 48, 240, 1200 and 6000 copies proviral DNA/assay). a Amplification parameters for HTLV II proviral DNA standard (Advanced Biotechnology Inc) as described in Section 1. The average threshold thermal cycle (Ct) was obtained from 10 individual runs in the monoplex format and seven individual runs in the biplex format of the calibration curve. b The linear regression parameter (intercept) of the calibration curve vs the log of the copy number per assay. c The linear regression parameter (slope) of the calibration curve vs the log of the copy number per assay. d % amplification efficiency as (10(2 1/slope) 2 1) £ 100 [21].

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

2.2. Performance with spiked specimens For the infected whole blood specimens, Two groups of five separate 0.25 ml blood specimens (20 total) were spiked, respectively, with 0, 9.6, 48, 240 and 1200 copies of

65

HTLV I or HTLV II DNA. The spiked specimens were then extracted by the normal protocol (see Section 1) and prepared for 10 individual amplifications. The specimens were then amplified four individual times (individual runs) by the multiplex DNA amplification procedure. Twenty

Fig. 3. Linear regression of HTLV standard. The HTLV I and II proviral DNA standard at 0.4, 1.9, 9.6, 48, 240, 1200 and 6000 copies per assay were amplified in a monoplex (A), HTLV I in the biplex (B) and HTLV II in the biplex (C) format under conditions described in Section 1. Amplification was detected and the threshold thermal cycle number (Ct) was collected and analyzed by a linear regression analysis. The root mean square (RMS) and slope of the standard curve are reported for each standard.

66

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

Fig. 3 (continued )

expected HTLV I negative specimens were not detected (100% specificity) and 10/12 specimens spiked with greater than 48 copies of HTLV I proviral standard per 10 assays were detected (83%) (Specimens spiked with 9.6 copies of proviral DNA would be expected to have viral loads below detectable levels—0.9 copies per assay). For HTLV II spiked specimens, 20 of 20 expected negative specimens were not detected for HTLV II (100% specificity) and 11/12 specimens that were spiked and extracted with 48 HTLV II proviral copies/10 extracted assays were detected (91.7% sensitivity). 2.3. HTLV proviral load in clinical specimens Three hundred and fifty-six specimens received by Specialty Laboratories for HTLV I/II DNA DetectR (Test Code 9896) by DNA amplification using a non-radioactive amplification product detection method consisting of two hundred-sixty-nine whole bloods, seventy-six CSF, and eleven miscellaneous tissues or body fluids were run with the multiplexed HTLV I/II ‘real-time’ procedure. The three positive HTLV specimens (one HTLV I and 2 HTLV II specimens) were detected and quantitated by both the monoplex and biplex HTLV I/II amplification formats. There were no additional specimens detected by the multiplexed HTLV amplification. The real-time detection corroborates the initial amplification results in a single amplification for both HTLV (I or II) targets without requiring post-amplification processing. The proviral load of HTLV for the positive specimens is reported as copies HTLV per 106 cells to control differences in DNA input and normalize possible differences of amplification efficiencies. The amount of human DNA in the HTLV I/II assay was determined by measuring the human b-globin gene. As for

the HTLV I/II calibration curves, the human calibration curve had regression parameters of 2 3.47 slope and 38.3 intercept and an efficiency of amplification of 95% (n ¼ 10). The values determined for the positive HTLV I specimen was 27.9 copies HTLV I per million lymphocytes (n ¼ 5; CV ¼ 35.4%) by the monoplexed format and 57.3 copies HTLV I per million lymphocytes (n ¼ 5; CV ¼ 213%) for the biplex assay format. The HTLV II specimens (two separate specimens) showed a much higher level of proviral copies per million lymphocytes. Specimen #1 had a mean HTLV II level of 954.3 copies per two million lymphocytes (n ¼ 5; 17.4%) for the monoplexed format and 1986.4 copies/million cells (n ¼ 5; CV ¼ 28.1%) for the biplexed assay. Specimen #2 had a mean copy number of 5576 copies HTLV II per 2 million lymphocytes (n ¼ 5; CV ¼ 23.3%) with the monoplexed assay and 11953 copies per million lymphocytes (n ¼ 5; CV ¼ 27.6%) for the biplex assay. Performance of both quantitative formats were similar in inter run accuracy reproducibility.

3. Discussion We have demonstrated the efficiency of a multiplexed proviral HTLV I/II DNA real-time quantitative amplification for determination of proviral DNA load in 356 clinical specimens received for clinical determination of viral infection. The procedure described allows either the HTLV I or II proviral quantitation in a single tube in biplex format. The assays allows one to screen specimens for both viruses in the same reaction decreasing the number of individual tests that must be run to detect which virus may be present and then quantitative the detected viral type.

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

Quantitation is determined by comparing the sample Ct to a five-fold dilution of a defined standard (HTLV I or II quantitated standard obtained from Advanced Biotechnologies Inc, Columbia, Maryland 21046) [24,25]. Comparison of the kinetics and efficiencies of amplification of the individual amplification (monoplexed format) and multiplexed amplification are similar yielding results that are basically identical [21]. Multiplexed quantitation of viral load by real-time DNA amplification allows the detection and quantitation of either HTLV I or II. The main advantage for this assay is requests for HTLV I or II DNA can be run with technology that reduces possible cross contamination (the amplification tubes are never opened) and allows all specimens to be amplified in the same amplification run reducing turn around time. The technology used allows reliable quantitation without post-amplification processing which makes it very suitable for clinical laboratory applications with a high throughput. For HTLV type 1, the gene target used is the tax gene, which has had some controversy in its presence in normal blood donors [22,23] as well as in HTLVseronegative myeloneuropathies [27]. However, screening 144 healthy donors at Specialty Laboratory did not confirm these findings but it must be understood should this test is used to screen for HTLV I proviral DNA, the tax gene target may have to be change to a target site more adequate for a broad spectrum of replicative and potentially infectious HTLV I/II strains. The biplex format for quantitation is a viable procedure to rapidly quantitate multiple infectious organisms and the purpose of this manuscript is to demonstrate the accuracy and reproducibility of the biplex format. The measurement of the human b-globin gene provides an amplification control and an accurate measure of cellular DNA. These values allow proviral DNA to be normalized to differences in DNA input in the amplification and/or the presence of PCR inhibitors. Reporting the results as viral copies per microgram DNA or cell number (2 copies of human DNA/cell) should be more value clinically in monitoring the long term levels of viral load. The usefulness of proviral HTLV I/II load has become an integral part of the clinical management of patients infected with viral organisms. Besides providing prognostic information on individual, the proviral load can be used to monitor response to therapeutic treatment [28]. Using a biplex format in conjunctions with human DNA quantitation provides a more efficient clinical procedure that is highly accurate with a very rapid turn around time.

References [1] Poiesz BJ. Etiology of acute leukemia, molecular genetics and viral oncology. In: Wienik CPH, Camelos GP, Dutcher JP, Kyle RA, editors. Neoplastic diseases of the blood. New York: Churchill Livingstone; 1995. p. 159.

67

[2] Poiesz BJ, Papsidero LD, Ehrlich G, et al. Prevalence of HTLV I associated T-cell Lymphoma. Am J Hematol 2001;66:32–8. [3] Gessian A, Barin F, Vernant JC, Gout O, Maurs L, Calender A, de Tge G. Antibodies to human T-cell lymphotropic virus type I in patients with tropical spastic paraparesis. Lancet 1985;2:407–10. [4] Osame M, Usuku K, Izumo S, Ijichi N, Amitani H, Igata A, Matsumoto M, Tara M. HTLV I associated myelopathy, a new clinical entity. Lancet 1986;1:1031– 2. [5] Mochizuki M, Watanabe T, Yamaguchi K, Takatsuki K, Yoshimura K, Shirao M, Nakashima S, Mon K, Araki S, Miyata N. HTLV I uvetis: a distinct clinical entity caused by HTLV I. Jpn J Cancer Res 1992;83:236 –9. [6] LaGrenade L, Hanchard B, Fletcher V, Cranston B, Blattner W. Infective dermatitis of Jamaican children: marker for HTLV I infection. Lancet 1990;336:1345–7. [7] Nishioka K, Maruyama I, Sato K, Kitajima I, Nakajima Y, Osame M. Chronic inflammatory arthropathy associated with HTLV-I. Lancet 1989;1:441. [8] Loughran TP, Coyle T, Sherman MP, et al. Detection of human T-cell leukemia/lumphoma virus type II in a patient with LGL leukemia. Blood 1992;80:300–9. [9] Poiesz B, Dube D, Dube S, et al. HTLV-II associated CD8 þ cutaneous T-cell lymphoma in an HIV-1 infected patient. N Engl J Med 2000;342:930–6. [10] Yoshida M, et al. Increased replication of HTLV-I in HTLV-I associated myelopathy. Ann Neurol 1989;26:331. [11] Gessian A, Saaal F, Gout O. High human T-cell lymphotropic virus type I proviral DNA load with polyclonal integration in peripheral blood mononuclear cells of French West Indies, Guianese, and African patients with tropical spastic paraparesis. Blood 1990;15: 428–33. [12] Ono A, Ikeda E, Mochizuki M, Matsuokas M, Yamaguchi K, Sawada T, Yamane S, Tokudome S, Watanabe T. Provirus load in patients with human T-cell leukemia virus type 1 uveitis correlates with precedent Graves’ disease and disease activities. Jpn J Cancer Res 1998;89:608 –14. [13] Nagai M, Usuku K, Matsumoto W, Kodama D, Takenouchi N, Moritoyo T, Hashiguchi S, Ichinose M, Bangham CR, Izumo S, Osame M. Analysis of HTLV-I proviral load in 202 HAM/TSP patients and 243 asymptomatic HTLV-I carriers,: high proviral load strongly predisposes to HAM/TSP. J Neurovirol 1998;4: 586–93. [14] Manns A, Miley WJ, Wilks RJ, Morgan OS, Hanchard B, Wharfe G, Cranston B, Maloney E, Wells SL, Blattner WA, Waters D. Quantitative proviral DNA and antibody levels in the natural history of HTLV-l infection. J Infect Dis 1999;180:1487– 93. [15] Miley WJ, Suryanarayana K, Manns A, Kubota R, Jacobson S, Lifson JD, Waters D. Real-time polymerase chain reaction assay for cellassociated HTLV type I DNA viral load. AIDS Res Hum Retroviruses 2000;16:665 –75. [16] Etoh E, Yamaguchi K, Tokudome S, Watanabe T, Okayama T, Okayama A, Stuver S, Mueller N, Takatsuki K, Matsuoka M. Rapid quantification of HTLV I provirus load: detection of monoclonal proliferation of HTLVB-I infected cells among blood donors. Int J Cancer 1999;81:859 –64. [17] Dehee A, Cesaire Desire N, Lezin A, Bourdonne O, Bera O, Plumella Y, Smadja D, Nicolas JC. Quantitation of HTLV I proviral load by a TaqMan real-time PVR assay. J Virol Methods 2002;102:37 –51. [18] Markoulatos P, Siafakas N, Moncany M. Multiplex polymerase chain reaction: a practical approach. J Clin Lab Anal 2002;16:47 –51. [19] Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate phenol cholorform extraction. Anal Biochem 1987;162:156–9. [20] Longo MC, Berninger MS, Hartley JL. Uracil DNA glycosylase to control carry-over contamination in polymerase chain reactions. Gene 1990;93:125 –8.

68

M.C. Estes, J.S. Sevall / Molecular and Cellular Probes 17 (2003) 59–68

[21] Meijerink J, Mandigers C, van de Locht L, To¨nnissen E, Goodsaid F, Raemaekers J. A novel method to compensate for different amplication efficiencies between patient DNA samples in quantitative real-time PCR. J Mol Diag 2001;3:55–61. [22] Zucker-Franklin D, Pancake BA. Human T-cell lymphotrophic virus type I tax among American blood donors. Clin Diagn Lab Immunol 1998;5:831 –5. [23] Cowan EP, Nemo GJ, Williams AE, Alexander RK, Vallejo A, Hewlett IK, Kak RB, Dezzutti CS, Gallahan K, Pancake BA, ZuckerFranklin D, McCurdy PR, Tabor E. Absence of human Tlymphotropic virus Type I tax sequences in a population of normal blood donors in the Baltimore, MD/Washington DC area: results from a multicenter study. Transfusion 1999;39:904 –9.

[24] Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative PCR. Genome Res 1996;6:986–94. [25] Bustin SA. Absolute quantitfication of mRNA using real-time reverse transcriptase polymerase chain reaction assay. J Mol Endocrinal 2000; 25:169–93. [26] Berger A, Preiser W. Viral genome quantification as a tool for improving management: the example of HIV, HBV, HCV and CMV. J Antimicrob Chemother 2002;49:713–21. [27] Galeno H, Ramirez E, Cartier L. HTLV-I provirus in sernegatve Chilean patients with tropical spastic paraparesis. Lancet 1996;2:1170. [28] Berger A, Preeiser W. Viral genome quantification as a tool for improving patient management: the example of HIV, HBV, HCV, and CMV. J Antimicrob Chemo 2002;49:713–21.