Stoichiometric Constraints in Bispecific Binding Networks

Why Stoichiometric Constraints Matter

A bispecific binding model can contain several antibody states: free antibody, antibody bound to the first antigen, antibody bound to the second antigen, and antibody bound to both. Each state is a different form of the same antibody molecule, not a separate antibody amount. Conservation equations make that distinction explicit. They provide a direct check on model equations, initial conditions, and proposed concentrations before interpreting the binding behavior.

The earlier material on state-based models of multispecific binding motivates representing each occupancy state separately. Here the focus is the stoichiometric accounting across those states: how many antibody and antigen molecules each state contains, and what totals must follow from that accounting.

Definition: Let $B$ denote free bispecific antibody, $X$ and $Y$ the two free antigens, $BX$ and $BY$ the singly bound antibody states, and $BXY$ the doubly bound state. In the one-binding-site-per-arm network, total antibody and antigen concentrations are $$ B_{\mathrm{tot}}=B+BX+BY+BXY,\qquad X_{\mathrm{tot}}=X+BX+BXY,\qquad Y_{\mathrm{tot}}=Y+BY+BXY. $$ Each term is a concentration in the same compartment and on the same concentration scale.

Model Assumptions and Conservation Equations

These equations apply to a closed binding network with one antibody molecule in every antibody-containing state, at most one $X$ molecule and one $Y$ molecule bound per antibody, and no additional states such as antibody aggregates or complexes with other ligands. The antigens bind to distinct arms, so $BXY$ contains one antibody, one $X$, and one $Y$. All species concentrations are nonnegative.

The network can be written with binding reactions such as $B+X\rightleftharpoons BX$, $B+Y\rightleftharpoons BY$, $BX+Y\rightleftharpoons BXY$, and $BY+X\rightleftharpoons BXY$. The route to $BXY$ does not change its composition: either route produces one molecule of each of the three molecular components. Adding the concentrations of the states that contain a given component therefore gives its total.

\begin{aligned} B_{\mathrm{tot}} &= B+BX+BY+BXY,\\ X_{\mathrm{tot}} &= X+BX+BXY,\\ Y_{\mathrm{tot}} &= Y+BY+BXY. \end{aligned}

A compact way to verify the accounting is to list the amount of each component in each species. In the order $(B,BX,BY,BXY,X,Y)$, the rows for antibody, antigen $X$, and antigen $Y$ are $(1,1,1,1,0,0)$, $(0,1,0,1,1,0)$, and $(0,0,1,1,0,1)$, respectively. For example, the $BXY$ column has a one in all three rows because the doubly bound complex contains one of every component. This bookkeeping is the basis of the conservation equations; it does not require the system to be at binding equilibrium.

For a binding-only system in a fixed compartment, the totals remain constant as molecules shift between free and bound states. If antibody or antigen is supplied, degraded, cleared, or transported across the compartment boundary, its local total need not remain constant. The conservation statement then applies to the binding reactions alone; the full time-dependent total also reflects those external fluxes.

Worked Concentration Checks

Worked Example: Calculating Totals From State Concentrations

Suppose a hypothetical model gives $B=2$ nM, $BX=3$ nM, $BY=1$ nM, $BXY=4$ nM, $X=6$ nM, and $Y=8$ nM. Calculate the totals and check the component counts.

The antibody total includes every antibody-containing state:

B_{\mathrm{tot}}=2+3+1+4=10\ \mathrm{nM}.

Free $X$ and every state containing bound $X$ contribute to its total:

X_{\mathrm{tot}}=6+3+4=13\ \mathrm{nM}.

Likewise, $Y$ is present free, in $BY$, and in $BXY$:

Y_{\mathrm{tot}}=8+1+4=13\ \mathrm{nM}.

All three totals are nonnegative and each bound complex is counted once for each component it contains. In particular, $BXY$ contributes 4 nM to each of the three totals, not 12 nM to any single total.

Worked Example: Detecting an Antibody-Balance Violation

Consider hypothetical reported states $B=1$ nM, $BX=4$ nM, $BY=3$ nM, and $BXY=3$ nM, with a specified antibody total of $10$ nM. Are these values feasible under the stated network?

The antibody amount implied by the states is

B+BX+BY+BXY=1+4+3+3=11\ \mathrm{nM}.

This exceeds the specified total by $11-10=1$ nM. The values cannot all be correct under the one-antibody-per-state model: every antibody molecule must appear in exactly one of the four states, so their concentrations must sum to $B_{\mathrm{tot}}$. Changing the free-antigen concentrations cannot repair this particular violation because neither free antigen appears in the antibody balance.

Worked Example: Detecting an Antigen-Balance Violation

Suppose a hypothetical system has $B_{\mathrm{tot}}=8$ nM, $X_{\mathrm{tot}}=5$ nM, and $Y_{\mathrm{tot}}=6$ nM. A proposed state assignment is $B=1$ nM, $BX=2$ nM, $BY=1$ nM, and $BXY=4$ nM. Check whether these antibody states can fit the totals.

First, the antibody balance is satisfied:

1+2+1+4=8\ \mathrm{nM}=B_{\mathrm{tot}}.

But the antibody-bound $X$ concentration alone is

BX+BXY=2+4=6\ \mathrm{nM}.

The specified $X_{\mathrm{tot}}$ is only 5 nM. Since free $X$ cannot be negative, adding any free $X$ would make the total at least 6 nM, not 5 nM. Thus the assignment is impossible even though its antibody balance is correct. The $Y$ bound amount is $BY+BXY=1+4=5$ nM, which is no greater than $Y_{\mathrm{tot}}=6$ nM; that balance would leave $1$ nM free $Y$. It is the $X$ balance, not the $Y$ balance, that fails.

Feasibility Conditions and a Short Justification

Given total concentrations and proposed antibody-state concentrations, the free species are determined by subtraction:

\begin{aligned} B &= B_{\mathrm{tot}}-BX-BY-BXY,\\ X &= X_{\mathrm{tot}}-BX-BXY,\\ Y &= Y_{\mathrm{tot}}-BY-BXY. \end{aligned}

These expressions give a direct feasibility test. The proposed bound-state concentrations must be nonnegative, and each calculated free concentration must also be nonnegative. Equivalently, the bound amounts cannot exceed their corresponding totals:

BX+BY+BXY\leq B_{\mathrm{tot}},\qquad BX+BXY\leq X_{\mathrm{tot}},\qquad BY+BXY\leq Y_{\mathrm{tot}}.

These inequalities are necessary because the omitted free concentrations are nonnegative. They are also sufficient for this network if the totals are nonnegative: subtracting the bound amounts then defines nonnegative $B$, $X$, and $Y$ that satisfy all three conservation equations. This is a stoichiometric feasibility test, not a claim that the states satisfy any particular binding-rate law.

A further immediate consequence is that no one complex concentration can exceed any total for a component it contains. In particular, $BXY\leq B_{\mathrm{tot}}$, $BXY\leq X_{\mathrm{tot}}$, and $BXY\leq Y_{\mathrm{tot}}$. Such bounds are useful quick checks, but checking each one separately is not enough: the sum of several bound states may exceed a total even when each state individually falls below it.

Diagnostics, Comparisons, and Limitations

When a simulated trajectory violates a total that should be conserved by binding alone, inspect the model before interpreting the predicted occupancy. Check whether every transition transfers the correct molecular components, whether the same state is included in all applicable totals, and whether the initial states satisfy the balances. Then check for accidental unit or compartment mismatches. A concentration in one compartment should not be added directly to a concentration in another without accounting for the amounts and volumes involved.

Small numerical deviations can arise from rounding or from numerical integration tolerances. Compare the imbalance with the scale of the total and the solver's precision; a tiny difference in rounded output is not equivalent to a substantial negative free concentration. In contrast, a large imbalance that grows during a binding-only simulation suggests a model equation, coding, or state-definition error. If external input, clearance, or transport is present, first account for its contribution before diagnosing a conservation failure.

Stoichiometric balances complement, rather than replace, other descriptions of binding. A state-based kinetic model tracks concentrations and their changes over time. An equilibrium calculation imposes additional binding conditions to determine state distributions. The conservation equations constrain either approach, but they alone do not specify how much antibody occupies each state or how quickly states interconvert. That requires additional kinetic or equilibrium assumptions.

The check also depends on the species list and assay interpretation. If the molecule can bind more than one copy of an antigen per arm, form aggregates, or occupy additional states, the component counts must be revised to reflect that stoichiometry. Similarly, an assay signal is not automatically the concentration of a specific species: a total-antibody assay may detect several states, while a free-antibody assay may detect only $B$. Apply the balances to concentrations defined for the modeled species, not to measurements whose state-specific response has not been established.

Key takeaway: Write totals by counting molecular components in each state. For this network, a valid assignment must have nonnegative states and must not use more antibody, $X$, or $Y$ than the corresponding total. A violation is evidence that the proposed states, totals, assumptions, or measurements are inconsistent.

Check Your Understanding

Answer each question using the one-binding-site-per-arm network defined above. Brief solutions follow the questions.

If $B=2$, $BX=1$, $BY=3$, and $BXY=2$ nM, what is $B_{\mathrm{tot}}$?
If $X_{\mathrm{tot}}=9$ nM, $BX=2$ nM, and $BXY=4$ nM, what is free $X$?
Can $BY=2$ nM and $BXY=5$ nM be feasible when $Y_{\mathrm{tot}}=6$ nM? Explain.
Why can a binding-only model conserve total antibody even while free antibody concentration changes?

Solutions. (1) $B_{\mathrm{tot}}=2+1+3+2=8$ nM. (2) $X=9-2-4=3$ nM. (3) No: those states contain $2+5=7$ nM of antibody-bound $Y$, already exceeding the 6 nM total; free $Y$ cannot be negative. (4) Binding transfers antibody between free and complex states but neither creates nor removes antibody, so the sum $B+BX+BY+BXY$ remains constant.