| Nucleic Acids Research | Pages |
Simultaneous detection of several oligonucleotides by time-resolved fluorometry: the use of a mixture of categorized microparticles in a sandwich type mixed-phase hybridization assay
Introduction
Materials And Methods
Categorized oligonucleotide-coated particles
Oligonucleotides and their fluorescent conjugates
Hybridization assays
Results
Categorization of the particles with prompt fluorophores
Hybridization
Discussion
Acknowledgements
References
Simultaneous detection of several oligonucleotides by time-resolved fluorometry: the use of a mixture of categorized microparticles in a sandwich type mixed-phase hybridization assay
ABSTRACT
INTRODUCTION
Mixed-phase hybridization assays based on immobilized oligonucleotide probes have received increasing popularity in pharmaceutical research, molecular genetics, medicine and diagnostics (1-4). Arrays consisting of oligonucleotides covalently bound to a solid surface, the so-called chips, are usually exploited in such assays (5-15). Microparticles have also been used for similar purposes (16-20), but thus far only in a batch-wise manner, which does not enable multiparametric measurements, i.e. simultaneous detection of more than one sequence from a single sample. In clinical diagnostics multiparametric measurements are, however, highly desirable, since a genetic disease can be caused by hundreds of different mutations. For example, over 500 mutations in the CFTR gene have been reported (21).
We have recently tried to develop a single-particle approach that allows a multiparametric mixed-phase hybridization assay (22-25). According to this format, a mixture of microparticles, each of which bears a given allele-specific probe and a reporter group(s) defining the particle category, is used as the solid phase in the assay. After hybridization, individual particles are subjected one by one to two parallel measurements, the first one identifying the particle category and the second quantifying the hybridization of the fluorescently tagged probe. While organic prompt fluorophores are used to create the particle categories, oligonucleotide probes labeled with photoluminescent lanthanide chelates are exploited in quantification of the hybridization. The latter markers exhibit some obvious advantages over conventional organic fluorophores (26-28): (i) the fluorescence lifetime of the order of 1 ms enables the application of time-resolved mode in the measurements; (ii) the Stokes' shift, i.e. the difference between the wavelength of emission and excitation, is large; (iii) the emission band is narrow; and (iv) the concentration quenching is negligible. Accordingly, the prompt background fluorescence emission may efficiently be eliminated, and the emission intensity depends linearly on concentration over an exceptionally wide range. Furthermore, different lanthanide ions (Eu3+, Tb3+, Sm3+ and Dy3+) may be utilized in a single assay, since their emission maxima differ considerably from each other.
We have previously described several methods for the covalent immobilization of oligonucleotides to polymeric particles (23,24). Somewhat surprisingly, post-synthetic immobilization of purified oligonucleotides did not provide any advantage over in situ chain assembly on the particles. Accordingly, the latter approach was used for quantification and optimization of a sandwich type hybridization assay illustrated in Figure
Figure 1. The principle of sandwich type hybridization assay. The particles employed were porous uniformly sized ([empty] = 50 µm) beads of SINTEF (Trondheim, Norway), made of a copolymer of glycidyl methacrylate (40%) and ethylene dimethacrylate (60%) (37% matrix) and derivatized with bis(3-aminopropyl)amine, yielding a 1 mmol/g density of primary amino functions. From the point of view of possible automation in future, the uniform size of particles is of importance. They were further derivatized for oligonucleotide synthesis with pyridinium (4,4[prime]-dimethoxytrityloxy)acetate and capped thoroughly with acetyl groups as described previously (24). Sequences 1-6 were assembled on these particles by normal phosphoramidite chemistry. According to dimethoxytrityl cation assay (29), the loading of the full-length sequences ranged from 15 to 18 µmol/g. The particle categories were defined by using a desired amount of either a fluorescein or dansyl tagged building block in one of the early steps of thechain assembly. The fluorescein tagged building block was a commercially available 1-(4,4[prime]-dimethoxytrityloxy)-2-[N-thiourea-(di-O-pivaloylfluorescein)-4-amino-butyl)propyl-3-O-[(2-cyano-ethyl)-N,N-
Table 1. 
MATERIALS AND METHODS
Categorized oligonucleotide-coated particles
diisopropylphosphoramidite] (7) and the dansyl tagged one a N4-(N-dansyl-6-aminohexyl)-2[prime]-deoxy-5[prime]-O-(4,4[prime]-dimethoxytrityl)cytidine 3[prime]-[O-(2-cyanoethyl)-N,N-diisopropyl]phosphoramidite (8) prepared in the laboratory as described below. The exact positions of these labels within the assembled chains are indicated in Table 1. The intensity of the fluorescence emission of these fluorophores was adjusted to a desired value by diluting 7 and 8 in the coupling mixture with thymidine phosphoramidite, keeping the total concentration constant (0.1 mol/l). The suitable dilutions were observed to be 1:3 and 1:10 for 8 ([8]/[T]), and 1:20 and 1:75 for 7 ([7]/[T]). When these dilutions were employed, the categories `high' and `low', and their combinations, as indicated in Table 2, could be unequivocally distinguished.
Compound
Sequence
1a-f
5[prime]-d(TATCATCTTTGGTTXYT)-3[prime]-P
1a: X = T; Y = none
b: X = 1:10 mixture of 8:T; Y = none
c: X = 1:3 mixture of 8:T; Y = none
d: X = 1:75 mixture of 7:T; Y = none
e: X = 1:10 mixture of 8:T; Y = 1:75 mixture of 7:T
f: X = 1:20 mixture of 7:T; Y = none
2
5[prime]-d(TATCATTGGTGTTCXT)-3[prime]-P
X = 1:10 mixture of 8:T
3
5[prime]-d(TTCTTGGAGAAGTXT)-3[prime]-P
X = 1:3 mixture of 8:T
4
5[prime]-d(ttctttgagaagtXT)-3[prime]-P
X = 1: 75 mixture of 7:T
5
5[prime]-d(TTCTTTGTGGCXYT)-3[prime]-P
X = 1:10 mixture of 8:T; Y = 1:75 mixture of 7:T
6
5[prime]-d(TTCTTGTGGCXT)-3[prime]-P
X = 1:20 mixture of 7:T
Table 2. . The particle categories employeda
| Category | Dansyl | Fluorescein |
| 1 | - | - |
| 2 | low | - |
| 3 | high | - |
| 4 | - | low |
| 5 | low | low |
| 6 | - | high |
The dansylated nucleoside (1 mmol; 0.84 g) was dried by coevaporation with MeCN (3 × 10 ml), and dissolved together with 2-cyanoethyl N,N,N[prime],N[prime]-tetraisopropylphosphordiamidite (0.403 ml) in dry MeCN (2 ml). Dry 1H-tetrazole (0.45 mol/l in MeCN, 2.17 ml) was added, and the reaction mixture was left to stand at ambient temperature. According to TLC (CH2Cl2, 3.4% MeOH, 1.7% Et3N), the reaction was completed in 1 h. Saturated aq. NaHCO3 (100 ml) was used to stop the reaction. The product was extracted with ethyl acetate (2 × 70 ml), dried over Na2SO4 and evaporated to dryness. The yield of 8 was nearly quantitative. The compound was used without further purification in the solid phase oligonucleotide synthesis. Rf 0.95 (CH2Cl2, 3.4% MeOH, 1.7% Et3N). 32P NMR (CDCl3): 149.16, 148.69.
Oligonucleotides and their fluorescent conjugates
The unlabeled oligodeoxyribonucleotides (9-14 in Table 3) used as targets were synthesized by the conventional phosphoramidite chemistry. The fluorescently labeled oligonucleotide probe (15), 5[prime]-(X5TCATGAGTCAAGTCTA), where X stands for N4-(6-aminohexyl)-2[prime]-deoxycytidine tethered to a photoluminescent europium(III) chelate, {2,2[prime],2[prime][prime],2[prime][prime][prime]-{{4[prime]-{4[prime][prime][prime]-[(4,6-dichloro-1,3,5,-tria-
zin-2-yl)-amino]phenyl}-2,2[prime]:6[prime],2[prime][prime]-terpyridine-6,6[prime][prime]-diyl}bis-(me-
thylenenitrilo)}tetrakis(acetato)}europium(III) (31), was used as a fluorescent probe in the sandwich type hybridization assays. This conjugate was prepared as described previously (32). The probe was observed to bear 3.1 chelates per oligomer, when assayed according to DELFIA protocol (33).
 |
Table 3.
| Compound | Sequence |
| 9 | 5[prime]-d(ACCAAAGATGATATTTTTAGACTTGACTCATGA)-3[prime] |
| 10 | 5[prime]-d(AACACCA[Delta][Delta][Delta]ATGATATTTTTAGACTTGACTCATGA)-3[prime] |
| 11 | 5[prime]-d(CTTCTCCAAGAACTATATAGACTTGACTCATGA)-3[prime] |
| 12 | 5[prime]-d(CTTCTCAAAGAACTATATAGACTTGACTCATGA)-3[prime] |
| 13 | 5[prime]-d(ACCACAAAGAACCCTGATAGACTTGACTCATGA)-3[prime] |
| 14 | 5[prime]-d(ACCACA[Delta]AGAACCCTGAGTAGACTTGACTCATGA)-3[prime] |
Hybridization assays
The hybridization assays were carried out in a Tris buffer (50 mmol/l, pH 7.5, 0.01% Tween 20 and 0.5 mol/l NaCl). Usually, 50 particles of each category were incubated in 10 µl of the buffer containing a known amount of target oligonucleotide 9-14 and the fluorescently labeled probe 15. The mixed-phase hybridization was typically allowed to proceed for 24 h under gentle agitation (Rotamix, 20 r.p.m.). The particles were washed twice with 200 µl of glycine buffer (50 mM, pH 10), containing 20% l-propanol. The particles were then transferred into a quartz capillary tube for measurement. The fluorescence emission of each particle was determined separately on a microfluorometer (22). Three different measurements were carried out with each particle: the fluorescence emission of (i) dansyl (excitation at 330 nm, emission at 515 nm, no time delay), (ii) fluorescein (excitation at 490 nm, emission at 515 nm, no time delay), and (iii) the europium chelate (excitation at 330 nm, emission at 616 nm, delay time 296 µs, time-window 300 µs). The excitation was attenuated 100-fold for dansyl and 1000-fold for fluorescein. The counts resulting from 1000 excitations were integrated for each measurement. In an multiparametric assay, up to 50 particles were measured. This ensured that at least three particles of each category became subjected to a measurement. Unspecific adsorption of the labeled probe (15) to the particles was estimated by treating the particles with 15 in the absence of the target oligonucleotide (9-14). The resulting background signal was subtracted from the signal observed in the presence of the target. Only these corrected signals, referring to sequence selective binding, are given throughout the paper. However, on examining the kinetics of hybridization, this kind of correction was not done, but only the background signal of particles themselves was subtracted.
RESULTS
Categorization of the particles with prompt fluorophores
In multiparametric assays, a mixture of microparticles bearing up to six different allele-specific oligonucleotide probes was used as a solid phase. The particle bearing a given oligonucleotide probe was additionally labeled with one or two prompt fluorophores in such a manner that the determination of the emission intensity of the reporter group(s) allowed identification of the particle and hence the sequence of the attached oligonucleotide probe. The prompt fluorophores chosen for this purpose were fluorescein and dansyl. The wavelength of excitation and emission of these two labels differ so markedly that the presence of one label does not severely disturb the detection of the other. Table 2 summarizes the six different categories employed. The fluorescein and dansyl labels were introduced as phosphoramidite building blocks 7 and 8 during the conventional phosphoramidite chain assembly of the oligonucleotide probe.
With both fluorescein and dansyl, the observed intensity of fluorescence emission was considerably varied as a function of the site of labeling. Incorporation of the reporter group close to the 3[prime]-terminus gave up to 2-fold as high intensities as 5[prime]-tethering. For this reason, the prompt fluorophore was always placed in the same position, i.e. at the second (or second and third) nucleosidic (or non-nucleosidic) unit calculated from the 3[prime]-terminus. It is also important to note that, owing to marked concentration quenching, the emission intensity of these fluorophores was not proportional to the amount of the labeled building blocks used in the chain assembly. To eliminate this concentration quenching, 7 and 8 were diluted with thymidine phosphoramidite. When the total concentration of the phosphoramidite building blocks in coupling was 0.1 mol/l, the applicable dilutions ([7] or [8]/[T]) were 1:3 and 1:10 for dansyl, and 1:20 and 1:75 for fluorescein. The use of these dilutions led to categories that could be unequivocally distinguished. Figure
A
 B  |
Figure 2. Distinction of the particle categories. (A) Dansyl signals for categories 1 ([closed square]), 2 ([open square]) and 3 ([closed circle]). (B) Fluorescein signals for categories 1 ([closed square]), 4 ([open circle]) and 6 ([open triangle]). The vertical lines show the upper and lower limits of each category. The signals shown are corrected ones [the background signal (average value of category 1) subtracted from the observed signal]. Mean values of categories 1, 2 and 3 in A were 0, 31630 and 88747 cps and the standard deviation of a mean 1062, 5186 and 7699 cps, respectively. Mean values of categories 1, 4 and 6 in B were 0, 74483 and 225821 cps and the standard deviations of a mean 352, 16546 and 38727 cps, respectively.
As indicated in Table 2, the categorization was usually based on the presence of a single fluorophore at a low or high concentration level. Only category 5 was characterized by simultaneous presence of both labels, fluorescein and dansyl, at a low concentration. Although in this case the two reporter groups were tethered to adjacent nucleosidic/non-nucleosidic units, viz. second and third unit from the 3[prime]-terminus, no energy transfer between the labels appeared to take place. The signal levels were similar to those observed when only one fluorophore at this concentration was used as a reporter group (cf. categories 2 and 4). Fluorescein signal was found to exhibit marked overflow to dansyl channel, but this overflow was constant (60%) and could hence be subtracted from the apparent dansyl signal.
Hybridization
To ascertain that the categorization of the particles has no influence on their hybridization properties, the six particle categories listed in Table 2 were prepared, but now in such a manner that they all bore a common oligonucleotide probe (1a-f). When a mixture of these particles was assayed with the complementary target (9) and fluorescently tagged probe (15), the measured europium(III) emission signals that quantified the hybridization were equal for particles of each category. In fact the observed independence of the hybridization efficiency on the particle category is expected, because always <25% of the particle-bound probes actually bore a prompt fluorophore, the bulk of the sequences containing thymidine in place of the fluorescently tagged structural unit.
Figure 3. The effect of additional particle categories and/or targets on the hybridization properties of a given particle category. The signal from particles 3 hybridized with target 11 and probe 15 plotted against the target concentration: ([squ]) in the absence of additional particle categories or targets, (o) in the presence of all six particle categories but no additional targets, and (x) in the presence of all six particle categories and targets. The particle density employed was 5 particles of each category in 1 µl. A sandwich-type model system for simultaneous detection of several synthetic oligonucleotide targets was then constructed. The targets were selected in such a way that they mimic different types of gene mutations, viz. a three base deletion (cf. targets 9 and 10), point mutation (cf. targets 11 and 12) and point deletion (cf. targets 13 and 14), related to cystic fibrosis gene mutations [Delta]F508, G542X and 1078delT, respectively. Mixtures of these targets were assayed with a mixture of particles 1-6, each of which bore an oligonucleotide probe complementary to the 5[prime]-terminal sequence of one of the targets 9-14. Targets 9-14 additionally contained a common 16 base long 3[prime]-terminal sequence designed to bind the fluorescently tagged oligonucleotide 15. Unspecific adsorption of 15 to particles 1-6 was estimated by incubating the particles with the labeled probe in the absence of the target oligomer 9-14. At 100 nM concentration of 15, <0.3% of 15 was adsorbed to the particles. The fluorescence emission resulting from this unspecific binding was always subtracted from the signal observed in the presence of targets 9-14. Only these corrected fluorescence intensities, referring to the sequence-selective hybridization, are given throughout this paper. They were further transformed to the number of target/fluorescent probe duplexes hybridized to the particles with the aid of a calibration line obtained by carrying out two parallel sets of hybridization reactions in the concentration range 0.17-50 nM of the target (9). The concentration of the fluorescent probe (15) was equal to that of the target. After incubation and washings, one set of particles was subjected to microfluorometer measurement from the particle surface, while from the other set, europium(III) ions were released to solution and determined according to the DELFIA protocol (33). The calibration line obtained (data not shown) was essentially identical with those reported previously (23-25). The length of the hybridizing sequence of the particle-bound oligonucleotide probes was varied from 9 to 13 nucleosidic units to obtain optimal selectivity of hybridization. The optimization was carried out as described earlier (25). On these bases the following lengths of the complementary region were selected: 13 bp in particles 1 and 2 to detect three base deletion (9 versus 10), 12 bp in particles 3 and 4 to detect a point mutation (11 versus 12), 10 bp in particle 5 and 9 bp in particle 6 to detect a point deletion (13 versus 14). The efficiency of hybridization, i.e. the amount of target bound compared to the total amount of target in the system, was then measured in the concentration range 0.17-17 nM of the target, the concentration of the fluorescent probe (15) being 100 nM. Consistent with the results obtained previously with non-categorized particles (25), the hybridization efficiency moderately increased with the increasing target concentration (data not shown). When only one category of particles was present, the hybridization efficiencies at 17 nM concentration of the target were: 1/9 89%, 2/10 72%, 3/11 72%, 4/12 62%, 5/13 60% and 6/14 36%. Neither the presence of particles belonging to another category nor the presence of additional targets had any effect on the efficiency of hybridization over the entire concentration range studied. Figure The possible cross-hybridization was further examined by using a mixture of all six particles (1-6) as a solid phase but only one target. The concentration of the target was varied from 50 pM to 50 nM at a constant concentration of 100 nM of the fluorescent probe (15). Figure
A
B
Figure 4. The specific and unspecific binding of targets 9 (A) and 14 (B) to particles 1-6, when particles of all categories are present. Categories: 1 ([closed square]) specific for target 9, 2 ([open square]), 3 ([closed circle]), 4 ([open circle]), 5 ([closed triangle]) and 6 ([open triangle]) specific for target 14. The signals are plotted against the target concentration employed.
Figure
As a further piece of evidence for the applicability of the present categorized particle approach, a multiparametric assay was carried out with four samples prepared by mixing synthetic oligonucleotides. Each sample contained a different combination of normal and/or mutated sequences related to cystic fibrosis mutations [Delta]F508, G542X and 1078delT discussed above. Particles 1 and 2 were aimed at recognizing the normal and mutated sequences of [Delta]F508, respectively, particles 3 and 4 the normal and mutated sequences of G542X and particles 5 and 6 the normal and mutated sequences of 1078delT. The results were interpreted as follows. The signals for the particles recognizing the normal and mutated sequence were determined. If the ratio of the intensities of these two signal (normal/mutated) was >5, the interpretation was homozygous normal. A value <0.2 was interpreted as homozygous mutant, and a value between 0.5 and 2 as heterozygous carrier. Table 4 summarizes the results obtained. With each sample, the interpretation of the results was unequivocal. No marked cross-hybridization took place, since the emission intensities were very similar to those observed when only one target was present. Furthermore, it is worth noting that the observed signal intensities could be converted to the concentrations of target in the sample, when the signal/[target] relationships presented in Figure
Figure 5. Kinetics of formation of sandwich hybrids on particles 1-6. The concentration of targets 9, 11 and 13 were 5, 1.7 and 17 nM, respectively. Particle density was 5 particles of each category in 1 µl and the temperature 25°C. For the categories, see Figure 4. Feasibility of multiparametric sandwich type hybridization assay using microparticles that allow simultaneous quantification of six oligonucleotides was demonstrated. Two organic prompt fluorophores were used for categorization of microparticles, and they were introduced as phosphoramidite building blocks during chain assembly. However, these building blocks had to be diluted with thymidine phosphoramidite to prevent the concentrations quenching (i.e. the `inner-filter effect'). This inner-filter effect occurs in concentrated mixtures of organic fluorophores, as nicely presented for labeled particles by Scott and Balasubramanian (34), and limits the applicable concentration range. Furthermore, dansyl is known to be an environmentally sensitive label having higher quantum yield in a hydrophobic environment (35). This was also observed in the present study, not only for dansyl but for fluorescein as well. For that reason the labels were placed to the 3[prime]-end of the immobilized oligonucleotide, i.e. near the particle. The six different categories created were distinguished unambiguously and the categorization did not affect the hybridization properties. The number of categories could even be increased since every particle category behaves independently, ignoring other particle categories and targets. The number of levels with one label can easily be increased from two to three or perhaps to four. Thus with these two fluorophores, >10 categories may be created. By increasing the number of labels, the number of possible categories can be increased to dozens, the exact number depending on the compatibility of labels. Especially the use of lanthanide chelates for categorization, not only for quantification of hybridization, is attractive, since they do not exhibit concentration quenching. The main advantage of this methodology is evidently the exceptionally large dynamic range. The intensity of the emission signal measured from a single particle is linearly related to the concentration of the target oligonucleotide over three orders of magnitude using one constant probe concentration. The major shortcoming of this methodology is the relatively slow hybridization. This shortcoming is typical for many mixed-phase systems (15,20,36) and could be effected in our assay by increasing the number of particles of one category (24) and by strengthening the mixing. Another limitation is that the length of allele-specific probes must be rather carefully optimized. For example, a point mutation studied (C->A) can be analyzed with 12mer probes (3 and 4) but the one base deletion used in this study has to be analyzed with 10mer and 9mer probes (5 and 6). The 11mer probes still exhibit cross-reactivities up to 100%. This need for optimization is not, of course, the problem of this methodology in particular, but a general feature arising from the fact that hybridization efficiency is a function of sequence. Systems, such as sequencing by hybridization (SBH), also suffer from unequal stabilities of as long duplexes (1). There are, anyway, possibilities to overcome this problem (37,38). Although the optimization of the length of the allele-specific probe increase work, the standard curve achieved after optimization enables the quantification of target oligonucleotide. As seen from Table 4, the multiparametric analysis applied in the present study does not only give yes/no answers, but it allows rather accurate quantification of various oligonucleotides. In summary, the multiparametric approach described in the foregoing is still at an early stage of development. Successful miniaturization, automatization and particle transport system are prerequisites for routine clinical applications. Nevertheless, the data presented here show that the chemical basis of the multiparametric assay is sound. As far as we can see, the slow kinetics of mixed phase hybridization constitute the major hurdle for further development. We have shown previously (24) that the hybridization may be greatly accelerated by increasing the density of particles in the sample, but simultaneously the sensitivity is decreased, since the target oligonucleotide is distributed evenly among all the particles, and hence the signal obtained from a single particle is diminished. It is also conceivable that the categorization of particles could be more conveniently carried out as a part of polymerization of the particles. Though further development and optimization of the assay format are still needed, some underlying features of the system are superior to those of existing assay systems. Above all, the exceptionally wide linear range of detection allows more accurate quantitative determination of several oligonucleotide concentrations from a single sample than the existing methods. Accordingly, the method, for example, shows great promise to the development of convenient quantitative PCR assay. Table 4. 
DISCUSSION
Entry
Cat.
Sequence
Signal
N/M
Interpretation
c(target)/
nMEstimated
c(target)/nM
1
1
dF508 N (9)
720963
52
N
5.0
4.2
2
M (10)
13814
3
G542X N (11)
250859
9.2
N
1.7
2.3
4
M (12)
27224
5
1078delT N (13)
1898214
143
N
17
15
6
M (14)
13235
2
1
dF508 N (9)
46393
1.0
heterozygous
0.5
0.35
2
M (10)
44456
carrier
0.5
0.40
3
G542X N (11)
1280972
142
N
8.3
12.3
4
M (12)
9011
5
1078delT N (13)
61905
>10
N
0.5
0.48
6
M (14)
n.p.
3
1
dF508 N (9)
4497464
196
N
3.3
2.2
2
M (10)
22906
3
G542X N (11)
13181
0.053
4
M (12)
247660
M
1.7
2.4
5
1078delT N (13)
546973
>10
N
5.0
4.4
6
M (14)
n.p.
4
1
dF508 N (9)
209590
11
N
1.7
1.4
2
M (10)
19547
3
G542X N (11)
2073206
0.91
heterozygous
17
20
4
M (12)
2283423
carrier
17
23
5
1078delT N (13)
18171
0.052
6
M (14)
348434
M
5.0
4.4
ACKNOWLEDGEMENTS
We thank Dr Ruth Schmid, SINTEF (Norway), for the generous gift of particles employed. We would like to thank Drs Harri Takalo and Veli-Matti Mukkala (Wallac Oy, Turku, Finland) for luminescent marker. Financial support from the Academy of Finland, the Research Council for Natural Sciences and Technology, is gratefully acknowledged.
REFERENCES
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 3 Dec 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
H. Harma, P. Tarkkinen, T. Soukka, and T. Lovgren
Miniature Single-Particle Immunoassay for Prostate-specific Antigen in Serum Using Recombinant Fab Fragments
Clin. Chem.,
November 1, 2000;
46(11):
1755 - 1761.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Print PDF (101K)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (17)
Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Articles by Hakala, H.
Articles by Lonnberg, H.
Search for Related Content
PubMed
PubMed Citation
Articles by Hakala, H.
Articles by Lonnberg, H.
Social Bookmarking
What's this?