Probe Density Considerations and Elongation of Self-Complementary Looped Probes Where Probes Are Attached to a Solid Phase

Background Molecular Stringency in Multiplexed Assays - A $elf-eomplementar\ oligonucleotide capture probe in. a "looped ^" conilguiatioπ may be used to adjust molecular stringency in an assa> Assay stringency relates to the positive results produced by an assa>, such that high stringency conditions generate relatively fewci positive results than lower stringency conditions Looped probes are described in WO 01 98765, entitled "Multianalyte Molecular Analysis Using Application-Specific Random Particle Arrays" and US Patent No 6 Jb 1,945 (assigned to Gen Probe, Inc ) Such a probe consists of a 5 ^' -terminal subsequence and a complementary 3 ^" -terminal subsequence, tethered by an unrelated subsequence, the two terminal subsequences capable of forming a duplex {"stem"), and the tether forming a loop, and either the 5'-temiina! subsequence of the 3 ^* -teiminai subsequence capable of forming a duplex xύth a taiget nucleic acid Hie probe may be attached to a solid phase such as an encoded micjopaiticle {'"bead"), b\ way of an appropriate functional modification of the 5'terminal subsequence or the loop subsequence

Using a fluorescence acceptor and a proximal fluorescence quencher (as discussed in VS Patent No 6.534.274). capture of a target nucleic acid is detected by way of detecting a transition from the Closed ("C") state of the capture probe to the Open ("O ^" ) state or the target-associated COT") state, the O-state contributing to "background" fluorescence, independent of target concentration {Fig, 1) In this competitive equilibrium, low stringency, favoring the closed stale, will i educe the likelihood of formation of the open (or other intermediate state, see Detailed Description, below) required for probe-target duplex formation, thereby diminishing the detection sensitiύt) Conversely, high stringency, the open state, likewise reduce the likelihood of target capture - by reducing the stability of any probe-target duplex ~ while producing imhsαinimale fluorescence, independent of captured target, thereby reducing specificity

Thus, the use of a looped probe calls foτ resolution of the conflict between detection sensitivity and specificity, preferably by operating near an optimal

stringency, determined by a choice of buffer conditions and operating temperature For typical buffer conditions, which generally are of low ionic strength, c g cot responding to salt concentrations of 5OmM, this step requites selection of an optima! detection temperature, preferably at or above the range of the midpoint of the melting cυne where specificity is optimal Optimal stringencies generally will depend on capture probe sequences, and on target configuration and/or length Thus, identifying the optimal stringency iange in a multiplexed assay thus becomes increasingly difficult with each different probe added, gi\en the dispersion of the melting curve profiles of a set of different probe-target complexes under given assay conditions Summary of the In vention

Disclosed are methods of enhancing detection sensitivity and expanding the range of stringencies compatible with detection of specific targets, especially uheie there is a low target concentration, as t>ρicalh encountered in, <?. ₍ y\ the detection of genomic material from infectious agents {see e g , Chen, Martinez &. Mulchandani, " ^" Molecular Beacons λ Real-Time Polymerase Chain Reaction Assay for Detecting SalftionelhC Analytical Biochemistry 280, 166- 172 (2000)) Also disclosed is a method of enhancing detection sensitivity b\ providing for target capture to a self complementary ("looped") probe, anchored, preferably by its loop subsequence, at a lateral densit) of at least a certain preset minimum, on a solid phase carrier, preferably a microparticte ("bead ^" )

KurtheT disclosed is a method of stabilizing a probe-target complex under conditions of high stringency by providing for target-mediated, enzyme-catalyzed elongation of the 3 ^" -terminal pfobe subsequence to convert the probe-target complex ( ^" O F ^" ), formed as a result of target capture and characterized by fluorescence, snto an elongation product ("eOT') of enhanced thermody namic stability (Fig. 2). The formation of the eO'l state can be detected by temperature cycling the eOT complex may be exposed to higher temperatures without loss of fluorescence - which would otherwise result, for a non-elongated complex (in the OT state), from the release of the target at the higher temperature and formation of the closed ("C") state of the probe - upon subsequent le.urn to lower tempeia.ure.

The formation of this elongation product has at least a three-fold benefit

(i) enhance the sensitivity of target detection - by converting the C state of the probe into the eOT state, even under conditions of extreme stringency, selected, for example, to ensure enzymatic efficiency particularly in homogeneous assay designs { (see t\ g. "Transcription Amplification System with Integrated Multiplex Detection; Functional Integration of Capture, Amplification and Multiplex Detection" filed 9/2/2005; Seria! No. 1 1/218838, incorporated by reference), this conversion ensures high detection sensitivity by accumulation of elongation product, over an extended period of time, by way of random fluctuations of the closed into the open (or related reactive intermediate, see below) state permitting target capture and enzyme-catalyzed elongation; to the extent that the eOT state is irreversible under prevailing assay conditions, this conversion is akin to a digital "ON ^1' signal;

(ϋ) enhance the range of optimal stringency of a multiplexed assay - essentially by raising melting temperatures and thereby avoiding operation in the range of temperatures coinciding with dispersion in the melting curves of multiple distinct probe-target pairs; and

(iii) enable the application of allele-specific detection and implementation of a phasing strategy, in analogy to the phasing method described in US Patent Application Serial No. 10/271 ,602, entitled: "Multiplexed Analysis of Polymorphic Loci by Concurrent interrogation and Enzyme-Mediated Detection, ^" incorporated by reference. Description of Figures

Fig. 1 is an illustration showing the closed ("C"). open ("O"). and target-associated ('"OT") states of a self-complementary ("looped") capture probe. Fig. 2A is an illustration showing the target-mediated, enzymatic elongation of a looped probe labeled with a fluorescence donor on the 5 '-terminal subsequence and an acceptor on the 3 '-terminal subsequence.

Fig. 2B is an illustration showing the target-mediated, enzymatic elongation of non- labeled looped probe.

Fig. 3 is an illustration comparing a volume element of solution containing uniformly distributed capture probes, and a volume element containing a microparticle and capture probes confined to a shell.

Fig. 4 A is an illustration showing the configuration of a homogenous assay performed using looped probes displayed on a pre~assembled random array of encoded beads.

Fig. 4B is an illustration showing an arrangement for performing a homogenous assay using looped probes displayed on a pre-assemblec! random array of encoded beads, where the array is mounted on an insert at the tip of a reaction tube and imaged in an inverted imaging arrangement.

Fig. 5 is a representation showing the capture of RNA target to bead-displayed self- compiementary capture probes in homogeneous BeadChip assays.

Fig. 6 is an illustration of target ""hopping" process and the escape process with concomitant shape relaxation.

Fig. 7A is an illustration of the effect of probe elongation on the melting curves of several probe-target complexes, and Fig. ?B is an illustration of the effect of randomly aborted probe elongation on the distribution of affinity constants.

Fig. 8 is an illustration of phasing, performed by elongation of allele-specific looped probes.

Fig. 9 is an illustration showing the configuration of a homogeneous assay performed using with iabeled looped probes displayed on encoded suspended beads.

Fig. 10 is an illustration showing the configuration of a homogeneous assay performed using with non-labeled looped probes displayed on encoded suspended beads. Fig. i 1 is an illustration of into an incubation chamber in place on a silicon wafer.

Fig. S2A is an illustration of a magnetic trap.

Fig. 12B shows the computed field distribution of quantities relevant to magneto- phoresis.

Fig. IJ shows bead-map plotted with Cy 3 against blue, showing three clusters of beads.

Fig. 14 shows the dose response of target interaction with specific and non-specific looped probes.

Detailed Description

1 ~ Mathematical Description of Molecular Stringency: Competitive Target Capture

In general, the interaction of a looped probe with a target nucleic acid will be go\ emed by a set of coupled equilibria between the non-fluorescent closed ("C ^" ) state, and the fluorescent open ("CT) state and the fluorescent target-associated ("OT") state Capture of a target nucleic acid is detected by way of defecting a transition from the C to the OI state The O state, which is not associated with the target, contributes to a "background fluorescence" The equations below describe mathematically the corresponding coupled equilibria The four input parameters are the initial iooped probe concentration [P]". initial target concentration [T] ⁰ , and the relevant equilibrium constants

In the most general situation, the target is permitted to interact not oπh with the open but also dύecti) with the closed state of the looped piobe {in a displacement reaction) so as to form a probe- target complex, hor molecular beacon probes in solution - beacons, in contrast to the looped probes considered here, arc designed to form a probe-target complex by \\a\ of the loop sequence and thus do not impose molecular stringency - Bonnet ct al reported a mathematical model applicable under conditions of excess target (see Bonnet et a!, Proc Natl Acad Sci USA VoI %. pp 6171-6176, Ma\ 1999, Biophysics) Here, we consider the more general situation, f.e , that there is usuall> low concentration of target and excess probe, in assays using solid phase-immobilized probes to detect targets in solution

Consider first looped probes, exposed to targets, the probes capable of adopting one of three states (s) a duplex state (associated with target), (ii) a closed state (the complementary stem subsequences forming a duplex), and (iii) an open state, for example in the form of an open random coil Iprex alent, for example, at high temperature) At equilibrium

where OT is the looped probe-target duplex, C is the probe in its closed state, O is the probe in the form of a random coil, and T is the free target. The normalized fluorescence at a given temperature should be the sura of the contribution from each of the three states:

/ ^■■ - α oc « [or J] + / _n J . [UC]4 V. [Uυ]

where α, β, and γ are the fluorescence quantum efficiency (QE) of the looped probe in each state, and

P" = [OT] ₊ [C] ₊ [O]

The law of mass action gives the following expression for the equilibrium constants governing the dissociation of the looped probe:

_K ^κ "

These afTinily constants are related by the following relation:

Two limiting cases of interest are:

Excess probe, i.e., /'" » T" :

The fraction of probes in each state can be expressed in terras of the equilibrium constants, K _c and K ₀ as follows.

ψ = r ^{ r + κ, + κl ^ι

•fcL λ' (K + A' Y

M κ,iκ _c +κ _t> y ^ι

Thus, the total fluorescence intensity is:

F - # r ^' (P' ^' + κ ₍ + K, ) ^"x + βc ₍ (A;. + ^ ) ^" ' + _? κ _n (κ _c í κ, _s y ^[ ,

Excess target, Le,, P ^'' « T ^' :

The fraction of probes in each state again can be expressed in terms of the equilibrium con slants. A ^' ,- and K ₀ as fo! lows ;

Ir]

[O]

Thus, the fluorescence intensity is.

F = [a ^r r -f βκ_ + 74CJF ^:> í λ:; +^J ^'! .

These equations may be simplified by assuming equality of quantum efficiencies (QE) in the duplex and open states, i.e., or - ^■ ;■' , and negligible QE in the closed state, i.e., β - Q :

Then, for the case of excess probe, i.e , T" « P" :

and similarly, for excess target, i.e. / ^>l! « T" :

Both expressions are equivalent to a Langmuir adsorption isotherm describing the capture of target to a probe-decorated solid phase in a process governed by a single effective affinity constant K^. ^~ (K _r +Kj'

The fraction of signal originating from the probe-target complex, compared to that originating from the open state of the probe, is given by:

Simplified Model: No Displacement - A similar result also is obtained by considering the target to interact only with the open form of the looped probe in accordance with a coupled equilibrium:

C <z>O+ T<?>OT

where Kj and K^ are the association equilibrium constants, namely:

_oφτh ^r[O]

^{1 J} 1 + K _; . [θ]

Similarly

K where d — ^■■ ; ,

(I ₊ Jf ₁ )

These two algebraic equations yield:

Then, fox excess probe, i.e.,

I O and similarly, for excess target, i.e

Both expressions are equivalent to a Langmuir adsorption isotherm describing the

15 capture of target to a probe-decorated solid phase in a process governed by a single effective affinity constant, K ₁ ^. ^~ OK,

The fraction of signal originating from the probe-target complex, compared to that originating from the open state of the probe is given by:

Both models thus generate similar mathematical expressions for [OT], namely :

5 where K _c n represents an association equilibrium constant governing the reaction Fí / oOr , between any of the states of the probe, P , and the target-associated

state, and P° and T ^υ respectively denote the initial concentrations of the probe and target. For the general model, K _e «- ^::: (K ₀ -MC ₀ ) ^" ' and for the simplified mode!, K _ej γ ^:::

Both modeb likewise generate similar expressions for the parameter η, namely:

where A, for the general model, is given by. λ ^~ ( H KJK^ and for the simplified model is given by λ ~ (HA ^' _/ )/ K ₁ . Under conditions of low coverage. [OT]ZP ⁰ « 1, //increases linearly with [OTj which, in this regime, is in turn linearly dependent on K^, . Hence, in this low coverage regime, an increase in K#r _> reflecting choice of ionic strength andZor temperature, will lead to an increase η, and hence detection sensitivity. This can be brought about by a choice in buffer conditions such that affinity K> or K _co decreases, which destabilizes the O state in favor of the OT state.

Probability of Target-Probe Encounter; Solution vs Solid Phase - For given target concentration, the probability of a target molecule encountering a probe is detenu ined by the effective concentration of probes With reference to Fig. 3, consider a test sphere of a radius r and a concentric shell of radius R r ^; δ the sphere displaying probes at a density σ ^■ - P" r~. The effective probe concentration within the shell is

given by Letting R decrease toward r, that is, in the limit δ-~ ^' 0, the local probe density approaches the limit in this limit, probes may be viewed as "condensed ^" on the bead surface. For example, given a bead of diameter 3.2 μm and a typical value of f" of 10° per bead, σ- 10 ^? μni ² . The effective probe concentration within a shell of dimension δ = 0.1 μm is thus:

[p.% ] * 3 K 10 ^s ymt ^' ] 0 ![/»?] * 3 x 1 (V x 10 ^ [ V/ J / 10 ' ' [/ ] 3λϊλ /

Typical conditions for target capture in solution involve a choice of probe concentration equal to the maximal anticipated target concentration Assuming a 5 dy namic range of 2 orders of magnitude, the probe conceouation vail exceed the lowest detectable target concentration by not more than 2 orders of magnitude Thus, in order to permit detection of target at a concentration of 5OnM (see Example \), a typical probe concentration will be 1 μM The effective probe concentration associated with the bead thus exceeds, b> at least 3 orders of magnitude, that typically

I O encountered in solution Accordingly, as a target approaches the solid phase carrier surface, it encounters probes with a far higher probability than that governing such encounters in solution, and this translates into a correspondingly higher local concentration of probe- target complexes This invention discloses immediately, below, a hopping model permitting the target to interact, during each encounter with the bead

15 sui face, v\ith not one but multiple probes, thereby extending its jtesidenee time near the surface

Enhanced Detection Sensitivity: Target "Happing" and Recapture — Experimental observations, described in greater detail in Example t and in Figs. 4 and 5. especially in the upper panel of Fig. 5, for a looped probe attached by its loop subsequence to a 0 microparticie ("bead"), indicate the response to display, in the regime of low target concentration, a substantially enhanced detection sensitivity as compared to the response of thai probe in solution

The enhancement is attributed to target "hopping" from occupied to nearby unoccupied capture probes f see Fig. 6A, E) That is. targets execute random walks (of 5 varying extent) on the surface by hopping from site to (unoccupied) site If "hopping" can occur sufficiently rapidly so as to leave the target conformation essentially unchanged and thus " ^" primed" for recapture (Fig. 6λ\ this process \ύll increase the residence time of the target at or near the surface Denoting b\ τ the characteristic relaxation time of the target conformation, ftom its constrained state it must adopt for 0 association with the carrier-displayed probe, to the unconstrained state it adopts as it "escapes" into the bulk solution (Fig. 6O. the distance d\\ between any occupied probe site and the nearest unoccupied site(s) so as to permit (random) "hopping" on a

tjrøescale n, < τ. Denoting by μj _t a characteristic hopping mobility, and corresponding diffusivity D^ ^::: (kT/M)μi ₁ , M representing the mass of the target molecule, this condition translates into (I _NN ² ' ^C ϊ\τ or, for the probe density, σ ^■ - d>;>f ² > 1/ D;,τ.

Phenonienologieally, the increase in target residence time manifests itself in the form of a reduction in the observed rate of dissociation. The ratio, kjkjo, of the observed to the "intrinsic" rate decreases with increasing probability of a target completing a "hop" from its current probe site to a nearby (unoccupied) probe site, and this probability, 0, in turn increases with the number of probes P ^(> provided on the surface, and with the unoccupied fraction.. \-ϊ\ of those probes. Thus, k _{ > may be represented in a form

where θ(/ ^>0 ,l - T) represents the probability of target recapture at a site close to the site of release; θ(/' ^Lt ,l - 1 ^" ) will be a monotonicaily increasing function of /*'' and 1 ^■ - r , and max(θ) < 1 .

Solving for F from the detailed balance equation, k _a H-F)T, ~ fcJZ yields:

where Kn ^:::: k _r ^' k ^~ j/, represents the affinity constant observed in the absence of target retention; in the limit of low target concentration, or small affinity constant T ~ KJ ^' \ ,

The observed affinity constant,

is enhanced at low target concentration, reflecting the large fraction of capture sites available to each target molecule; K decreases toward its "intrinsic" value at high coverage Regardless of its detailed form, the recapture probability function, relates an increase in observed affinity to an increase in total surface probe density and/or decrease in coverage. By enhancing the observed affinity, this cooperative effect arising from target hopping between densely grafted probes on a solid surface favors complex formation and thus accounts for an enhanced sensitivity,

The arguments advanced herein are not limited to the seif-corøpiementary ("looped") probes employed here, and will apply to any target (or ligand) capture to solid-phase displayed capture probes (or receptors) at low target (or ligarsd) concentration. Interfaciai Polarization - At high stringency, capture especially of short targets will occur within a polarized interfaciai region of elevated ionic strength, and hence under conditions of lower stringency as compared to conditions in the bulk solution. For example, for a 50-mM bulk NaCI concentration, this interfaciai region extends to a characteristic length \/κ -30 A beyond the surface of the solid phase carrier. Given the increased effective target concentration, this will further stabilize the OT state, a conclusion which also follows from the analysis of the mathematical description described above (see Eq 1 ). Under these conditions, an effect such as a coυnteri on- mediated attraction of short range (Ha & Liu, Phys Rev Letts. 79, pp 1289 - 1292 ( ^" 1997)) may contribute to target retention within the interfaciai region. Expanded Dynamic Range - The experimental observations described in the Examples below also indicate the response of looped probes anchored to a solid surface to display a more than two-fold expansion of dynamic range as compared to that observed in solution,

At typical grafting densities of at least 10 ^? probes per bead, a solid phase assay , especially in the regime of low target concentration, corresponds to conditions of excess probe. Under the assumption, a - γ . β - 0, discussed above, and under the further assumption λ ^" \ >> K,, , the absolute fluorescence intensity assumes the form:

This expression, describes an increase in the intensity of fluorescence emitted by looped probes with increasing probe density. That is, the response, given by the slope. in fluorescence intensity as a function of variations in target concentration, will affect the intensity of emitted fluorescence. For example, under conditions described in Example 1 , K _c * 0.1/&V/ , so that, if the grafting density, and hence P ^iJ is varied from (an equivalent of) 10 nM to (the equivalent of) 1OmM, the

response in fluorescence signal intensity can be varied over an order of magnitude, from OA ex to ex:

The broadening in the response is reminiscent of that observed when comparing the response of a polyclonal antibody to that of a monoclonal antibody (Tarnok, Harabsch, Chen 8ι Varro, Clinical Chemistry 49, No 6, pp 1000-1002, 2003) However, as decribed herein, anchored looped probes, grafted at high density, also display an enhanced detection sensitivity at low target concentration. This effect, which has not been described in connection with immunoassay designs replacing a monoclonal capture antibody by a polyclonal capture antibody, is attributed here to an enhanced observed ("effective") affinity at low coverage in accordance with a target bopping model .

In accordance with the target bopping model, a cooperative effect related to probe grafting density enhances the affinity observed at low coverage, thereby further contributing to the heterogeneity in the response in a manner that is favorable to generating an expanded dynamic range of target detection. At low target concentration, the response is dominated by the enhanced affinity arising from target retention near the surface, and at high target concentration, the response is dominated by the low affinity associated with low grafting density. That is, the expanded dynamic range reflects the contributions of enhanced sensitivity at low coverage, and those of solid phase carriers of Sower affinity at high coverage.

2 - Formation of eOT State: Enhancing Operating Range and Detection Sensitivity - The use of a looped probe calls for operation within a range of optima! stringencies that is determined by a trade-off between detection sensitivity and specificity. Conditions of low stringency will stabilize the C state, thereby rendering target capture more difficult and reducing detection sensitivity. Conversely, conditions of high stringency will destabilize both the C state and the OT state, as evident from the results of the detailed mathematical description provided herein above, thereby reducing specificity, in the extreme, the open state of the probe will produce fluorescence even in the absence of target. Optimization of specificity generally will dictate selection of an operating temperature near the melting temperature of the relevant probe-target complex.

However, as this choice also reduces the stability of the probe-target complex, it reduces detection sensitivity Conversely, a choice of lower stringency increases the sensitivity, but compromises the specificity of the response. When detection of target by capture to looped probes is to be performed concurrently with enzymatic target amplification (or other enzyme-catalyzed target manipulation) in a homogeneous format, or subsequent to such manipulations, but without intervening separation step, in a ~ ^l single-tube ^" format, the choice of optimal stringencies may be further constrained, ϊn practice, high stringency is preferred: for example, the conditions of Example 1 , involving the formation of a duplex of 20 base pairs, provide for SOmM sa 11 and an operating tern perature of 42 C .

Optimal stringencies generally will depend not only on specific capture probe sequences, but on target configuration and/or length, and the task of identifying the operating range of stringencies in a multiplexed assay thus becomes increasingly difficult, given the dispersion of the melting curve profiles of a set of different probe- target complexes under given assay conditions. The design of a multiplexed assay format calling for the concurrent detection of multiple targets by capture to matching probes, will thus further restrict the choice of optimal stringencies which depend on the stability of individual probe-target complexes.

Thus, target -mediated elongation of (the 3 'terminal subsequence of) a self- complementary probe provides a method of stabilizing probe-target complexes by converting the OT state into the elongated ("eOT") state and thereby a method of expanding the operating range particularly of multiplexed nucleic acid detection while simultaneously enhancing the sensitivity of detection. Elongation may be performed using DNA target and a DNA polymerase or RNA target and a Reverse Transcriptase (R.T), as described in the co-pending application included herein by reference. The probe is constructed so as eliminate "self-priming", either by providing strictly blunt ends of the stem, or preferably by providing an "overhanging" 3 ^" terminus. Expanding the Operating Range ~ The enhanced thermodynamic stability of the eOT state manifests itself in a shift to higher temperature of the melting curve: generally, the longer the template, the larger shift, hi contrast, since the .V terminal subsequence of the probe remains unmodified, the C --> O transition follows its original melting

curve, in a multiplexed assay, this shift of the dispersive portion of the melting curves of different probe-target complexes to higher temperature, renders the system more forgiving in terms of selecting a high operating temperature: as illustrated in this situation Fig. 7A, the ability to operate at high temperature ensures high stringency and hence specificity, and the ability remain outside of the range of dispersion simultaneously ensures high sensitivity. Non-uniform probe elongation, as a result of randomly aborted probe elongation reactions, would produce a polydisperse length distribution and would further broaden this distribution of affinity constants. Such an increase in heterogeneity will manifest itself in an increase in the dispersion of the (shifted) melting curves (see Fig. 7B): that is, randomly aborted elongation reactions provide a means of expanding the dynamic range of the assay.

Enhancing the Sensitivity - The enhanced stability of the eOT state also translates into enhanced detection sensitivity, as a result of shifting the equilibrium of the competitive probe-target interaction to the duplex state by converting OT states, essentially irreversibly, into stable eOT states. Phenonienologkally, this conversion corresponds to a a reduction of the observed rate of dissociation, and corresponding increase in the observed affinity of the probe-target interaction: to the extent that it is irreversible, this process, given sufficient time, will consume all available target

The enhancement in detection sensitivity afforded by generation of the {essentially irreversible) eOT state is particularly effective when operating in a regime of stringency permitting only the transient formation of an OT state. Random fluctuations producing the transient formation of a probe-target-enzyme-substrate intermediate will mediate the (essentially) irreversible conversion of a fraction of this intermediate OT state into an eOT state, leading, over time, to accumulation of eOT state and depletion of target,. The '"zippering~up" of the intermediate OT state producing the eOT state, akin to the turn of a ratchet, permit operation in a regime of low stringency without loss of detection sensitivity. 3 — Attele-specific Detection and Phasing

As with al tele-specific detection of nucleic acids generally, looped probes may be used to advantage in connection with Elongaii on-mediated Multiplexed Analysis of Polymorphisms (eMAP™; sec US Application Serial No. 10/271,602). In this

application, the use of a looped probe has the additional benefit of permitting control of molecular stringency so as to improve allele άϊ sen mi nation b\ target capture In particular, eMλP using looped capture probes which simultaneous!) serve as elongation primers permit the application of phasing, either in the mode described in detail in VS Application Serial No 10/271,602 (incorporated by reference), or by combining the stringent control of annealing conditions afforded by the design of specific stem subsequences with aileie-specific elongation of a 3" -terminal subsequence whose 3 ^' terminus is designed not to display complementarity with the 5 ¹ - tenninal subsequence so as to eliminate the possibility of self-priming T hat is, as illustrated in Fig. & the configuration of a first variable site, located within the portion of the sequence capable of annealing to the 3 ^" -tcrminal subsequence of the probe is detected by prcfeiential capture of the matching allele, and the configuration of a second v ariable site, located in juxtaposition to the 3 'terminus {or proximal position) of the probe, is detected fa> elongation (or lack thereof) Elongation products may be formed under conditions permitting incorporation of fluorescently labeled dNTPs or may be formed with unlabeled dNTPs and decorated by a fluorescentiy labeled hybridization probe, such a decoration probe can be designed to be directed to an additional polymorphic site of interest located in the elongated probe sequence Example I Homogeneous; Bemlehip Assay Using Looped Probes

A homogenous BeadChip assay format, shown in Fig, 1, was implemented by providing a variable gap configuration set to a large value during target capture and a smaller value during recording of assay images from a random encoded arraj of beads displaying self-complementary probes as well as positive and negative controls I he reaction volume was sealed by encapsulation of the reaction with mineral oil (from Sigma-λldrich)

BeadChips were prepared to contain a random arraj composed of 4,000 beads of four t\pes of color-encoded microparticles ("beads") on a 375-μtn thick <10ϋ" n- type Silicon substrate Color-coding was achieved by staining the beads in accordance with a solvent tuning method described in US Application Serial No, 10 348.155 (incorporated b\ reference) Stained beads weie functionaU/ed b) covalent

attachment of strepta\ idin to permit subsequent attachment of biotim lated self- complementary ("looped ^"1 ) probes, illustrated ni Fig. 1

One probe, displayed on one t>pe of bead, contained a 20-nt capture sequences specific to a 20-mer single-stranded target, the other probe contained an unrelated 20- mer sequence Three type of beads were respectively fuπctionali/ed with a target- specific ("matched") probe, a mismatched probe serving as a negative control, and a biotinj lated and C> 3 -modi fled oligonucleotide (" ¹ A lO ^" ) serving as an intensity reference, a fourth tv pe of bead, left tm-functionalized was added to dilute the array composition BeadChips were affixed to glass substrates using an epo>cy adhesive C ^" Loctite' ^" )and a polydimethylsiloxane (PDMS) spacer, either 400 μm or 1 ,000 μm in thickness, was cast, PDMS conforms well to Oat surfaces and provides a reliable seal, given its negligible thermal expansion up to iOO ^c C Two 400-μm spacers were placed adjacent to the mounted BcadChip, and two 1000-μm spacers were placed next to the 400-μm spacers, a glass covers! ip of 0 15 mm thickness was cut to fit the separation of the 1 OϋO-μrn spacers

To perform the assay, 1 5-μl of reaction mix containing specific target at a particular concentration was pipette-tuinsferred to the chip surface, the reaction volume was closed by fixing the coverslip via two PDMS pads placed onto the 1 000-μni spacers, and transferring 5-μl of mineral oil into the gap, capillary forces ensure that the oil Cjuiekl\ encircles and isolates the reaction volume After completion of the reaction, the coverslip was shifted so as to come to rest on the 400-μm spaceis to foim a 25-μm gap for optical interrogation

The result of titrating a 20-mer RNA target on a Beadchip using this setup is shown in Fig. SA at a temperature of 42C and in Fig, SB at two additional incubation temperatures, followed by imaging at room temperature Fluorescence intensity readings, normalized using the Al O fluorescence, are shown along with normalized data recorded from the same assay performed in solution, using a fixed looped probe concenuation. of 0 l μ\1 Compared to the solution response, the reaction with the bead-displaj ed probes displays a much broader detection dynamic range of target (3 logs) and substantially enhanced sensitivity at low target concentration

Example 2: Homogenous Assay in Suspension of Encoded Heath

The loopcd-probe design also can be used in a homogenous format with encoded beads in suspension, as described in TlS Patent No 6,25U(J^h US Application Serial No 10 204.7*W (incorporated by reference) As shown in Fig. 9, a reaction mixture in a sealed incubation chamber, or caitridge, may contain T7-tagged DNA template, components for in-vitro transcription reaction such as a T7 RNA polymerase, Vt el I known in the art, and looped-probe functionali/ed color-coded beads, each color corresponding to a unique capture probe sequence Preferably, encoded magnetic beads are used (sec US Application No S 1/218.838). and a random an ay of such beads is assembled in real time following completion of the assay, as described in US Patent No 6.2Sl.69 J . L S Application Serial No 10 204.79 ^Q

Two sets of magnetic beads (Spherotech, 4 10 μm in diameter, ρ- 1 13 g/ml), one encoded with a green d\e b\ solvent-tuning (RF.F Sohent Tuning}, the other left uncoiored. are covatently functional ized with Strepaύdin for attachment of a biotinlyated looped probe One probe, displayed on the green beads, contains a 10-nt capture sequence specific to a 20-nicr HIV single-stranded target the other probe contains a 10-nt sequence unrelated to HlY The looped probes are labeled with a Cs ? fluorescence dye on the 5' -terminal subsequence and a Blackhole quencher on the 3 ^' - tcrminal subsequence Buffer containing all the reaction ingredients is adjusted in density by properly mixing with 20% Ficoll PM 70 separation medium (λmersham) in D2O (Aklrich, p- 1 18 g ml, rj- 10 cpj The reaction suspension is then brought to 0 2S% solid content

In-vitro transcnption is pei formed in the sealed chamber, oi in a sealed cartridge, containing suspended beads (see also the detailed descriptions in the co- pending application included herein by iefeτence> The reaction is initiated bs raising the temperature to a predetermined value optimizing the efficiency of the I ⁷ RNA polymerase, the "hot start" mechanism well known in the art. also may be employed to initiate the reaction Real-time Array Assembly and Detection ~ The cartridge is placed into a magnetic field configuration designed to permit the formation of a random array of beads Beads are fust magnetically napped at the semiconductor surface and the reaction buffer

exchanged for assembly buffer, prev iously disclosed, preferred for the subsequent step an AC voltage (typically '- I Ypp, ^v ' 1 kHz) is applied to the electrodes and a spot on the substrate, defined by an aperture in the projection optics, is illuminated (typically with a power of 30 mW/mrn~ generated b> a 12W l 0OW Halogen Lamp), and a converging eleαrokinetie flow directed toward the illuminated spot is induced πcai the semiconductor surface Under the influence of both electrokinetic and magnetic-dipole- repuisive forces, beads gather in the illuminated iegion but remain separated from each other Finally, beads are "annealed" into a dense-packed ordered planar assembly Images are then recorded with a CCD camera (Apogee) In an alternative arrangement, the fluorescence signal associated with the open state of the looped probe may be detected by inserting the reaction mix into a flow cytometer which also peπnits decoding of the beads and hence deteiminatioπ of sequences corresponding to each assay signal

Example HL Homogeneous Binding Assay in Suspension Using Looped Probes Immobilized on Magnetic Beads

Looped probes were immobilized on color-encoded magnetic miciopartides ("beads") for use in a homogeneous binding assay Briefly, magnetic beads of -4 micron diameter were synthesized by standard methods and color-encoded as set forth in US Application Ko K* 348, 1 (>5, incorporated by reference Next, encoded beads were modified bv covalent attachment of Neutravidin to cpoxy groups on the beads to permit attachment of a "perfect -match (PM) ^" biotin> Sated looped probe, a "no-match (NM)" biotinyjated looped probe, and a biotinvlated positive control, in the form of a Cj 3-laheled oligonucleotide As in the previous examples, looped-probes contain a donor dye and an acceptor d\ e at their respective 5' and 3 ^" ends λliquots of probe-decorated, encoded magnetic beads were pooled in one test tube for determination of RNA target concentrations

Io determine the response of the probes, target RNAs were serial!) diluted (I 2) in reaction buffer (50 mM Tris (pH S 0), 0 I m\1 EDTA, 50 mM NaCl, 0 2%

Tween 20) and were then each incubated with an aliquot of pooled magnetic beads in a test tube Following incubation for IOmin at room temperature, a 0 5 μl aliquot of each

bead suspension was transferred - without washing - into an incubation chamber on a silicon wafer (fig. Ji) for image acquisition.

Trapping of magnetic beads was realized in a magnetic trap shown in Fig. S 2 A. This device comprises a bottom actuation element and a disposable top element that may host a channel system or a static reactor. In this example, it hosted an incubation chamber, as shown in Fig. 12A, which was formed by sandwiching 0.5 μl bead suspension droplet between a solid substrate and a 0.2-mm glass cover slip with 100- μm separation, and then by encapsulating the liquid phase with mineral oil. The magnetic actuator consists of a magnetic core, a coil, and high-permeability alloy layers that tune the field flux. In this particular embodiment, the device generates a magnetic field that is localized in a 1/8" circular region To form an array of magnetic beads, a typical current below 100 niA was sufficient to generate a flux density gradient exceeding by more than two orders of magnitude that of an untuned coil without significantly increasing the flux density (< 200 Gauss). Illustrated in Fig. 12B is the computed field distribution of quantities relevant to magneto-phoresis, namely, equipotential curves of ~tr. a quantity proportional to magneto-phoretic potential of an induced magnetic dipole moment, and vectors of its gradient, which is proportional to the relavent force The induced magnetic field induces the magnetic beads in suspension to migrate towards a substrate. Once in proximity to the solid support, the beads interact with each other repulsively and reorganize into arrays in the reaction buffer. The beads are in a random state before the magnetic field is turned on.

In this experiment, fol lowing incubation, bead suspension from each tube was transferred into the magnetic trap and, on activation, organized into arrays in accordance with the method described above. Optical interrogation was performed using fluorescence microscope (Nikon Eclipse ESOO). linage snapshots were taken through different optical filters, which are bright field, Cy 3 filter (F5, 500 ms), green filter (FS, 200 ms), and blue field (F5, 150ms), respectively. Images were processed using a Matlab code. Each single bead was identified and its corresponding Cy 3 intensity was then registered to its blue intensity, ϊn a "bead-map ^" (Fig. 13} plotted with Cy3 against blue, three clusters of beads can be seen and can be categorized to be Bl , B2. B3, from left to right, respectively. The Cy 3 intensity of B2 cluster indicates

the magnitude of RNA-hinding to the looped probes of specific type. After normalizing to the positive control (B 1 ) for each sample. The dose response of target interaction with specific and non-specific looped probes are summarized in Figure S 4, with error bars representing standard deviation of the mean intensities.

It should be understood that the terms, expressions and examples herein are exemplary only and not limiting, and that the scope of the invention is defined only in the claims which follow, and includes all equivalents of the subject matter of the claims.

">">