Concentration assessment

Abstract

Method for analyzing a sample Su, wherein sample Su comprises an unknown concentration CM_u of a ligand-specific monovalent binding species M and further comprises a ligand-specific multivalent binding species B, the method comprising the steps of an activated sensor surface A1L1 providing step actM comprising providing a sensor surface A1 having a ligand L1 immobilized thereon, wherein the ligand L1 can be monovalently bound by the ligand-specific monovalent binding species M and the ligand L1 can be multivalently bound by the ligand-specific multivalent binding species B, a calibration step calM yielding a monovalent calibration function FA1L1_M, recording in step mesM for the sample Su comprising the unknown concentration CM_u of the ligand-specific monovalent binding species M a sensorgram SGu using the activated sensor surface A1L1, selecting in step selM a response value rA1L1_M_u or the slope of the response value sA1L1_M_u from the sensorgram Su at a selection time tA1L1_M_s, retrieving in step retrM the concentration CM_u of the ligand-specific monovalent binding species M in the sample Su by calculating the concentration form the response value rA1L1_M_u or the slope of the response value sA1L1_M_u using the monovalent calibration function FA1L1_M.

Description

[0001] Concentration Assessment

[0002] Technical Field of the Invention

[0003] The present invention is located in the technical field of analysis of interactions between a target molecule and a ligand at a sensor surface, and more particularly to systems and methods for enabling detailed determination of concentration of certain species therein.

[0004] Background of the Invention

[0005] Analytical sensor systems that can monitor interactions between molecules, such as biomolecules, in real time are gaining increasing interest. Such systems typically enable the determination of binding, kinetics, affinity, specificity and / or concentration of molecules. Such molecules are often contained in samples, which comprise other components or molecules. It has turned out that for such application, optical biosensors are particularly useful. Such biosensors are usually referred to as interaction analysis sensors or bio-specific interaction analysis sensors. A representative of a system using a biosensor is the BIACORE™ instrumentation sold by Cytiva, which uses surface plasmon resonance (SPR) for detecting interactions between molecules at a sensing surface without any need for labels (cf. Figure 1). During passing a respective sample over the sensor surface, the progress of binding can be measured as a response value from the sensor. Thus, a direct reflection of the rate at which an interaction between molecules is occurring can be observed.

[0006] A typical output from systems such as the BIACORE™ system is a graph or curve describing the progress of a molecular interaction with time, including an association phase part and a dissociation phase part (cf. Figure 2). This binding curve, which is usually displayed on a computer screen, is often referred to as a “sensorgram” . With the BIACORE™ system (and analogous sensor systems) it is thus possible to determine in real time without the use of labelling, and often without purification of the substances involved, not only the presence and concentration of a particular molecule in a sample, but also additional interaction parameters, including kinetic rate constants for binding (association) and dissociation in the molecular interaction as well as the affinity for the interaction being assessed. The association rate constant (a) and the dissociation rate constant ( can be obtained by fitting the resulting kinetic data for one or preferably several different sample analyte concentrations to mathematical descriptions of interaction models in the form of differential equations. The affinity (expressed as the association equilibrium constant KA or the dissociation equilibrium constant KD) can be calculated from the association and dissociation rate constants.

[0007] Traditionally, calibration curves are created from multiple measurements of the same sample with different, known concentration of a specific target molecule. For each of these measurements, a sensorgram is recorded. The calibration curves are created by either plotting the response value after the analyte injection at a fixed time point against the concentration of the target molecule or by plotting the slope of the response at a fixed time point against the concentration of the target molecule (cf. Figure 4) .

[0008] Summary of the Invention

[0009] However, the problem with such calibration curves is that only the total concentration, i.e. of all species binding the ligand-activated sensor surface, can be measured and analyzed at once. However, in various processes, it is needed to measure the concentration of one of these specific target molecules in the presence of other target molecules. For example, in antibody production, a cell culture sample is analyzed, which comprises the antibody itself, potentially homodimers thereof, 1 Abs, free kappa and / or lambda light chains, etc. All of these components could potentially bind to the ligand-activated sensor surface (depending on the ligand immobilized on the sensor surface). Therefore, it is an object of the present invention to provide a method for analyzing the concentration of a ligand-specific monovalent binding species in the presence of a ligand-specific bivalent binding species in a sample.

[0010] It has now surprisingly found out that above-mentioned object can be achieved by a method for analyzing a sample Su, wherein sample Su comprises an unknown concentration CM_U of a ligand-specific monovalent binding species M and further comprises a ligand-specific multivalent binding species B, the method comprising the steps of: an activated sensor surface A 1LI providing step actM comprising providing a sensor surface A 1 having a ligand L1 immobilized thereon, wherein the ligand L1 can be monovalently bound by the ligand-specific monovalent binding species M and the ligand L1 can be multivalently bound by the ligand-specific multivalent binding species B, a calibration step calM yielding a monovalent calibration function FAILI_M, recording in step mesM for the sample Su comprising the unknown concentration CM_U of the ligand-specific monovalent binding species M a sensorgram SGUusing the activated sensor surface A 1LI , selecting in step selM a response value rAiLi_M_u or the slope of the response value SA ILI_M_U from the sensorgram Su at a selection time tAi _M_s, retrieving in step retrM the concentration CM_U of the ligand-specific monovalent binding species M in the sample Su by calculating the concentration form the response value fAiLi_M_u or the slope of the response value SAILI_M_U using the calibration function FAILI_M-

[0011] The advantage of the present invention is that in SPR measurements not only the total concentration of all species binding to the ligand on the sensor surface can be measured, but also a concentration of only a ligand-specific monovalent binding component comprised in the sample. It is in particular a specific advantage of the present invention that the concentration of only a ligand-specific monovalent binding component can be measured in the presence of a ligand-specific multivalent binding component. Thereby, the method of the present invention enables more detailed and faster analysis of samples in SPR measurements.

[0012] Definitions

[0013] The term ‘sensorgram’ as used herein denotes a graph or curve describing the progress of a molecular interaction in dependence of time. A typical sensorgram is depicted in Figure 2. Hence, a sensorgram comprises a base line (A), an association phase (B), optionally a plateau phase (c), a dissociation phase (D), and optionally a quitting point (E). The molecular interaction is typically represented by a response value of a sensor, preferably a biosensor. Such a response value can e.g. be recorded by a SPR measurement.

[0014] The term ‘ligand-specific monovalent binding species’ or ‘ligand-specific monovalent binding component’ as used herein denotes a specific target molecule analyzed by the method of the present invention. The present invention is based on the principle of specific binding between target molecules and ligands immobilized on a sensor. The ligand can have one or more specific binding sites, which may bind to one or more binding sites of a target molecule. The ligand-specific monovalent binding species is a target molecule, which forms only one of such bindings to the ligand on the sensor surface. It should be noted that the valency of the binding of the target molecule depends on the number and character of the binding sites provided by the ligand. Hence, the valency of a target molecule might vary over the type of ligand used. This, however, can also be advantageously used as outlined herein, in that with multiple measurements, specific concentrations can be calculated from differences of concentrations resulting from such multiple measurements. The term ‘ligand-specific multivalent binding species’ or ‘ligand-specific multivalent binding component’ as used herein denotes a target molecule, which forms at least two of the bindings to the ligand on the sensor surface as explained for the ligandspecific monovalent binding species.

[0015] The term ‘ligand-specific bivalent binding species’ or ‘ligand-specific bivalent binding component’ as used herein denotes a target molecule, which forms exactly two of the bindings to the ligand on the sensor surface as explained for the ligand-specific monovalent binding species.

[0016] The term ‘monovalent concentration’ as used herein denotes the concentration of all ligand-specific monovalent binding species present in the sample as analyzed in the method of the present invention.

[0017] The term ‘total concentration’ as used herein denotes the concentration of all ligandspecific monovalent binding species and all ligand-specific multivalent binding species present in the sample as analyzed in the method of the present invention.

[0018] The term ‘multivalent concentration’ as used herein denotes the concentration of all multivalent binding species present in the sample as analyzed in the method of the present invention.

[0019] The term ‘bivalent concentration’ as used herein denotes the concentration of all bivalent binding species present in the sample as analyzed in the method of the present invention.

[0020] The term ‘maximum response value’ as used herein denotes the maximum response value found in a sensorgram. Usually, the maximum response value is achieved towards the end of the association (B, cf. Figure 2), such as in the plateau phase (c). The maximum response value of a sensorgram does not necessarily need to be at the association-dissociation equilibrium. Usually, reaching such an equilibrium takes time and does not contribute to information gain. Hence, usually, the association phase is disrupted before reaching such equilibrium. However, in case of high dissociation rate constants, the maximum response value could also represent such an equilibrium.

[0021] The term ‘monovalent calibration curve’ as used herein denotes a graph or curve representing a plot of the response value or the slope of a response value from a sensorgram measured by a sensor comprising a ligand versus the concentration of the ligand-specific monovalent binding species, wherein the response values are selected at a fixed time located in the dissociation phase of the sensorgram, preferably are selected at a fixed time later than the time of the maximum of response values in the sensorgram. The term ‘monovalent calibration function’ as used herein denotes the function underlying the monovalent calibration curve, i.e. the function of the response value of a sensor comprising a ligand versus the concentration of the ligand-specific monovalent binding species.

[0022] The term ‘total calibration curve’ as used herein denotes a graph or curve representing a plot of the response value or the slope of a response value from a sensorgram measured by a sensor comprising a ligand versus the concentration of the ligand-specific monovalent binding species and ligand-specific multivalent binding species, wherein the response values are selected at a fixed time located after end of sample injection, in the dissociation phase of the sensorgram, or at a fixed time during the early association phase when slope is selected.

[0023] The term ‘total calibration function’ as used herein denotes the function underlying the total calibration curve, i.e. the function of the response value of a sensor comprising a ligand versus the concentration of the ligand-specific monovalent binding species and ligand-specific multivalent binding species.

[0024] The term ‘active sensor surface’ as used herein denotes a sensor surface, on which a ligand has been immobilized and which is ready for measurement.

[0025] Short Description of the Figures

[0026] Figure 1 is a schematic side view of a biosensor system based on SPR.

[0027] Figure 2 is a representative sensorgram showing detector response versus time for the interaction between an analyte and an immobilized binder for the analyte.

[0028] Figure 3 is a schematic representation of the emicizumab antibody as used in example IE1 .

[0029] Figure 4 is the total calibration curve of the emicizumab sample recorded on the Fc- ligand as used in IE1 .

[0030] Figure 5 is the monovalent calibration curve of the emicizumab sample recorded on the Fc-ligand as used in IE1.

[0031] Figure 6 is a schematic representation of the bsAb2 antibody as used in example IE2.

[0032] Figure 7 is the monovalent calibration curve of the bsAb2 sample recorded on the \4x-ligand as used in IE1. Detailed Description of the Invention

[0033] As indicated above, the present invention relates to a method and a system, which will be described in the following in more detail preceded by a general description of the sensor technology and the concentration determination in such technology.

[0034] Sensor technology

[0035] As mentioned above, the present disclosure relates to the evaluation of binding response data obtained for a target molecule at a plurality of concentrations. From this, one or more interaction parameters for the interaction may be determined. Typically, the experimental binding data is obtained by sensor-based technology, which studies the molecular interaction and presents the results in real time as the interactions progress. To aid understanding, some brief background concerning such sensor-based technology will now be provided.

[0036] Chemical sensors or biosensors (also referred to as ‘sample analysis systems’ herein) are typically based on label-free techniques, detecting a change in a property of a sensor surface, such as e.g. mass, refractive index, or thickness for the immobilized layer, but there are also sensors relying on some kind of labelling. Typical sensor detection techniques include, but are not limited to, mass detection methods, such as optical, thermo-optical and piezoelectric or acoustic wave methods (including e.g. surface acoustic wave (SAW) and quartz crystal microbalance (QCM) methods), and electrochemical methods, such as potentiometric, conductometric, amperometric and capacitance / impedance methods. With regard to optical detection methods, representative methods include those that detect mass surface concentration, such as reflection-optical methods, including both external and internal reflection methods, which are angle, wavelength, polarization, or phase resolved, for example evanescent wave ellipsometry and evanescent wave spectroscopy (EWS, or Internal Reflection Spectroscopy), both of which may include evanescent field enhancement via surface plasmon resonance (SPR), Brewster angle refractometry, critical angle refractometry, frustrated total reflection (FTR), scattered total internal reflection (STIR) (which may include scatter enhancing labels), optical wave guide sensors; external reflection imaging, evanescent wave-based imaging such as critical angle resolved imaging, Brewster angle resolved imaging, SPR-angle resolved imaging, and the like. Further, photometric and imaging / microscopy methods, “per se” or combined with reflection methods, based on for example surface enhanced Raman spectroscopy (SERS), surface enhanced resonance Raman spectroscopy (SERRS), evanescent wave fluorescence (TIRF) and phosphorescence may be mentioned, as well as waveguide interferometers, waveguide leaky mode spectroscopy, reflective interference spectroscopy (RlfS), transmission interferometry, holographic spectroscopy, and atomic force microscopy (AFR).

[0037] Commercially available biosensors include the afore-mentioned BIACORE™ system instruments, manufactured and marketed by Cytiva, Uppsala, Sweden, which are based on surface plasmon resonance (SPR) and permit monitoring of surface binding interactions in real time between a bound ligand and a target molecule of interest. In this context, a “ligand” is a molecule that has a known or unknown affinity for a given target molecule and includes any capturing or catching agent immobilized on the sensor surface, whereas “target molecule” includes any specific binding partner thereto that is typically injected and flows over or past the ligand on the sensor surface.

[0038] While in the detailed description that follows the present invention is illustrated in the context of SPR spectroscopy, and more particularly the BIACORE™ system, it is to be understood that the present invention is not limited to this detection method. Rather, any affinity-based detection method where a target molecule binds to a ligand immobilized on a sensing surface may be employed, provided that a change at the sensing surface can be measured which is quantitatively indicative of binding of the target molecule to the immobilized ligand thereon.

[0039] The phenomenon of SPR is well known, suffice it to say that SPR arises when light is reflected under certain conditions at the interface between two media of different refractive indices, and the interface is coated by a metal film, typically silver or gold. In the BIACORE™ instruments, the media are the sample and the glass of a sensor chip, which is contacted with the sample by a microfluidic flow system. The metal film is a thin layer of gold on the chip surface. SPR causes a reduction in the intensity of the reflected light at a specific angle of reflection. This angle of minimum reflected light intensity varies with the refractive index close to the surface on the side opposite from the reflected light, in the BIACORE™ system the sample side.

[0040] A schematic illustration of the BIACORE™ system is shown in Figure 1. Sensor chip 101 has a gold film 102 supporting capturing molecules (ligands) 103, e.g. antibodies, exposed to a sample flow with target molecules 104, e.g. an antigen, through a flow channel 105. Monochromatic p-polarised light 106 from a light source 107 (e.g. LED) is coupled by a prism 108 to the glass / metal interface 109 where the light is totally reflected. The intensity of the reflected light beam 110 is detected by an optical detection unit 111 (e.g. photodetector array).

[0041] A detailed discussion of the technical aspects of the BIACORE™ instruments and the phenomenon of SPR may be found in U.S. Patent No. 5,313,264. More detailed information on matrix coatings for biosensor sensing surfaces is given in, for example, U.S. Patent Nos. 5,242,828 and 5,436, 161. In addition, a detailed discussion of the technical aspects of the biosensor chips used in connection with the BIACORE™ instruments may be found in U.S. Patent No. 5,492,840. The above publications as well as any other publications, patent applications, patents, or other references mentioned in this disclosure are incorporated by reference in their entirety.

[0042] When target molecules in the sample bind to the capturing molecules (ligand) on the sensor chip surface, the concentration, and therefore the refractive index, at the surface changes and an SPR response is detected. Plotting the response against time during the course of an interaction will provide a quantitative measure of the progress of the interaction. Such a plot, or kinetic or binding curve (binding isotherm), is usually called a sensorgram, also sometimes referred to in the art as an “affinity trace" or “affinogram" . In the BIACORE™ system, the SPR response values are expressed in resonance units (RU). One RU represents a change of 0.0001 ° in the angle of minimum reflected light intensity, which for most proteins and other biomolecules corresponds to a change in concentration of about 1 pg / mm2on the sensor surface. As sample containing a target molecule contacts the sensor surface, the capturing molecule (ligand) bound to the sensor surface interacts with the target molecule in a step referred to as “association". This step is indicated on the sensorgram by an increase in response (RU) as the sample is initially brought into contact with the sensor surface. Conversely, “dissociation" normally occurs when the sample flow is replaced by, for example, a buffer flow. This step is typically indicated on the sensorgram by a drop in response (RU) over time as target molecule dissociates from the surface bound ligand.

[0043] A representative sensorgram (binding curve) for a reversible interaction at the sensor chip surface is presented in Figure 2. The sensorgram is representative of an interaction involving an immobilized capturing molecule (ligand), for example an antibody, interacting with a binding partner (target molecule) in a sample. The binding curves produced by biosensor systems based on other detection principles mentioned above will have a similar appearance. The vertical axis (y-axis) indicates the response (here in resonance units, RU) and the horizontal axis (x-axis) indicates the time (here in seconds). Initially, buffer is passed over the sensing surface giving the baseline response A in the sensorgram. During sample injection, an increase in signal is observed due to binding of the target molecule to the ligand. This part B of the binding curve is usually referred to as the “association phase". Eventually, a steady state condition is reached at or near the end of the association phase where the resonance signal plateaus at C (this state may, however, not always be achieved). It is to be noted that herein the term “steady state" is used synonymously with the term “equilibrium" (in other contexts the term “equilibrium" may be reserved to describe the ideal interaction model, since in practice binding could be constant over time even if a system is not in equilibrium). At the end of sample injection, the sample is replaced with a continuous flow of buffer and a decrease in signal reflects the dissociation, or release, of target molecule from the surface. This part D of the binding curve is usually referred to as the “dissociation phase". The analysis is optionally ended by a regeneration step where a solution capable of removing bound target molecule from the surface, while (ideally) maintaining the activity of the ligand, is injected over the sensor surface if the timelength to complete dissociation in buffer becomes unpractical. This is indicated in part E of the sensorgram. Injection of buffer restores the baseline A and the surface is now ready for a new analysis.

[0044] From the profiles of the association and dissociation phases B and D, respectively, information regarding the binding and dissociation kinetics is obtained. The height of the resonance signal at C represents affinity (the response resulting from an interaction being related to the change in mass concentration on the surface). This will now be explained in more detail below in the context of the various afore-mentioned approaches for determining interaction parameters of a target molecule-ligand interaction at the sensor surface.

[0045] General concentration determination

[0046] The time-dependent binding of two components A and B is expressed in formula (I). Thereby, it can be seen that the rate of the complex formation is proportional to the concentration of A and B, the rate of the dissociation is proportional to the concentration of AB. Thus, the association rate constant kais the proportionality constant for the association and the dissociation rate constant kd the proportionality constants for dissociation. Rates are measured in per second: the units of kaare per molar per second (M'1s'1), and the units of kd are per second (s-1)

[0047] In SPR measurements, the measured response R scales linearly with the complex concentration of the associated components [AB], Hence, the formula (I) can be changed to formula (II).

[0048] In formula (II), C is the concentration of the target molecule and Rmaxis the response signal at saturation of the target molecule on the (activated) sensor surface. Formula (II) thus describes the so-called net rate, i.e. the difference between the association and the dissociation. The net rate describes what can be observed during the association phase on the sensor surface. In that phase, both association and dissociation can be observed at the same time. During dissociation the analyte concentration is zero. Thus, also no association can occur. Consequently, during the dissociation phase only dissociation can be observed.

[0049] SPR experiments are usually conducted by injection of a certain volume sample having a certain concentration (either known or unknown) of the target molecule. After the end of such sample injection, the target molecule concentration is zero. Hence, before this point has been reached, usually association dominates and after this point has been reached dissociation dominates. In the latter situation, the dissociation rate depends on the complex concentration, now described by the response R, and the dissociation rate constant kd, cf. formula (III).

[0050] In the interpretation of SPR measurements, usually the terms ‘avidity’ and ‘affinity’ are used, which are explained in the following. Thereby, the term affinity as used herein denotes a single interaction between a target molecule A and a ligand on the surface B. It should be understood that a target molecule may be able to establish more than one interaction to ligands B on the surface, wherein each of these interactions has a specific affinity, which may vary according to the nature of the binding.

[0051] More precisely, affinity is defined as the equilibrium dissociation constant KD (with the unit of moles per liter, M). Equilibrium constants are defined in terms of the ratio of the concentration of components A and B and the concentration of complex AB at equilibrium, such as shown in formula (IV). Hence, the equilibrium constants are related to the kinetic rate constants. The set of formulas below show the relationship between the association and dissociation rate at equilibrium . Formula (V) describes the association rate and formula (VI) describes the dissociation rate. Formula (VII) describes the situation at equilibrium when the association rate equals the dissociation rate. The formulas (VIII) and (IX) show the relationship between the equilibrium constants and the rate constants. Finally, formula (X) shows that KA is the reciprocal of KD.

[0052] ^1= ka[A][B] (V)

[0053] ^l = -kd[AB] (VI)

[0054] / ca[A][B] = / cd[AB] (VII) r _ >

[0055] A~ [4][B] “ kd(VIII)

[0056] The term avidity as used herein on the other hand denotes the accumulated strength of multiple affinities of individual non-covalent binding interactions, such as between a target molecule A and a ligand on the surface B. The term avidity is also referred to as functional affinity.

[0057] It should be understood that avidity is not simply the sum of all affinities of interactions of a target molecule A with ligands on the surface B. The reason therefore is that individual binding events increase the likelihood of occurrence of other interactions (i.e. , increase the local concentration of each binding partner in proximity to the binding site). Hence, avidity should not be thought of as the mere sum of its constituent affinities but as the combined effect of all affinities participating in the biomolecular interaction.

[0058] Generally, higher avidity of target molecules to the ligand on the sensor surface decreases the dissociation rate.

[0059] Higher avidity can thus be achieved by increasing the number of interaction sites. The individual binding interactions do not have to have the same affinity or binding kinetics. However, generally, for example in a cell culture comprising produced antibodies and side products, the available binding sites are comparable, since antibodies are homodimers, mostly even identical.

[0060] Avidity from homodimers, homotrimers, or homotetramers forming the same types of bindings, will have identical dissociation (and association) rate constants per binding, but with different numbers of bindings per target molecule to the ligand-activated sensor surface. This is most evident when comparing monovalently and bivalently bound target molecules. In this case, the dissociation rate constant kd and the association rate constant (ka) remain the same for each binding of the bivalently (kdb, kab and monovalently (kdm, kam) bound target molecules. However, a decreased dissociation rate can be observed for the bivalently bound target molecule in comparison to the monovalently bound target molecule, resulting in kdm > kdb- the decreased dissociation rate is caused by the dissociation from two binding sites instead of one. Thus, kdb can be roughly described as kdm2. The present invention uses that finding to determine the concentration of ligand-specific monovalent binding species in the presence of ligandspecific multivalent binding species. Thereby, one basic assumption is that the dissociation rate of the bivalent / multivalent sample is close to zero and thus negligible. This is easily tested by injecting the same concentration of a monovalent and a bivalent / multivalent sample. If the bivalent / multivalent concentration has negligible dissociation, the bivalent / multivalent dissociation rate could be assumed to be close to zero. This will also be the most common situation since the typical range of biological interactions with regard to the dissociation is from around 10"5to 10'2s-1for dissociation rate constants.

[0061] To measure an unknown concentration of the ligand-specific monovalent binding species (unknown “monovalent concentration") in a sample, a calibration curve for the monovalent concentration or “monovalent calibration curve" needs to be created. Such a monovalent calibration curve can be created from multiple measurements of the same sample with different, known concentrations of a ligand-specific monovalent binding target molecule. For each of these measurements, a sensorgram is recorded. The calibration curves are created by either plotting the response value after the analyte injection has ended (i.e. during the dissociation phase) at a fixed time point against the concentration of the target molecule or by plotting the slope of the response after the analyte injection has ended (i.e. during the dissociation phase) at a fixed time point against the concentration of the ligand-specific monovalent binding target molecule. If response values are used to establish the monovalent calibration curve, the highest response value at the end of the sample injection (i.e. the association phase) is used to align the base lines of the sensorgrams.

[0062] Method of the present Invention

[0063] In the following the method of the present invention is explained in detail. a) Determination of the monovalent concentration of M

[0064] The most general embodiment of the present invention relates to a method for analyzing a sample Su, wherein sample Su comprises an unknown concentration CM_U of a ligand-specific monovalent binding species M and further comprises a ligandspecific multivalent binding species B, the method comprising the steps of: an activated sensor surface A 1LI providing step actM comprising providing a sensor surface A 1 having a ligand L1 immobilized thereon, wherein the ligand L1 can be monovalently bound by the ligand-specific monovalent binding species M and the ligand L1 can be multivalently bound by the ligand-specific multivalent binding species B, a calibration step calM yielding a monovalent calibration function FAILI_M, recording in step mesM for the sample Su comprising the unknown concentration CM_U of the ligand-specific monovalent binding species M a sensorgram SGUusing the activated sensor surface A 1LI , selecting in step selM a response value fAiLi_M_u or the slope of the response value SA ILI_M_U from the sensorgram Su at a selection time IA-IL-I_M_S, retrieving in step retrM the concentration CM_U of the ligand-specific monovalent binding species M in the sample Suby calculating the concentration form the response value fAiLi_M_u or the slope of the response value SAILI_M_U using the calibration function FAILI_M-

[0065] The difference between this method and the methods described in the prior art is that the calibration function is a monovalent calibration function. As defined earlier herein, the monovalent calibration function is prepared from response values of multiple sensorgrams having different known concentrations of the ligand-specific monovalent binding species, wherein the response values have been taken from the dissociation phase of the sensorgrams, i.e. correspond to a fixed time later than the time of the maximum response value. The specific advantage of this method is that the concentration of the ligand-specific monovalent binding species can be determined in the presence of ligand-specific multivalent binding species. The underlying principle enabling this measurement is that the dissociation rate of the ligand-specific multivalent binding species is significantly lower than the dissociation rate of the ligand-specific monovalent binding species.

[0066] It should be understood that the italic-set abbreviations used in the description of the present invention are not equations but only used to exactly define the terms used.

[0067] Furthermore, it should be understood that the ligand-specific monovalent binding species can be a single species or group of species. Hence, the step retrM can be used to determine the concentration of a single species, if the M is a single species. However, if this is not the case, but the concentration of single species is wanted to be determined by the method of the present invention, further steps as described later herein are necessary.

[0068] Hence, preferably, in the method according to the present invention the dissociation rate constant kdB of the ligand-specific multivalent binding species to the ligand is lower than the dissociation rate constant kdM of the ligand-specific monovalent binding species. More preferably, the dissociation rate of the ligand-specific multivalent binding species is close to zero and thus negligible. More preferably, the ratio kdM / kdB of the dissociation rate constant kdM of the ligandspecific monovalent binding species to the dissociation rate constant kdB of the ligandspecific multivalent binding species is more than 100, even more preferably more than 200, still even more preferably more than 500, and most preferably more than 1000. Usually, the ratio kdM / kdB is lower than 10000, preferably lower than 5000. This ensures that the difference in dissociation rates is high enough to assume that the dissociation rate of the ligand-specific multivalent binding species is close to zero and thus negligible

[0069] Also preferably, the dissociation rate constant kdM of the ligand-specific monovalent binding species is higher than 10'5, more preferably higher than 10'4, and most preferably higher than 5x1 O'4. Usually, the dissociation rate constant kdM of the ligandspecific monovalent binding species is lower than 10'1.

[0070] Preferably, the sample is present in liquid form, more preferably in solution, most preferably in aqueous solution, which may be optionally buffered. While the present disclosure exemplifies the method of the present invention by separation of antibodies, it generally can be used with any combination of target molecules having different numbers of binding sites and ligands, which can be immobilized to a sensor surface and which can monovalently and multivalently bind such target molecules. Thereby, the interactions between the target molecules and the ligands can be covalent interactions, antibody epitope interactions, hydrophobic interactions, electrostatic interactions, dipole-dipole interactions, and hydrogen bonding interactions. Most preferably, the interaction between the target molecules and the ligands is an antibody epitope interaction.

[0071] The sample can comprise other components besides the ligand-specific multivalent binding species and the ligand-specific monovalent binding species. Such other components are ligand-specific non-binding species and as such do not create any change in response in the measurement. Hence, preferably, the sample Suis prepared from a mixture derived from a reaction, more preferably a mixture from a bioreactor optionally further comprising cells, and / or solvents.

[0072] In a preferred embodiment of the present invention, the ligand-specific monovalent binding species M and / or the ligand-specific multivalent binding species B independently from each other are substances naturally occurring in animal or plant tissue, are produced by naturally occurring microorganisms or enzymes, and / or are produced by modified, preferably genetically modified, microorganisms or enzymes. Thus, the method of the present invention can be used to analyze samples from naturally occurring mixtures of biomolecules, or samples from mixtures of biomolecules produced by either natural processes including enzymes or microorganisms or by human-designed processes including enzymes or microorganisms.

[0073] As such, the ligand-specific monovalent binding species M and / or the ligand-specific multivalent binding species B independently from each other are preferably selected from the list consisting of antibodies, proteins, viruses, organic substances, DNA molecules, RNA molecules, shorter nucleotides molecules, exosomes, cells (such as used in cell therapy), or specific target molecules.

[0074] In the most preferred embodiment of the present invention, the ligand-specific monovalent binding species M and / or the ligand-specific multivalent binding species B independently from each other are antibodies or fragments of antibodies, preferably antibodies or fragments of antibodies both produced by the same cell type.

[0075] The ligand as used in the method of the present invention can be chosen depending on the interaction between the target molecule and the ligand.

[0076] Ligands for covalent coupling interactions can be selected from carboxylic ligands, amine ligands, and gold ligands. Ligands for hydrophobic interactions can be selected from liposomes and hydrophobics, i.e. long chain alkanes.

[0077] Most preferably, the ligands used in the method according to the present invention are ligands for antibody epitope interactions. In antibody epitope interactions, the ligands used provide respective binding sites for specific binding sites to be commonly found on antibodies.

[0078] Antibodies are proteins having a molecular weight of about 150 kDa and a size of about 10 nm. Generally, they are structured in three globular regions thereby forming a Y shape. Human and mammalian antibodies usually consist of four polypeptide chains: two identical heavy chains and two identical light chains, which are connected by disulfide bonds. [

[0079] Each chain consists of a series of domains, wherein domains are comparable sequences of amino acids (about 110 amino acids per domain). Light chains consist of one variable domain VL and one constant domain CL, while heavy chains contain one variable domain VH and three to four constant domains CH1, CH2, CH3, [CH4].

[0080] Two arms of the Y-shape represent two antigen-binding fragments (Fab), each of which usually comprises a VL, a VH, a CL, and a CH1 domain. The third arm of the Y shape is the crystallizable fragment (Fc).

[0081] Potential binding sites of an antibody are each Fc-region and each Fv region, wherein the latter provides a binding site of the variable domain of the heavy chain VH and the variable domain of the light chain VL. Hence, three structurally differing binding sites are offered by an antibody: the Fc-region, the ^-domain and the \4-domain. However also CL and CH1 to CH4 domains can be used as potential binding sites.

[0082] It should be noted that depending on the origin of the antibody in question, not each of these binding sites must be present or functional. Hence, in particular in recombinant produced variants of antibodies it is common practice to either modify the structure of one of the fragments or domains to not provide any of the binding sites at all or to modify the binding sites in a way that they do not bind to common receptors anymore.

[0083] Thus, ligands for antibody capture coupling preferably comprise molecules which provide binding sites for the Fc-region, the ^-domain, the \4-domain, CL domain, CH 1 to CH4 domains, or mixtures thereof. More preferably ligands for antibody capture coupling comprise molecules which provide binding sites for the Fc-region, the \ / H- domain, the \4-domain, or mixtures thereof.

[0084] An example of such a ligand is staphylococcal protein A (SpA), containing domains capable of binding to the Fc-regions of IgG immunoglobulins from different species. These domains are commonly denoted as the E-, D-, A-, B- and C-domains. Hence, preferably, ligands for antibody capture coupling comprise molecules which provide binding sites for the Fc-region as used in the method according to the present invention are SpA-based proteins. SpA-based proteins are commercially available, for example, in the form of MabSelect™ SuRe, MabSelect™ SuRe LX, MabSelect™ PrismA protein A and HiScreen Fibro™ PrismA from Cytiva, Uppsala, Sweden. Specific preferred ligands are described in WO 2023 / 046886 A1.

[0085] Ligands with binding sites for the domain

[0086] An example of such a ligand is also staphylococcal protein A (SpA), also containing domains capable of binding to the Fab-regions of IgG immunoglobulins from different species. Hence, preferably, ligands for antibody capture coupling comprise molecules which provide binding sites for the ^-domain as used in the method according to the present invention are SpA-based proteins. SpA-based proteins are commercially available, for example, in the form of MabSelect™ SuRe, MabSelect™ SuRe LX, MabSelect™ PrismA protein A and HiScreen Fibro™ PrismA from Cytiva, Uppsala, Sweden.

[0087] Specific preferred Protein A and Protein A-derived ligands have binding affinity for VH- domains are described in WO 2023 / 174900 A1. Ligands with binding sites for the domain

[0088] For antibody fragments, such as Fab, single-chain variable fragments (scFv), bispecific T-cell engagers (BiTEs), domain antibodies etc., which lack the Fc-region but have a kappa light chain subclass 1 ,3 or 4, ligands comprising Protein L derived from Peptostreptococcus magnus are preferably used. Specific preferred Protein L derived ligands are described in WO 2023 / 030995 A1.

[0089] Protein L matrices are commercially available as for instance Capto™ L from Cytiva™.

[0090] The calibration step for identification of the monovalent calibration function FA ILI_M involves at least two measurements of a sensorgram of samples having different known concentrations of the ligand-specific monovalent binding species M and respective linear fitting of a linear function through these two data points, thereby yielding the monovalent calibration function FA ILI_M- However, ideally, the measurement is carried out multiple times to achieve higher accuracy.

[0091] Hence, in a preferred embodiment of the method of the present invention, the calibration step calM comprises the following steps:

[0092] ■ rrimax times recording in step cal_mesM(m) a sensorgram SG(m) for a sample Sk(m) comprising a known concentration CM(m) of the ligand-specific monovalent binding species M using the activated sensor surface A 1LI , wherein the sensorgram SG(m) comprises response values rAiLi_M(m) related to the time relative to the start time of the recording step, wherein CM(m) differs from all concentrations in the range of from CM(1) to CM(mmax) except for CM(m), wherein m is a natural number in the range of from 1 to mmax, and wherein mmax is 2 or higher;

[0093] ■ aligning in step cal_algM the sensorgrams SG(1) to SG(mmax) at the maximum response value,

[0094] ■ extracting in step cal_extrM the response value rAiLi_M(m) or the slope of the response value SA IL I_M(ITI) from each sensorgram SG(m) at a time point subsequent to the time corresponding to the maximum response value and providing a datapoint d(m) comprising the response value rAiLi_M(m) or the slope of the response value SA IL I_M(ITI) and the corresponding known concentration CM(m) of the ligand-specific monovalent binding species M yielding mmax datapoints, and ■ fitting in step cal_fitM the monovalent calibration function FAILI_M from the mmax datapoints.

[0095] In the aligning step cal_algM, the sensorgrams SG(1) to SG(mmax) are aligned in that the response value having the maximum value is set to zero. Usually, the maximum response value denotes the start of the dissociation phase. Hence, the sensorgrams SG(1) to SG(mmax) are aligned so that the start of the dissociation phase for all sensorgrams is at the same absolute response value, i.e. preferably at zero. This has the advantage that response values taken from the sensorgrams after the maximum response value can be used to prepare the monovalent calibration curve representing the dissociation behavior in dependency of the concentration of the ligand-specific monovalent binding species. Hence, in case the step cal_extrM is carried out by extracting the slope of the response value SA IL I_M(ITI) from each sensorgram SG(m), the aligned step cal_algM is not necessarily needed.

[0096] Preferably, in the method according to the present invention, mmax is 3 or higher, preferably 5 or higher, more preferably 10 or higher. This has the advantage of higher precision of the relationship between concentration of the ligand-specific monovalent binding species and the response value. Usually, mmax is not higher than 30.

[0097] As discussed earlier, the measurement step cal_mesM(m) in the calibration step calM is carried out by injecting a sample comprising known concentrations CM(m) of the ligand-specific monovalent binding species (target molecule) leading to association of the target-molecule to the ligand on the sensor surface and measuring the response value. Furthermore, after the sample has run out, the measurement is continued merely with solvent and buffer, which leads to dissociation of the ligand-associated target molecule. Hence, step cal_mesM(m) comprises measurement of the association and dissociation phase. Thus, preferably, each step cal_mesM(m) of the method according to the invention comprises the steps cal_mes_assM(m) and cal_mes_dissM(m), wherein step cal_mes_assM(m) comprises measuring the response values rM_ass(m) of the activated sensor surface A 1LI during contacting a solution comprising a known concentration CM(m) of the at least one monovalent binding species M and not comprising any of the at least one multivalent binding species B with the activated sensor surface A 1LI yielding an associated activated sensor surface A lLi_M_ass(m) until a maximum response value of the at least one monovalent binding species M to the activated sensor surface A 1LI has been reached, step cal_mess_dissM(m) comprises measuring the response values rAiLi_M_diss(m) of the associated activated sensor surface A 1Li_M_ass(m) during contacting a solution not comprising any of the at least one monovalent binding species and not comprising any of the at least one multivalent binding species with the associated activated sensor surface A 1Li_M_ass(m) thereby yielding a dissociated activated sensor surface A 1i_i_diss(m) and adding the response values rAiLi_M_diss(m) to the response values rAiLi_M_ass(m) yielding the total response values rAiLi_M(m).

[0098] It should be noted that to enhance the accuracy of the calibration, each concentration of the ligand-specific monovalent binding species can be measured more than one time resulting in more than one sensorgram SG(m) per concentration Civi(m).

[0099] Furthermore, in another preferable embodiment of the method according to the present invention, the calibration step calM can be carried out more than one time. For example, it can be advantageous to record two calibration curves: one before the measurement step mesM and one after the measurement step mesM has been finished. In such a method, effects of, e.g., ligands loosing binding activity during measurement can be compensated. In such case samples measured more towards the end are compared against the second calibration curve. The second calibration has furthermore the advantage to be a good quality indicator. In such case, the two calibration curves are overlayed. The more the two calibration curves equal each other, the better the used method and thus the results are.

[0100] An example for a monovalent calibration function FAILI_M is depicted in Figure 5 and 7. As the calibration function represents the dissociation phase, the response value decreases in line with complex concentration. As explained above, the change of response value in this function can be attributed to the dissociation of the ligandspecific monovalent binding species only, as the ligand-specific multivalent binding species is not dissociating. Hence, the change can be used to determine the concentration of the ligand-specific monovalent binding species independently of the presence and the amount of ligand-specific multivalent binding species in the sample.

[0101] Any conceivable ligand-specific monovalent binding species and ligand combination can be used in the method of the present invention as long as the respectively ligandspecific multivalent binding species does not dissociate or only very slowly dissociates in comparison to the ligand-specific monovalent binding species.

[0102] However, preferably, the interaction between the ligand and the ligand-specific monovalent binding species is as described in the following.

[0103] Preferably, the ligand L 1 has a binding site LBSM- Also preferably, the ligand-specific monovalent binding species M has a binding site MBSLI bindable to LBSM and the ligand-specific monovalent binding species M has no further binding site bindable to L1. This ensures, that only a monovalent binding between the ligand and the ligandspecific monovalent binding species can be formed.

[0104] Furthermore, preferably, the ligand-specific multivalent binding species B has a binding site BBSLI bindable to the ligand L1 and a binding site BBSL2 bindable to the ligand L1. This ensures that the ligand-specific multivalent binding species B forms at least two bindings to the ligand L1. It should be noted that it is not necessary for establishing two bindings to the ligand-specific multivalent binding species B, as the number of ligand molecules in comparison to the number of target molecules is usually very high. Hence, it may be sufficient if the ligand has one binding site LBSB, which is able to bind to both, BBSLI and BBSL2- Thereby, BBSLI and BBSL2 can be structurally identical (BBSLI = BBSL2) or structurally different binding sites (BBSLI t BBSL2\ If BBSLI and BBSL2 are structurally identical, it is sufficient that the ligand L1 had one binding site LBSB, which is able to bind to both BBSLI and BBSL2- If BBSLI and BBSL2 are structurally identical, the ligand L1 preferably has a binding site LBSBL and a binding site LBSB2, wherein binding site BBSLI is bindable to LBSBL and binding site BBSL2 is bindable to LBSBL- In an alternative, less preferred embodiment, if BBSLL and BBSL2 are structurally identical, two ligands L1a and L1b are present on the activated sensor surface, wherein L1a has binding site BBSLL is bindable to LBSBL and L1b has binding site BBSL2 is bindable to LBSBL-

[0105] In a preferred embodiment, the target molecule is an antibody and MBSL, BBSLL, and / or BBSL2 are selected from a Fc-region, a ^-domain, a \4-domain, a CL domain, or CH1 to CH4 domains. Respectively, preferably, the ligand L1 is a protein and the binding sites LBSM, LBSB, LBSBL, LBSB2 independently from each other are epitopes for these respective Fc-region, ^-domain, \4-domain, a CL domain, or CH1 to CH4 domains.

[0106] More preferably, the target molecule is an antibody and MBSL, BBSLL, and / or BBSL2 are selected from a Fc-region, a ^-domain, or a \4-domain. Respectively, more preferably, the ligand L1 is a protein and the binding sites LBSM, LBSB, LBSBL, LBSB2 independently from each other are epitopes for these respective Fc-region, VH- domain, or \4-domain. b) Determination of the total concentration of M and B

[0107] In a preferred embodiment of the method of the present invention not only the monovalent concentration is determined, but also the total concentration. Thus, preferably, in the method according to the present invention, the concentration CMB_U in sample Su of the ligand-specific monovalent binding species M and the ligand- specific multivalent binding species B is unknown and the method further comprises the steps of: an activated sensor surface A2L2 providing step actMB comprising providing a sensor surface A2 having a ligand L2 immobilized thereon, wherein the ligand L2 can be bound by the ligand-specific monovalent binding species M and the ligand L2 can be bound by the ligand-specific multivalent binding species B, a calibration step CSIMB yielding a total calibration function FA2L2_MB, recording in step mesMB for the sample Su comprising the unknown concentration CMB_U of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species B a sensorgram SGUusing a sensor surface A2, selecting in step SGIMB a maximum response value fA2L2_MB_u or the slope of a response value not being the maximum response value SA2L2_MB_U from the sensorgram Su, retrieving in step retrMB the total concentration CMB_U of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species B in the sample Su by calculating the concentration from the response value fA2L2_MB_u or the slope of the response value SA2L2_MB_U using the total calibration function FA2L2_MB-

[0108] It should be understood that in step SGIMB, the slope SA2L2_MB_U of a response value not being the maximum response value can be selected. Preferably, the slope SA2L2_MB_U is selected at a response value at an early report point, more preferably at a response value in the range of from 15 to 20 s after start of sample injection. The slope at the maximum response value will be low, as the surface usually starts to saturate. This slows down the rate of the binding significantly and make the slope less suitable for being used in the method according to the present invention. Alternatively, the maximum response value is selected, wherein the maximum response value usually appears after the end of the sample injection.

[0109] The preferred embodiments as defined for ligand L1 are equally eligible ligand L2. Ligand L1 as used for the determination of the monovalent calibration function and ligand L2 used for the determination of the total calibration function can be identical or different. The choice of ligand depends on the species to be determined. Different ligands bind different species or groups of species. Hence, in one preferred embodiment, ligand L 1 and L2 are identical. In this case the method according to the present invention may further comprises the steps of: calculating in step retrB the concentration CB_U using the following formula (XI)

[0110] CB_U= CMBU~CM_U (XI).

[0111] It should be noted that the ligand-specific monovalent binding species or the ligandspecific multivalent binding species can be groups of species. Hence, the step retrB can be used to determine the concentration of a single species, if the difference of species between B and M gives a single species. However, if this is not the case, but the concentration of single species are wanted to be determined by the method of the present invention, further steps as described later herein are necessary.

[0112] In accordance with the calibration step used for determining the monovalent calibration function FAILI_M, the calibration step for identification of the total calibration function FA2L2_MB involves at least two measurements of a sensorgram of samples having different known concentrations of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species MB and respective linear fitting of a linear function through these two data points, thereby yielding the total calibration function FA2L2_MB- However, ideally, the measurement is carried out multiple times to achieve higher accuracy.

[0113] Hence, in the method according to the present invention, the calibration step CSIMB comprises the following steps:

[0114] ■ nmax times recording in step cal_mesMB(n) a sensorgram SG(n) for a sample Sk(n) comprising a known concentration CMB(n) of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species B using a sensor surface A2, wherein the sensorgram SG(n) comprises response values fA2L2_MB(n) related to the time relative to the start time of the recording step, wherein CMB(n) differs from all concentrations in the range of from CMB(1) to CMB(nmax) except for CMB(II), wherein n is a natural number in the range of from 1 to nmax, and wherein fimax is 2 or higher;

[0115] ■ aligning in step cal_algMB the sensorgrams SG(1) to SG(nmax) at the base line after the start time and before any increase of the response value rA2L2_MB(n),

[0116] ■ extracting in step cal_extrMB a response value rA2L2_MB(n) or the slope of a response value SA2L2_MB(D) at a fixed time before or at the maximum response from each sensorgram SG(n) providing a datapoint d(n) comprising the response value fA2L2_MB(n) or the slope of the response value SA2L2_MB(n) and the corresponding known concentration CMB(D) of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species B yielding fimax datapoints, and

[0117] ■ fitting in step cal_fitMB the total calibration function FA2L2_MB from the nmax datapoints.

[0118] In the aligning step cal_algMB, the sensorgrams SG(1) to SG(nmax) are aligned in that the response value having the lowest value is set to zero. Usually, the lowest response value denotes the initial baseline of the measurement and thus the start of the association phase. Thus, the baseline of the measurement is preferably set to zero. Hence, the sensorgrams SG(1) to SG(nmax) are aligned so that the start of the association phase for all sensorgrams is at the same absolute response value, i.e. preferably at zero. This has the advantage that response values taken from the sensorgrams at a fixed time before or at the maximum response value or at an early point at the beginning of the association, more preferably at a response value in the range of from 5 to 30 s after start of sample injection, can be used to prepare the total calibration curve representing the association behavior in dependency of the concentration of the ligand-specific monovalent binding species and the ligandspecific multivalent binding species. Hence, in case the step cal_extrMB is carried out by extracting the slope of the response value SA2L2_MB(D) from each sensorgram SG(n), the aligned step cal_algMB is not necessarily needed.

[0119] Preferably, in the method according to the present invention, nmaxis 3 or higher, preferably 5 or higher, more preferably 10 or higher. Usually, nmaxis not higher than 30.

[0120] As discussed earlier, the measurement step cal_mesMB(n) in the calibration step calMB is carried out by injecting a sample comprising known concentrations CMB(n) of the ligand-specific multivalent binding species and ligand-specific monovalent binding species (target molecule) leading to association of the target-molecule to the ligand on the sensor surface and measuring the response value. Furthermore, after the sample has run out, the measurement can be continued merely with solvent and buffer, which leads to dissociation of the ligand-associated target molecule. Hence, step cal_mesMB(n) comprises measurement of the association and optionally of a dissociation phase. Hence, preferably, in the method according to the present invention, each step cal_mesMB(n) comprises the step cal_mes_assMB(n), wherein step cal_mes_assMB(n) comprises measuring the response values rMB_ass(n) of the activated sensor surface A2L2 during contacting a solution comprising a known concentration CMB(H) of the at least one monovalent binding species M and the at least one multivalent binding species B with the activated sensor surface A2L2 yielding an associated activated sensor surface A2i_2_MB_ass(n) until a maximum response value of the at least one monovalent binding species M and the at least one multivalent binding species B to the activated sensor surface A2L2 has been reached or before a maximum response value of the at least one monovalent binding species M and the at least one multivalent binding species B to the activated sensor surface A2L2 has been reached, preferably before a maximum response value of the at least one monovalent binding species M and the at least one multivalent binding species B to the activated sensor surface A2L2 has been reached.

[0121] An example for a total calibration function FA2L2_MB is depicted in Figure 4. As the calibration function represents the association phase, the response value increases in line with complex conentration. As explained above, the change of response value in this function can be attributed to the association of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species B, as both species associate with the ligand L2.

[0122] Usually, the dissociation phase is not of interest for the measurement of the total concentration, as it is not needed for its calculation. Hence, preferably, each step cal_mesMB(n) does not comprise the step cal_mes_dissMB(n) after the step cal_mes_assMB(n). This increases the speed of the method.

[0123] However, if a bi- / multivalent calibrant is used to set up the calibration curve and an unknown sample with both entities is analyzed, the concentration of the bi- / multivalent can be measured after the dissociation of the monovalent component. In such a case, preferably, each step cal_mesMB(n) comprises the step cal_mes_dissMB(n) after the step cal_mes_assMB(n), wherein step cal_mes_dissMB(n) comprises measuring the response values rA2L2_MB_diss(n) of the associated activated sensor surface A2i_2_MB_ass(n) during contacting a solution not comprising any of the at least one monovalent binding species and not comprising any of the at least one multivalent binding species with associated activated sensor surface A2i_2_MB_ass(n) thereby yielding a dissociated activated sensor surface A2i_2_diss(n) and adding the response values rA2L2_MB_diss(n) to the response values rA2L2_MB_ass(n) yielding the total response values rA2L2_MB(n).

[0124] Preferably, the ligand L2 has the same properties as the ligand L 1. Hence, preferably, the ligand L2 has a binding site LBSM- Also preferably, the ligand-specific monovalent binding species M has a binding site MBSL2 bindable to LBSM and the ligand-specific monovalent binding species M has no further binding site bindable to L2. This ensures, that only a monovalent binding between the ligand and the ligand-specific monovalent binding species can be formed.

[0125] Furthermore, preferably, the ligand-specific multivalent binding species B has a binding site BBSLI bindable to the ligand L2 and a binding site BBSL2 bindable to the ligand L2. This ensures that the ligand-specific multivalent binding species B forms at least two bindings to the ligand L2. It should be noted that it is not necessary for establishing two bindings to the ligand-specific multivalent binding species B, as the number of ligand molecules in comparison to the number of target molecules is usually very high. Hence, it may be sufficient if the ligand L2 has one binding site LBSB, which is able to bind to both, BBSLI and BBSL2. Thereby, BBSLI and BBSL2 can be structurally identical (BBSLI = BBSL2) or structurally different binding sites (BBSLI t BBSL2\ If BBSLI and BBSL2 are structurally identical, it is sufficient that the ligand L2 had one binding site LBSB, which is able to bind to both BBSLI and BBSL2- If BBSLI and BBSL2 are structurally identical, the ligand L2 preferably has a binding site LBSBL and a binding site LBSB2, wherein binding site BBSLI is bindable to LBSBL and binding site BBSL2 is bindable to LBSBL- In an alternative, less preferred embodiment, if BBSLL and BBSL2 are structurally identical, two ligands L2a and L2b are present on the activated sensor surface, wherein L2a has binding site BBSLL is bindable to LBSBL and L2b has binding site BBSL2 is bindable to LBSBL-

[0126] In a preferred embodiment, the target molecule is an antibody and MBSL, BBSLL, and / or BBSL2 are selected from a Fc-region, a ^-domain, a \4-domain, a CL domain, or CH1 to CH4 domains. Respectively, preferably, the ligand L2 is a protein and the binding sites LBSM, LBSB, LBSBL, LBSB2 independently from each other are epitopes for these respective Fc-region, ^-domain, \4-domain, a CL domain, or CH1 to CH4 domains.

[0127] In a more preferred embodiment, the target molecule is an antibody and MBSL, BBSLL, and / or BBSL2 are selected from a Fc-region, a ^-domain, or a \4-domain. Respectively, more preferably, the ligand L2 is a protein and the binding sites LBSM, LBSB, LBSBL, LBSB2 independently from each other are epitopes for these respective Fc-region, ^-domain, or \4-domain.

[0128] Preferably, LBSM and LBSBL are the same binding sites. More preferably, LBSM, LBSBL, and LBSB2 are the same binding sites.

[0129] Most preferably, ligands L1 and L2 have the same molecular structure. c) Determination of the total concentrations of B1 and B2 As explained above, the situation might arise that depending on the ligand the ligandspecific multivalent binding species B comprises a group of species. However, certain applications might require that the concentration of single ligand-specific multivalent binding species should be determined. This could be achieved by a ligand, which selectively binds only bivalently to the species in question. However, not always such a ligand-species combination is achievable. However, the problem could be solved by a ligand L3, which binds differently to the ligand-specific multivalent binding species B than ligand L2. Hence, if the total concentration is determined with ligands L2 and subsequently with L3, the difference in concentration measured with L2 and L3 can be attributed to the species bivalent binding to L2, but not L3 or vice versa.

[0130] Hence, preferably, in the method according to the present invention the ligand-specific multivalent binding species B comprises an unknown concentration CBI_U of a first ligand-specific multivalent binding species B1 and an unknown concentration CB2_U of a second ligand-specific multivalent binding species B2 and the method further comprises the steps: an activated sensor surface A3L3 providing step actMBi comprising providing a sensor surface A3 having a ligand L3 immobilized thereon, wherein the ligand L3 can be bound by the ligand-specific monovalent binding species M and the ligand L3 can be bound by the first ligand-specific multivalent binding species B1, but the ligand L3 cannot be bound by the second ligandspecific multivalent binding species B2; a calibration step CBIMBI yielding a total calibration function FA3L3_MBI, recording in step mesMBi for the sample Su comprising the unknown concentration CB I_U of the first ligand-specific multivalent binding species B1 a sensorgram SGUusing a sensor surface A3, selecting in step SGIMBI a response value rA3L3_MBi_u or the slope of a response value SA3L3_MB I_U from the sensorgram Su, and retrieving in step retrMBi the total concentration CMBI_U of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species B in the sample Su by calculating the concentration from the response value rA3L3_MBi_u or the slope of the response value SA3L3_MBI_U using the total calibration function FA3L3_MBI.

[0131] For the same reasons as set out for the determination of the total calibration function

[0132] FA2L2_MB , also in the determination of the total calibration function FA3L3_MBI at least two samples comprising known concentration CMBI ( of the ligand-specific monovalent binding species M and the first ligand-specific multivalent binding species B 1 are measured. Hence, preferably, in the present invention, the step CBIMBI comprises the following steps:

[0133] ■ fmax times recording in step cal_mesMBi(f) a sensorgram SG(f) for a sample Sk(f) comprising a known concentration CMBi(f) of the ligand-specific monovalent binding species M and the first ligand-specific multivalent binding species B 1 using a sensor surface \3, wherein the sensorgram SG(f) comprises response values rA3L3_MBi(f) related to the time relative to the start time of the recording step, wherein CMBI ( differs from all concentrations in the range of from CMB I(1) to CMBi (fmax) except for CMBi(f), wherein f is a natural number in the range of from 1 to fmax, and wherein fmax i s 2 or higher;

[0134] ■ aligning in step cal_algMBi the sensorgrams SG(1) to SG(fmax) at the base line after the start time and before any increase of the response value rA3L3_MBi(f),

[0135] • extracting in step cal_extrMBi a response value fA3L3_MBi(f) or a slope of the response value SA3L3_MBi(f) at a fixed time before or at the maximum response from each sensorgram SG(f) providing a datapoint d(f) comprising the response value rA3L3_MBi(f) or the slope of the response value SA3L3_MBi(f) and the corresponding known concentration CMB I( of the ligand-specific monovalent binding species M and the first ligand-specific multivalent binding species B 1 yielding fmax datapoints, and

[0136] ■ fitting in step cal_fitMBi a calibration function FA3L3_MB I from the fmax datapoints.

[0137] Preferably, the method of the present invention further comprises the steps of: calculating in step retrMBi the concentration CB2_U using the following formula (XII)

[0138] CB2_U= CMB_U ~CMB1_U (XII).

[0139] Hence, if ligands L2 and L3 have been chosen to differ only in one ligand-specific multivalent binding species B2, the concentration thereof can be determined respectively.

[0140] For the same reasons as set out for the determination of the total calibration function FA2L2_MB , also each step cal_mesMBi(f) comprises the step cal_mes_assMBi(f) and optionally step cal_mes_dissMBi(f), wherein step cal_mes_assMBi(f) comprises measuring the response values rMBi_ass(f) of the activated sensor surface A3L3 during contacting a solution comprising a known concentration CMB I( of the at least one monovalent binding species M and the first multivalent binding species B 1 with the activated sensor surface A3L3 yielding an associated activated sensor surface A3i_3_MBi_ass(f) until a maximum response value of the at least one monovalent binding species M and the first multivalent binding species B 1 to the activated sensor surface A3L3 has been reached.

[0141] Preferably, step cal_mes_dissMBi(f) comprises measuring the response values rA3L3_MBt_diss(f) of the associated activated sensor surface A3i_3_MBi_ass(f) during contacting a solution not comprising any of the at least one monovalent binding species and not comprising any of the at least one multivalent binding species with associated activated sensor surface A3i_3_MBi_ass(f) thereby yielding a dissociated activated sensor surface A3i_3_diss(f) and adding the response values rA3L3_MBi_diss(f) to the response values rA3L3_MBi_ass(f) yielding the total response values rA3L3_MBi(f.

[0142] Preferably, fmax is 3 or higher, more preferably 5 or higher, especially preferably 10 or higher.

[0143] The general properties as set out for ligand L2 also apply for ligand L3. However, ligands L2 and L3 differ in that the binding to the ligand-specific multivalent binding species and / or to the ligand-specific monovalent binding species differs such that the ligand-L2-specific multivalent binding species and / or the ligand-L2-specific monovalent binding species comprise a different number of species than the ligand - L2-specific multivalent binding species and / or the ligand-L2-specific monovalent binding species. Preferably the difference is exactly one species.

[0144] Hence, preferably, the ligand L3 has a binding site LBSBI_I and / or a binding site LBSBI_2 the first ligand-specific multivalent binding species B 1 has a binding site B 1BSL bindable to LBSBI_I and a binding site BBSL2 bindable to LBSBI_2, the ligand L3 has no binding site LBSB2 and / or the second ligand-specific multivalent binding species B2 has no binding site B2BSL bindable to LBSB2-

[0145] More preferably, the ligand L3 additionally has a binding site LBSM, the ligand-specific monovalent binding species M has a binding site MBSL bindable to LBSM, the ligand-specific monovalent binding species M has no further binding site bindable to L3.

[0146] Preferably, L1 and L3 have a different molecular structure and / or L2 and L3 have a different molecular structure. d) Determination of the monovalent concentrations of M1 and M

[0147] As explained above, another situation might arise that depending on the ligand the ligand-specific monovalent binding species M comprises a group of species. However, certain applications might require that the concentration of single ligand-specific monovalent binding species should be determined. This could be achieved by a ligand, which selectively binds only monovalently to the species in question. However, not always such a ligand-species combination is achievable. Nevertheless, the problem could be solved by a ligand L4, which binds differently to the ligand-specific monovalent binding species B than ligand L1. Hence, if the monovalent concentration is determined with ligands L1 and L4, the difference in concentration can be attributed to the species monovalent binding to L1, but not L4 or vice versa.

[0148] Hence, preferably, in the method according to the present invention the ligand-specific monovalent binding species M comprises an unknown concentration CMI_U of a first ligand-specific monovalent binding species M1 and an unknown concentration CM2_U of a second ligand-specific monovalent binding species M2 and the method further comprises the steps: an activated sensor surface A4L4 providing step actMBi comprising providing a sensor surface A4 having a ligand L4 immobilized thereon, wherein the ligand L4 can be bound by the first ligand-specific monovalent binding species M1, but the ligand L4 cannot be bound by the second ligandspecific monovalent binding species M2 or the ligand L4 is multivalently bound by the second ligand-specific monovalent binding species M2; a calibration step calMi yielding a monovalent calibration function FA4L4_MI, recording in step mesMi for the sample Su comprising the unknown concentration CMI_U of the first ligand-specific monovalent binding species M1 a sensorgram SGUusing a sensor surface A4, selecting in step selMi a maximum response value rA4L4_M-t_u or the slope of the maximum response value SA4L4_MI_U from the sensorgram Su, retrieving in step retrMi the total concentration CMI_U of the first ligand-specific monovalent binding species M1 in the sample Su by calculating the concentration from the response value rA4L4_Mi_u or the slope of the response value SA4L4_MI_U using the calibration function FA4L4_MI-

[0149] Preferably, the method according to the present invention further comprises the steps of: calculating in step retrMBi the concentration CB2_U using the following formula (XIII) cM2 u= cM_u ~cMl_u (XI 11,

[0150] For the same reasons as set out for the determination of the monovalent calibration function FA ILI_M, also in the determination of the monovalent calibration function FA4L4_MI at least two samples comprising known concentration CMI (I) of the ligandspecific monovalent binding species M1 and the first ligand-specific multivalent binding species B are measured. Hence, preferably, in the present invention, the step cahi comprises the following steps:

[0151] ■ Imax times recording in step cal_mesMi(l) a sensorgram SG(I) for a sample Sk(l) comprising a known concentration CMI(I) of the ligand-specific monovalent binding species M1 using a sensor surface A4, wherein the sensorgram SG(I) comprises response values fA4L4_Mi(l) related to the time relative to the start time of the recording step, wherein CMI(I) differs from all concentrations in the range of from CMI(1) to CMi(lmax) except for CMI(I), wherein I is a natural number in the range of from 1 to lmax, and wherein lmax is 2 or higher;

[0152] ■ aligning in step cal_algMi the sensorgrams SG(1) to SG(lmax) at the base line after the start time and before any increase of the response value fA4L4_Mi(l),

[0153] • extracting in step cal_extrMi the response value fA4L4_Mi(l) or the slope of the response value SA4L4_MI (I) from each sensorgram SG(I) at a time point subsequent to the time corresponding to the maximum response value and providing a datapoint d(l) comprising the response value fA4L4_Mi(l) or the slope of the response value SA4L4_MI (I) and the corresponding known concentration CMI (I) of the first ligand-specific monovalent binding species M1 yielding lmax datapoints, and

[0154] ■ fitting in step cal_fitMi a calibration function FA4L4_MI from the lmax datapoints. In the aligning step cal_algMi, the sensorgrams SG(1) to SG(lmax) are aligned in that the response value having the maximum value is set to zero. Usually, the maximum response value denotes the start of the dissociation phase. Hence, the sensorgrams SG(1) to SG(lmax) are aligned so that the start of the dissociation phase for all sensorgrams is at the same absolute response value, i.e. preferably at zero. This has the advantage that response values taken from the sensorgrams after the maximum response value can be used to prepare the monovalent calibration curve representing the dissociation behavior in dependency of the concentration of the ligand-specific monovalent binding species. Hence, in case the step cal_extrMi is carried out by extracting the slope of the response value SA4L4_MI(I) from each sensorgram SG(I), the aligned step cal_algMi is not necessarily needed.

[0155] Preferably, in the method according to the present invention, lmax is 3 or higher, preferably 5 or higher, more preferably 10 or higher. This has the advantage of higher precision of the relationship between concentration of the ligand-specific monovalent binding species and the response value. Usually, lmax is not higher than 30.

[0156] As discussed earlier, the measurement step cal_mesMi(l) in the calibration step calMi is carried out by injecting a sample comprising known concentrations CMI(I) of the ligand-specific monovalent binding species (target molecule) leading to association of the target-molecule to the ligand on the sensor surface and measuring the response value. Furthermore, after the sample has run out, the measurement is continued merely with solvent and buffer, which leads to dissociation of the ligand-associated target molecule. Hence, step cal_mesMi(l) comprises measurement of the association and dissociation phase. Thus, preferably, each step cal_mesMi(l) comprises the steps cal_mes_assMi(l) and cal_mes_dissMi(l), wherein step cal_mes_assMi(l) comprises measuring the response values rMi_ass(l) of the activated sensor surface A4L4 during contacting a solution comprising a known concentration CMI(I) of the at least one monovalent binding species M1 and not comprising any of the at least one multivalent binding species B with the activated sensor surface A4L4 yielding an associated activated sensor surface A4L4_Mi_ass(l) until a maximum response value of the at least one monovalent binding species M1 to the activated sensor surface A4L4 has been reached, step cal_mess_dissMi(l) comprises measuring the response values rA4L4_Mi_diss(l) of the associated activated sensor surface A4L4_M-i_ass(l) during contacting a solution not comprising any of the at least one monovalent binding species and not comprising any of the at least one multivalent binding species with the associated activated sensor surface A4L4_Mn_ass(l) thereby yielding a dissociated activated sensor surface A4L4_diss(l) and adding the response values rA4L4_Mi_diss(l) to the response values rA4L4_Mi_ass(l) yielding the total response values rA4L4_Mi(l) .

[0157] Preferably, ligands L1 and L4 have a different molecular structure and / or L2 and L4 have a different molecular structure. Hence, more preferably ligand L4 has a binding site LBSMI, the first ligand-specific monovalent binding species M1 has a binding site M1BSL bindable to LBSMI, the second ligand-specific monovalent binding species M2 has no binding site bindable to L4 or the second ligand-specific monovalent binding species M2 has two or more binding site bindable to L4.

[0158] System of the present Invention

[0159] The present invention further relates to a system for analyzing a sample Su, wherein sample Su comprises an unknown concentration CM_U of a ligand-specific monovalent binding species M and further comprises a ligand-specific multivalent binding species 8, wherein the system is suitable for carrying out the method according to the present invention.

[0160] Preferably, the system according to the present invention comprises a sensor surface configured to detect binding interactions between the ligand-specific monovalent binding species M and further comprises a ligand-specific multivalent binding species B and a ligand immobilized at the sensor surface and a computing device configured to cause the system to carry out the method according to the present invention.

[0161] The system according to the present invention preferably is a surface plasmon resonance (SPR) system. In that case, registering a sensor response indicative of binding of the ligand-specific monovalent binding species M and further comprises a ligand-specific multivalent binding species 8 to binding sites of the ligand may be based on surface plasmon resonance.

[0162] Experimental Part

[0163] In the following the principle of the present invention is exemplified in view of the separation of antibodies from a sample.

[0164] Materials and Methods a) Measurement For measuring the response value and / or the slope thereof, Biacore 8K and 8K+ instruments were used. Biacore Insight control Software version 5.0.18.22102 has been used for operating the instrument. Biacore Insight Evaluation Software version 5.0.18.22102 and Excel (Microsoft 365) have been used for data handling. b) Material and Reagents

[0165] Sensor Chip CM5, HBS-EP+ buffer, amine coupling kit were obtained from Cytiva (Uppsala, Sweden).

[0166] Calibration reagents

[0167] Hemlibra® emicizumab) obtained from Apoteket AB (Uppsala, Sweden) was used for calibration in the experiment that included the emicizumab cell culture sample and purified in-house produced bsAb2 (Bispecific antibody with one VLK and one Vi ) was used for calibration in the experiment that included bsAb2 cell culture samples.

[0168] Ligands

[0169] Vuc-ligand’. A modified kappa light chain-binding polypeptide, SEQ ID No 22, as described in WO 2023 / 030995 A1 has been used. This ligand binds to the VLK domain of the light chain of the investigated antibody.

[0170] SEQ ID No: 22 from WO 2023 / 030995 A1 :

[0171] PKEEVTIKAH LIYADGKTQT AEFKGTFEEA TAEAYRYADL LAKEHGKYTV DVADKGYTLQ IKFAGKEKTP EE

[0172] FcVH3-ligand'. A mutated immunoglobulin-binding polypeptide, SEQ ID No 11 , as described in WO 2016 / 079033 A1 have been used. This polypeptide binds to Fc and VH3 regions of immunoglobulins.

[0173] SEQ ID No 11 from WO 2016 / 079033 A1 :

[0174] VDAKFDKEAQ EAFYEILHLP NLTEEQRNAF IQSLKDDPSV SKAILAEAKK LNDAQAPK

[0175] Fc-ligand'. A Fc binding polypeptide, SEQ ID No 7, as described in WO 2023 / 046886 A1 has been used. This polypeptide binds to the Fc region of immunoglobulins.

[0176] SEQ ID No 7 from WO 2023 / 046886 A1 :

[0177] VDAKFDKEAQ EAFYEILHLP NLTEEQRNAFI QSLKDDPSGS KAILAEAKKV NDAQAPK Sensor surface preparation

[0178] The VLK- ligand, FcVH3-ligand, and Fc-ligand respectively, were each immobilized on the sensor surface by using direct amine coupling to the activated dextran using the following method. The sensor chip (CM5, obtained from Cytiva (Uppsala, Sweden)) was activated in a Biacore system by an injection of EDC / NHS (50:50) for 7 min, followed by an injection of 30 pg / mL of ligand dissolved in 10 mM acetate at pH 4.5 for 7 min. The remaining binding sites were blocked by injecting ethanolamine, pH 8.5, for 7 min. The coupling procedures were carried out at a flow rate of 10 pl / min at 25 °C. Immobilization of FcVH3-ligand and Fc-ligand resulted in immobilization levels between 5000- 6000 RU. Immobilization of VLK- ligand resulted in an immobilization level between 7200-7300 RU.

[0179] Calibration

[0180] Standards of emicizumab and bsAb2 were prepared in HBS-EP+ buffer. The concentrations used for construction of the calibration curves were 0.125, 0.25, 0.5, 1 , 2, 4, 8, 16 and 32 pg / mL.

[0181] Cell culture sample (emicizumab)

[0182] A cell culture clone of emicizumab (stock, concentration 1 g / L) was as diluted 1 :200 in HBS-EP+ buffer into a final concentration of 5 pg / mL.

[0183] Cell culture sample (bsAb2)

[0184] Thirteen cell culture samples of bsAb2 (A to M) from the clonal screening and selection phase have been investigated. The samples were diluted 1 :25 in HBS- EP+ buffer.

[0185] Assay procedure

[0186] The diluted cell culture samples of emicizumab and bsAb2 were injected across the immobilized sensor surface at a flow rate of 10 pL / min for 120 s. The bound sample was then allowed to dissociate for 300 s. The increase in response due to sample binding and decrease in response due to dissociation was recorded. The surface was then regenerated by injection of a 30 s pulse of a 10 mM Glycine-HCI buffer pH 1.5. The analysis temperature was 25°C.

[0187] General procedure

[0188] A useful application for the method of the present invention could be the selection of the best clones early during a cell line development (CLD) process. The best clones are the clones that produce high titers of the desired antibody and low levels of unwanted byproducts, such as mispaired antibodies and unpaired heavy chains, i.e. 1 Abs, etc.

[0189] Examples IE1 and IE2 comprise three ligands with different binding properties. The FcVH3-ligand binds to two regions, on the heavy chain specific the Fc region and the VH3 region. The \4x-ligand binds to the variable domain of light chains (LC) of kappa type (VL = variable light chain). Finally, the Fc-ligand binds only to the specific Fc- region of the heavy chain as is also bound by the FcVH3-ligand but lacking the VH3 binding ability.

[0190] The choice of ligands to include in the analysis depends on the binding properties of the antibody one wishes to produce. For example, FcVH3-ligand gives valuable information for emicizumab. Emicizumab is a Fc-engineered bispecific antibody. The Fc engineering lead to different Fc binding properties on one of its heavy chains. Thus, components of a cell culture of emicizumab bind differently to the FcVH3-ligand compared to the Fc-ligand. bsAb2 on the other hand has identical Fc binding properties on both heavy chains and therefore binds with both its heavy chains to Fc binding ligands. Therefore, bsAb2 gives the same total concentration when being measured on FCVH3 or Fc-ligands.

[0191] It is important to remember that the total concentration (pg / mL) in a sample includes all components binding to the ligand on the sensor surface. The proportion of the concentration in the sample that binds to the ligand on the sensor surface depends on the binding properties of the immobilized ligand and the binding capabilities of the target molecule, i.e. the antibody in this case. Thus, the measured total concentration but also the monovalent concentration can differ depending on the ligand used. The fact that different ligands measure different total concentrations and that one can distinguish monovalent concentration from total concentration are the two principles that enable the determination of the proportion of unwanted byproducts in the cell culture samples.

[0192] Example IE1 (emicizumab cell culture sample)

[0193] Generally, the sample of emicizumab (cf. Figure 3) production could comprise the species as shown in the following Table 1.

[0194] Table 1: Components comprised in the emicizumab cell culture and respective binding behavior towards different ligands

[0195] * It is not fully understood why free kappa light chains dissociate very slowly. Affinity chromatography studies have shown that free kappa light chains are the last fraction eluted. The first fraction eluted is monovalent binding antibody (K / A) , followed by bivalent binding Antibody (K / K) . The second fraction is dimeric free kappa light chains, and the last fraction is monomeric kappa light chain. Without wishing to be bound by theory this behavior is interpreted in that small entities such as free kappa light chains jump between ligand binding sites and therefore cannot come off easily.

[0196] Hence, the Fc-ligand binds only to the Fc-region of the VH1 bearing heavy chain. Thereby, homodimer VH3 and 14 Ab VH3 are not bound at all by the Fc-ligand.

[0197] Furthermore, the FcVH3-ligand forms bivalent bindings to all forms as can be expected in a sample of a cell culture of emicizumab except for free kappa light chains and both 14 Abs. Reason is that, as explained above, the FcV^S-ligand provides binding sites for Fc on one heavy chain and for VH3 on the other heavy chain only for emicizumab.

[0198] Finally, as the VLK- ligand only provides binding sites to the kappa light chains, it can directly distinguish between a mixture comprising emicizumab as well as its homodimers in comparison to a mixture comprising both 14 Abs.

[0199] Figure 4 shows the calibration curve recorded on the Fc-ligand to determine the total concentration of a sample. This measurement constitutes a concentration determination procedure as known from the prior art. Respective measurements were also carried out on the FCVH3 and the W-ligand.

[0200] Figure 5 shows the calibration curve used to determine the concentration of the monovalent binding component of a sample on the Fc-ligand. The slope of response at a fixed time point during the dissociation phase is plotted against the concentration and gives the concentration of the dissociating monovalent binding component in the sample. As the slope of response has been used, no alignment of the sensorgrams had been necessary. Calibration curves from the VLK- ligand and FcVH3-ligand are not shown as those curves were only used to determine the total sample concentration according to the procedure as known from the prior art.

[0201] The total concentration determination (i.e. procedure known from the prior art) using the FcVH3-ligand showed that the emicizumab cell culture sample contained a total concentration of 4.76 pg / mL when being measured on the FcVH3-ligand. The FCVH3- ligand binds to the Fc region on one of the heavy chains and to the VH3 region on the other heavy chain of emicizumab. Hence, the measured total concentration corresponds to all antibody material in the sample having VH3 and / or Fc binding capacity and, thus, to all conceivable species except free kappa light chains (cf. Table 1).

[0202] The total concentration determination (i.e. procedure known from the prior art) using the Fc-ligand resulted in a value of 4.12 pg / mL, which corresponds to emicizumab, homodimer VH1 and 14 Ab VH1.

[0203] The concentration determination of the Fc-ligand-specific monovalent binding components (i.e. procedure according to the invention) on the Fc-ligand resulted in a value of 3.8 pg / mL. The Fc-ligand-specific monovalent binding components primarily include emicizumab but also 1 Ab VH1 (cf. Table 1 above). Thus, no differentiation is possible between these species by measuring on the Fc-ligand.

[0204] Since emicizumab has a common kappa chain, only both 14 Abs will bind monovalent to the W-ligand (cf. Table 1 , for an explanation as regards free kappa light chains it is referred to the starred comment therein). Thus, both 14 Abs can be quantified on the W-ligand by measurement of the monovalent concentration. However, no dissociation at all could be observed when determining the total concentration (i.e. procedure known from the prior art) on the W-ligand. Hence, this shows that the sample does not contain any measurable amount of both 14 Abs.

[0205] Thus, the fraction that binds monovalent to the Fc ligand, which was measured to be 3.8 pg / mL) contains only emicizumab. The emicizumab concentration in the measured sample therefore was thus 3.8 pg / mL, which corresponds to 79.8% of the total concentration. The difference between the total concentration measured on the FcVnS-ligand and the Fc-ligand is 0.64 pg / mL. This difference provides a measure of the proportion of material that does not bind to the Fc-ligand, which in this case corresponds to homodimer VH3 and 14 Ab VH3 (cf. Table 1). As has been shown above, no 14 Ab VH3 has been present in the sample. Thus, the difference in concentration between the total concentration measured on the FcVH3-ligand and the Fc-ligand can be attributed to the homodimer VH3 only. Homodimer VH3 therefore constitutes 13.4% of the total concentration.

[0206] The difference between the total concentration measured on the Fc-ligand and the concentration of Fc-specific monovalent binding components measured on the Fc- ligand corresponds to bivalent binding material. This bivalent binding material is in case of the Fc-ligand the homodimer with Fc-binding ability VH1. Hence, the homodimer VH1 concentration is therefore 0.32 pg / mL, which corresponds to 6.7% of the total concentration.

[0207] The total concentration measured on the W-ligand was determined to be 5.84 pg / mL. This concentration also includes free kappa light chains in addition to the total antibody material bivalently bound on the FcVH3-ligand (cf. Table 1). The difference between the total concentration measured on the W-ligand and the total concentration measured on FCVH3 is 1.08 pg / mL. Thus, the concentration of free kappa chain is 1.08 pg / mL, which constitutes 18.5% of the total material.

[0208] A summary of the results is shown in Tables 2 and 3 below. The emicizumab cell culture clone was also analyzed with LC-MS the results correspond well with the results shown in table 3.

[0209] Table 2: Total concentration measured with various ligands and concentration of Fc- specific monovalent binding component measured with Fc-ligand

[0210] Table 3: Measured or calculated concentrations of specific components in the sample of the cell culture of emicizumab

[0211] * Calculation in square brackets a Concentration of Fc-specific monovalent binding component measured with Fc- ligand b Total concentration measured with Fc-ligand c Total concentration measured with FcVH3-ligand d Total conentration measured with IZ / .K-ligand

[0212] Example IE2 (bsAb2 cell culture samples)

[0213] Generally, the sample of bsAb2 (cf. Figure 6) production could comprise the species as shown in the following Table 4. The structure of bsAb2 is similar to Plamotamab. Hence, one monomer is structured according to a normal antibody monomer, thereby having a heavy chain ( / 7C) and a light chain (LC). This heavy chain has binding capabilities to the Fc binding sites (Fc-ligand or FcVH3-ligand), but contains a VH1 region and, hence, cannot bind to VH3 binding sites (FcV^S-ligand). The other monomer is one single peptide, which is a single chain variable fragment (scFv) covalently bound to a Fc region. Hence, the bsAb2 antibody consists of three different polypeptides: HC, LC and scFv-Fc. The light chain (LC) is of the kappa type and, thus, has binding capabilities to the \4K-ligand. The scFv-Fc polypeptide has binding capabilities to the Fc binding sites (Fc-ligand and FcVH3-ligand). Thus, bsAb2 already differs from emicizumab in that both heavy chains comprise capabilities to the Fc binding sites. However, the VL domain of the scFv-Fc polypeptide is of lambda type and, thus, has no binding capabilities to the \4x-ligand. On the other hand, the VH domain scFv-Fc polypeptide is of VH3 type and, thus, has binding capabilities to the FcVH3-ligand.

[0214] Table 4: Components comprised in the bsAb2 cell culture and respective binding behavior towards different ligands (cf. Figure 6)

[0215] * It is not fully understood why free kappa light chains dissociate very slowly. Affinity chromatography studies have shown that free kappa light chains are the last fraction eluted. The first fraction eluted is monovalent binding antibody (K / A) , followed by bivalent binding Antibody (K / K) . The second fraction is dimeric free kappa light chains, and the last fraction is monomeric kappa light chain. Without wishing to be bound by theory this behavior is interpreted in that small entities such as free kappa light chains jump between ligand binding sites and therefore cannot come off easily.

[0216] Hence, the Fc-ligand binds to the Fc-region of both heavy chains. Thereby, homodimers VH1 and VH3 and bsAb2 are bound bivalent, wherein both 14 Abs are bound only monovalently by the Fc-ligand. Free kappa light chains are not bound at all by the Fc-ligand.

[0217] Furthermore, the FcVH3-ligand forms at least bivalent bindings to all forms as can be expected in a sample of a cell culture of bsAb2 except for 14 Ab VH1 as well as free kappa light chains. Reason is that, as explained above, the FcV^S-ligand provides binding sites for both Fc and VH3.

[0218] Finally, as the VLK- ligand only provides binding sites to the kappa light chains, will, bsAb2 and 14 Ab VH1 bind monovalently whereas homodimer VH1 and free kappa light chain binds bivalently. Homodimer VH3 and 1 / 2 Ab VH3 are not bound by the W-ligand.

[0219] Figure 7 shows the calibration curve used to determine the monovalent concentration for the \4x-ligand, i.e. concentration of the \4x-ligand-specific monovalent binding species. As detailed above, the ligand binding properties of bsAb2 differ from emicizumab of example IE1. Therefore, different measurements are compared in examples IE1 and IE2. Hence, in case of bsAb2, the total concentration was measured on the Fc-ligand, while the monovalent concentration was measured on the \4x-ligand. The total concentration determination (i.e. procedure known from the prior art) using the Fc-ligand showed that the bsAb2 cell culture samples contained a total concentration in the range of from 18.5 to 408 pg / mL when being measured on a Fc- assay. The Fc-ligand binds to the Fc region of both heavy chains of bsAb2. Thus, the measured total concentration corresponds to all antibody material in the sample except for free kappa light chains (cf. Table 4 above).

[0220] The concentration determination of the W-ligand-specific monovalent binding components (i.e. procedure according to the invention) on the VLK- ligand resulted in a concentration value in the range of from 9.12 to 394 pg / mL. The \Z / K-ligand-specific monovalent binding components primarily include bsAb2 but also 14 Ab VH1 comprising kappa light chains (cf. Table 4 above). None of the clones produced any 1 Abs (confirmed by LC-MS analysis) and the monovalent concentration is therefore equivalent with the pure bsAb2 concentration, which is reflected by the concentration determination of the W-ligand-specific monovalent binding components.

[0221] Hence, as the ratio between the total concentration and the monovalent concentration determines the selection of top clones, the clones having such a ratio closest to 100% are clones having the lowest concentration of unwanted material.

[0222] Table 5 shows the results from this analysis. The table 5 additionally includes the results from a Biacore sandwich assay. Eight of the samples shown in the table (samples A, B, C, D, G, H, K and M) have also been analyzed with LC-MS and correspond well to the results shown in Table 5.

[0223] Table 5: Results of measurement of thirteen samples of a bsAb2 cell culture calculated by VLK (mono) / Fc (total) measured and calculated using Biacore sandwich assay

Claims

- 43 -Claims1. Method for analyzing a sample Su, wherein sample Su comprises an unknown concentration CM_U of a ligand-specific monovalent binding species M and further comprises a ligand-specific multivalent binding species B, the method comprising the steps of: an activated sensor surface A 1LI providing step actM comprising providing a sensor surface A 1 having a ligand L1 immobilized thereon, wherein the ligand L1 can be monovalently bound by the ligand-specific monovalent binding species M and the ligand L1 can be multivalently bound by the ligand-specific multivalent binding species B, a calibration step calM yielding a monovalent calibration function FAILI_M, recording in step mesM for the sample Su comprising the unknown concentration CM_U of the ligand-specific monovalent binding species M a sensorgram SGUusing the activated sensor surface A 1LI , selecting in step selM a response value fAiLi_M_u or the slope of the response value SAILI_M_U from the sensorgram Su at a selection time tAiLi_M_s, and retrieving in step retrM the concentration CM_U of the ligand-specific monovalent binding species M in the sample Su by calculating the concentration form the response value fAiLi_M_u or the slope of the response value SAILI_M_U using the calibration function FA ILI_M-2. The method according to claim 1 , wherein the calibration step calM comprises the following steps:■ rrimax times recording in step cal_mesM(m) a sensorgram SG(m) for a sample Sk(m) comprising a known concentration CM(m) of the ligandspecific monovalent binding species M using the activated sensor surface A 1LI, wherein the sensorgram SG(m) comprises response values rAiLi_M(m) related to the time relative to the start time of the recording step, wherein CM(m) differs from all concentrations in the range of from CM(1) to CM(mmax) except for CM(m), wherein m is a natural number in the range of from 1 to mmax, and wherein mmax is 2 or higher;- 44 -■ aligning in step cal_algM the sensorgrams SG(1) to SG(mmax) at the maximum response value,■ extracting in step cal_extrM the response value rAiLi_M(m) or the slope of the response value SA ILI_M(ITI) from each sensorgram SG(m) at a time point subsequent to the time corresponding to the maximum response value and providing a datapoint d(m) comprising the response value rAiLi_M(m) or the slope of the response value SA ILI_M(ITI) and the corresponding known concentration CM(m) of the ligand-specific monovalent binding species M yielding m max datapoints, and■ fitting in step cal_fitM the monovalent calibration function FAILI_M from the rrimax datapoints.

3. The method according to claims 1 or 2, wherein the ligand L1 has a binding site LBSM, the ligand-specific monovalent binding species M has a binding site MBSL bindable to LBSM, the ligand-specific monovalent binding species M has no further binding site bindable to L1, the ligand L1 has a binding site LBSBI and a binding site LBSB2, the ligand-specific multivalent binding species B has a binding site BBSLI bindable to LBSBI and a binding site BBSL2 bindable to LBSB2-4. The method according to claims 1 to 3, wherein the concentration CMB_U in sample Suof the ligand-specific monovalent binding species M and the ligandspecific multivalent binding species B is unknown, wherein the method further comprises the steps of: an activated sensor surface A2L2 providing step actMB comprising providing a sensor surface A2 having a ligand L2 immobilized thereon, wherein the ligand L2 can be bound by the ligand-specific monovalent binding species M and the ligand L2 can be bound by the ligand-specific multivalent binding species B, a calibration step CSIMB yielding a total calibration function FA2L2_MB, recording in step mesMB for the sample Su comprising the unknown concentration CM_U of the ligand-specific monovalent binding species M a sensorgram SGUusing a sensor surface A2,- 45 - selecting in step SGIMB a maximum response value rA2L2_MB_u or the slope of the maximum response value SA2L2_MB_U from the sensorgram Su, and retrieving in step retrMB the total concentration CMB_U of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species B in the sample Su by calculating the concentration from the response value fA2L2_MB_u or the slope of the response value SA2L2_MB_U using the calibration function FA2L2_MB-5. The method according to claim 4, wherein the concentration CB_U in sample Su of the ligand-specific multivalent binding species B is unknown, wherein the method further comprises the steps of: calculating in step retrB the concentration CB_U using the following formula (XI)CB_U= CMBU~CM_U (XI)6. The method according to claims 4 or 5, wherein the ligand L2 has a binding site LBSM, the ligand-specific monovalent binding species M has a binding site MBSL bindable to LBSM, the ligand-specific monovalent binding species M has no further binding site bindable to L2, the ligand L2 has a binding site LBSBI and a binding site LBSB2 the ligand-specific multivalent binding species B has a binding site BBSLI bindable to LBSBI and a binding site BBSL2 bindable to LBSB2-7. The method according to claims 4 to 6, wherein LBSM and LBSBI are the same binding sites, preferably wherein LBSM, LBSBI , and LBSB2 are the same binding sites, most preferably wherein L1 and L2 have the same molecular structure.

8. The method according to any of the preceding claims 4 to 7, wherein the ligandspecific multivalent binding species B comprises an unknown concentration CB I_U of a first ligand-specific multivalent binding species B1 and an unknown concentration CB2_U of a second ligand-specific multivalent binding species B2, and wherein the method further comprises the steps: an activated sensor surface A3LS providing step actMBi comprising providing a sensor surface A3 having a ligand L3 immobilized thereon,wherein the ligand L3 can be bound by the ligand-specific monovalent binding species M and the ligand L3 can be bound by the first ligandspecific multivalent binding species B1, but the ligand L3 cannot be bound by the second ligand-specific multivalent binding species B2, a calibration step CBIMBI yielding a total calibration function FA3L3_MBI, recording in step mesMBi for the sample Su comprising the unknown concentration CB I_U of the first ligand-specific multivalent binding species B1 a sensorgram SGUusing a sensor surfacewherein the sensor surface is activated by immobilizing a ligand L3 thereon yielding an activated sensor surface A3L3, wherein the ligand L3 can be bound by the ligand-specific monovalent binding species M and the ligand L3 can be bound by the first ligandspecific multivalent binding species B1, but the ligand L3 cannot be bound by the second ligand-specific multivalent binding species B2; selecting in step SGIMBI a maximum response value rA3L3_MBi_u or the slope of the maximum response value SA3L3_MB I_U from the sensorgram Su, and retrieving in step retrMBi the total concentration CMBI_U of the ligand-specific monovalent binding species M and the ligand-specific multivalent binding species B in the sample Su by calculating the concentration from the response value rA3L3_MBi_u or the slope of the response value SA3L3_MBI_U using the calibration function FA3L3_MBI.

9. The method according to claim 8, wherein the method further comprises the steps of: calculating in step retrMBi the concentration CB2_U using the following formula (XII)CB2_U= CMB_U ~CMB1_U (XII).

10. The method according to any of the preceding claims, wherein the sample Su is prepared from a mixture from a bioreactor comprising microorganisms, such as yeasts, bacteria, and / or fungi, cells, viruses, and / or solvents, wherein the sample Su is preferably prepared from a mixture from a bioreactor comprising cells, and / or solvents.11 . The method according to any of the preceding claims wherein the ligand-specific monovalent binding species M and / or the ligand-specific multivalent binding species B independently from each other are substances naturally occurring inanimal or plant tissue, are produced by naturally occurring microorganisms or enzymes, and / or are produced by modified, preferably genetically modified, microorganisms or enzymes, wherein the ligand-specific monovalent binding species M and / or the ligand-specific multivalent binding species B independently from each other are preferably selected from the list consisting of antibodies, proteins, viruses, organic substances, DNA molecules, RNA molecules, shorter nucleotides molecules, exosomes, cells (such as used in cell therapy), or specific target molecules, wherein the ligand-specific monovalent binding species M and / or the ligand-specific multivalent binding species B independently from each other most preferably are antibodies or fragments of antibodies, preferably antibodies or fragments of antibodies both produced by the same cell type.

12. The method according to any of the preceding claims 3 to 11 , wherein each of the binding sites MBSL, BBSLI, and BBSL2 are selected from the fragment crystallizable region FC of the antibodies or fragments of antibodies, the variable heavy chain region VH of the antibodies or fragments of antibodies, or the variable light chain region VL.

13. A system for analyzing a sample Su, wherein sample Su comprises an unknown concentration CM_U of a ligand-specific monovalent binding species M and further comprises a ligand-specific multivalent binding species 8, wherein the system is suitable for carrying out the method according to any of the preceding claims 1 to 12.

14. The system according to claim 13, further comprising a sensor surface configured to detect binding interactions between the ligand-specific monovalent binding species M and further comprises a ligand-specific multivalent binding species B and a ligand immobilized at the sensor surface and a computing device configured to cause the system to carry out the method according to the present invention.

15. The system according to claim 14, wherein the system is a surface plasmon resonance (SPR) system.

Images