SPIKESbind
  • Development Tool
  • Self-Assessment Tool
  • Quiz
  • -log Ki Table
  • Assumptions of the Model
-log Ki Table
Open in new window

Hypothetical ligands have been included in the model to increase its applicability.

Ligand A and Ligand B are examples of ligands that have marked differences in affinities (-log Ki values) for receptor subtypes (not evident with established ligands).

You may nominate affinities (-log Ki values) for Ligands C and D.

Competing
Ligand
-log Ki Value
M1 M2 M3 M4 M5
Established
Atropine 9.0 8.8 9.3 8.9 9.2
Pirenzepine 8.2 6.5 6.9 7.4 7.2
Methoctramine 6.7 7.7 6.0 7.0 6.3
Darifenacin 7.8 7.0 8.8 7.7 8.0
MT-3 6.7 5.9 6.0 8.1 6.0
S-Secoverine 8.0 7.9 7.7 7.7 6.5
Solifenacin 7.6 6.8 7.9 7.0 7.5
DAU-5884 8.9 7.1 8.9 8.5 8.1
Hypothetical
Ligand A 9.0 9.0 6.0 6.0 2.0
Ligand B 8.0 5.0 2.0 5.0 8.0
Ligand C * * * * *
Ligand D * * * * *
Assumptions of the Model
  1. All of the usual assumptions associated with Law of Mass Action that typically describe drug binding to receptors
    • All receptors are equally accessible to ligands.
    • All receptors are either free or bound to ligand - i.e. the model ignores any states of partial binding.
    • Neither ligand nor receptor are altered by binding.
    • Binding is reversible.

  2. Assumptions associated with competition binding studies
    • Only a small fraction of both the labelled and unlabelled ligands has bound, so the free concentration is virtually the same as the added concentration.
    • There is no co-operativity – binding to one binding site does not alter affinity at another site.
    • The experiment has reached equilibrium.
    • Binding is reversible and follows the law of mass action.

  3. Other model-specific assumptions
    • Analysis is conducted using % Specific Binding (no non-specific binding)
    • [3H-QNB] used in studies was << KA of 3H-QNB for receptors present, i.e. Cheng-Prusoff equation reduces to IC50 ≅ Ki for all competing ligands.

Provided below are worked examples (of increasing complexity) that demonstrate how visual inspection of competition binding curves and the application of the SPIKES approach can often be used to definitively determine which receptor subtypes are present within a particular cell/tissue and if more than one receptor is present, their relative densities. Pay particular attention to the notes provided at the end of most of the examples as they provide special insight into the process.

The questions posed in the Quiz Tool are similar in style and difficulty to the worked Examples #4, #5 and #6 (the Quiz Tool does not present triphasic competition binding curves as posed in worked Example #7).

SPIKES Examples

  • Example #1
  • Example #2
  • Example #3
  • Example #4
  • Example #5
  • Example #6
  • Example #7

Apply the SPIKES approach to the following competition binding curve for MT-3 to determine which M cholinoceptor subtype(s) is (are) present in the cell/tissue.

Shape / Position

Visual inspection of the MT-3 competition binding curve indicates that it is inverse sigmoidal, but is it steep sigmoidal or shallow sigmoidal? For a competition binding curve to be considered steep sigmoidal then the extent of %Specific Binding covered by logIC50 ± 1.0 should be greater than 80% - if it is less than 80% then the curve is considered shallow sigmoidal. Firstly, determine the logIC50 value for MT-3, i.e. -8.1. Then determine the %Specific binding values at the [MT-3] that is 1.0 log unit below the –logIC50 (-9.1) and at 1.0 log unit above the –logIC50 (-7.1). As shown in the figure below, the levels of %Specific binding are 90.9% and 9.1%, respectively. The difference (90.9% – 9.1% = 81.8%) is greater than 80% - thus, the competition binding curve is considered steep sigmoidal. This indicates that MT-3 is binding to a population of receptors for which it has similar affinity – for MT-3 that would be either a pure population of M4 receptors for which it has high affinity (-logKi = 8.1) or a population of M1 and/or M2 and/or M3 and/or M5 receptors for which it has a much lower affinity (range of –logKi values 5.9-6.7).

IC50 / Ki value

As a very low concentration of radioligand was used in the study (compared to the Kd of the radioligand), the –logIC50 ~ -logKi and so the –logKi value for MT-3 = 8.1. If a higher concentration of radioligand had been used then the Cheng-Prusoff equation would have been used to convert the IC50 value into the Kivalue.

Elimination / Summation

When we compare the –logKi value of 8.1 to the known –logKi values of MT-3 for M1 (6.7), M2 (5.9), M3 (6.0), M4 (8.1) and M5 (6.0) the –logKi value of 8.1 is ≥ 1 log unit different to the known –logKi values of MT-3 binding to M1, M2, M3 and M5 receptors – and only similar to the –logKi value of 8.1 expected for MT-3 binding to M4 receptors.

Thus, visual inspection of the MT-3 competition binding curve indicates that it is highly unlikely that the cell/tissue contains significant populations of either M1, M2, M3 or M5 receptors (eliminated because experimental and known –logKi values for MT-3 ≥ 1 log unit different), and the only reasonable conclusion is that the cell/tissue contains a homogeneous population of M4 receptors.

Apply the SPIKES approach to the following competition binding curve for S-secoverine to determine which M cholinoceptor subtype(s) is (are) present in this cell/tissue.

Shape / Position

Visual inspection of the S-secoverine competition binding curve indicates that it is inverse sigmoidal, but is it steep sigmoidal or shallow sigmoidal? For a competition binding curve to be considered steep sigmoidal then the extent of %Specific Binding covered by logIC50 ± 1.0 should be greater than 80% - if it is less than 80% then the curve is considered shallow sigmoidal. Firstly, determine the logIC50 value for each competing ligand, which in this example is -6.5. Then determine the %Specific binding values at the [S-secoverine] that is 1.0 log unit below the –logIC50 (-7.5) and at 1.0 log unit above the –logIC50 (-5.5). As shown in the figure below, the levels of %Specific binding are 90.9% and 9.1%, respectively. The difference (90.9% – 9.1% = 81.8%) is greater than 80% - thus, the competition binding curve is considered steep sigmoidal. This indicates that S-secoverine is binding to a population of receptors for which it has similar affinity – for S-Secoverine that would be either a pure population of M5 receptors for which it has low affinity (-logKi = 6.5) or a population of M1 and/or M2 and/or M3 and/or M4 receptors for which it has much higher affinity (range of –logKi values 7.7-8.0).

IC50 / Ki value

As a very low concentration of radioligand was used in the study (compared to the Kd of the radioligand), the –logIC50 ~ -logKi and so the –logKi value for S-secoverine = 6.5. If a higher concentration of radioligand had been used then the Cheng-Prusoff equation would have been used to convert the IC50 value into the Kivalue.

Elimination / Summation

When we compare the –logKi value of 6.5 to the known –logKi values of S-secoverine for M1 (8.0), M2 (7.9), M3 (7.7), M4 (7.7) and M5 (6.5) the –logKi value of 6.5 is ≥ 1 log unit different to the known –logKi values of S-secoverine binding to M1, M2, M3 and M4 receptors – and only similar to the –logKi value of 6.5 expected for S-secoverine binding to M5 receptors.

Thus, visual inspection of the S-secoverine competition binding curve indicates that it is highly unlikely that the cell/tissue contains significant populations of M1, M2, M3 or M4 receptors (eliminated because experimental and known –logKi values ≥ 1 log unit different), and the only reasonable conclusion is that the cell/tissue contains a homogeneous population of M5 receptors.

Note: Examples #1 (MT-3 to identify M4 receptors) and #2 (S-secoverine to identify M5 receptors) are examples whereby the use of a –logKi value obtained from a single competing ligand can successfully identify which receptor is present in a cell/tissue. Similarly, from the list of established ligands provided, DAU-5884 could also be used to identify M2 receptors in a cell/tissue. However, in all other situations more than one receptor-selective ligand would need to be used within a competition binding study to conclusively identify the M receptor subtypes present in a cell/tissue.

Apply the SPIKES approach to the following competition binding curve for S-secoverine to determine which M cholinoceptor subtype(s) is (are) present in this cell/tissue.

Shape / Position

Visual inspection of the S-secoverine competition binding curve indicates that it is inverse sigmoidal, but is it steep sigmoidal or shallow sigmoidal? For competition binding curves to be considered steep sigmoidal then the extent of %Specific Binding covered by logIC50 ± 1.0 should be greater than 80% - if it is less than 80% then the curve is considered shallow sigmoidal. Firstly, determine the logIC50 value for each competing ligand, which in this example is -7.7. Then determine the %Specific binding values at the [S-secoverine] that is 1.0 log unit below the –logIC50 (-8.7) and at 1.0 log unit above the –logIC50 (-6.7). As shown in the figure below, the levels of %Specific binding are 90.9% and 9.1%, respectively. The difference (90.9% – 9.1% = 81.8%) is greater than 80% - thus, the competition binding curve is considered steep sigmoidal. This indicates that S-secoverine is binding to a population of receptors for which it has similar affinity – for S-Secoverine that would be either a pure population of M5 receptors for which it has low affinity (-logKi = 6.5) or a population of M1 and/or M2 and/or M3 and/or M4 receptors for which it has higher affinity (range of –logKi values 7.7-8.0).

IC50 / Ki value

As a very low concentration of radioligand was used in the study (compared to the Kd of the radioligand), the –logIC50 ~ -logKi and so the –logKi value for S-secoverine = 7.7. If a higher concentration of radioligand had been used then the Cheng-Prusoff equation would have been used to convert the IC50 value into the Kivalue.

Elimination / Summation

When we compare the visually-determined –logKi value of 7.7 to the known –logKi values of S-secoverine for M1 (8.0), M2 (7.9), M3 (7.7), M4 (7.7) and M5 (6.5) the visually-determined –logKi value of 7.7 is ≥ 1 log unit different to the –logKi values expected for S-secoverine binding to M5 receptors (6.5), but similar to the –logKi values expected for S-secoverine binding to M1, M2, M3 and M4 receptors (7.7-8.0).

Thus, visual inspection of the S-secoverine competition binding curve indicates that it is highly unlikely that the cell/tissue contains a significant population of M5 receptors (eliminated because experimental and known –logKi values ≥ 1 log unit different), but it is not possible to conclude whether the cell/tissue contains M1 &/or M2 &/or M3 &/or M4 receptors. Clearly additional competition binding curves using ligands that can differentiate between (are selective for) M1, M2, M3 and M4 receptors would need to be conducted.

Note: Determining which of the 5 receptor subtypes are present within a cell/tissue may require conducting a competition binding study that utilises 5 competing ligands, each of which shows at least some selectivity for a particular receptor subtype – for example, using pirenzepine (M1-selective over M2, M3 and M5 receptors), methoctramine (M2-selective over M1, M3 and M5 receptors), darifenacin (M3-selective over M1, M2 and M4 receptors), MT-3 (M4-selective over all other M subtypes), and S-secoverine (M5-selective over all other M subtypes) – as shown in Example #4 below.

Consider an example where a competition binding study was conducted using the five ligands – pirenzepine, methoctramine, darifenacin, MT-3 and S-secoverine. Apply the SPIKES approach to the following competition binding curves to determine which M cholinoceptor subtype(s) is (are) present in this cell/tissue.

Shape / Position

Visual inspection of the all the curves indicates that that they are inverse sigmoidal, but are they steep sigmoidal or shallow sigmoidal? For competition binding curves to be considered steep sigmoidal then the extent of %Specific Binding covered by logIC50 ± 1.0 should be greater than 80% - if it is less than 80% then the curve is considered shallow sigmoidal. Notice in the Table below the analysis of the Shape / Position of each of the curves, and the conclusion that all competition binding curves are steep sigmoidal. This is suggestive that each of the competing ligands is binding to a single population of receptors for which they have a particular affinity (M1 or M2 or M3 or M4 or M5). It is unlikely that there is more than one receptor subtype present as none of the five competition binding curves is shallow sigmoidal.

Competing Ligand Visually-determined logIC50 value> Specific Binding (%) at logIC50-1 Specific Binding (%) at logIC50+1 Difference Shape
Pirenzepine -6.5 (-7.5) 90.9 (-5.5) 9.1 90.9 - 9.1 = 81.8% Steep
Methoctramine -7.7 (-8.7) 90.9 (-6.7) 9.1 90.9 - 9.1 = 81.8% Steep
Darifenacin -7.0 (-8.0) 90.9 (-6.0) 9.1 90.9 - 9.1 = 81.8% Steep
MT-3 -5.9 (-6.9) 90.9 (-4.9) 9.1 90.9 - 9.1 = 81.8% Steep
S-secoverine -7.9 (-8.9) 90.9 (-6.9) 9.1 90.9 - 9.1 = 81.8% Steep

IC50 / Ki value

As a very low concentration of radioligand was used in the study (compared to the Kd of the radioligand), the –logIC50  -logKi. The –logKi values are displayed in the table below along with the known –log Kivalues of these ligands for all receptor subtypes.

Competing ligand Visually-determined -logKi value Known -logKi values
M1 M2 M3 M4 M5
Pirenzepine 6.5 8.2 6.5 6.9 7.4 7.2
Methoctramine 7.7 6.7 7.7 6.0 7.0 6.3
Darifenacin 7.0 7.8 7.0 8.8 7.7 8.0
MT-3 5.9 6.7 5.9 6.0 8.1 6.0
S-secoverine 7.9 8.0 7.9 7.7 7.7 6.5

Elimination / Summation

By sequentially comparing the visually-determined and known –logKi values for each competing ligand and using the principles discussed above:

Pirenzepine eliminates the likelihood of M1 receptor being present in the cell/tissue
– visually-determined –logKi value of 6.5 is ≥ 1 log unit different than that expected for pirenzepine binding to the M1 receptor (8.2)

Methoctramine eliminates the likelihood of M3 and M5 receptors being present in the cell/tissue
– visually-determined –logKi value of 7.7 is ≥ 1 log unit different than that expected for methoctramine binding to the M3 (6.0) & M3 (6.3) receptors

Darifencin eliminates the likelihood of M3 receptors being present in the cell/tissue
– visually-determined –logKi value of 7.0 is ≥ 1 log unit different than that expected for darifenacin binding to the M3 receptor (8.8)

MT-3 eliminates the likelihood of M4 receptors being present in the cell/tissue
– visually-determined –logKi value of 5.9 is ≥ 1 log unit different than that expected for MT-3 binding to the M4 receptor (8.1)

S-secoverine eliminates the likelihood of M5 receptors being present in the cell/tissue
– visually-determined –logKi value of 7.9 is ≥ 1 log unit different than that expected for S-secoverine binding to the M5 receptor (6.5)

Thus, visual inspection of the presented competition binding curves indicates that the cell/tissue did not contain significant populations of M1 receptors (eliminated by pirenzepine), M3 receptors (eliminated by methoctramine and darifenacin), M4 receptors (eliminated by MT-3) or M5 receptors (eliminated by S-secoverine). Thus, the only reasonable conclusion is that the cell/tissue contains a homogeneous population of M2 receptors.

Note: The visually-determined–logKi values for each of the competing ligands is similar to their known –logKi values at M2 receptors, confirming the conclusion drawn.

Consider an example where a competition binding study was conducted using six ligands – 5 established ligands (pirenzepine, methoctramine, darifenacin, MT-3, S-secoverine) and the hypothetical ligand Ligand C. Apply the SPIKES approach to the following competition binding curves to determine which M cholinoceptor subtype(s) is (are) present in this cell/tissue.

Shape / Position

Competing Ligand Visually-determined logIC50 value> Specific Binding (%) at logIC50-1 Specific Binding (%) at logIC50+1 Difference Shape
Pirenzepine -7.1 (-8.1) 90.5 (-6.1) 9.5 90.5 - 9.5 = 81.0% Steep
Methoctramine -6.2 (-7.2) 90.5 (-5.2) 9.5 90.5 - 9.5 = 81.0% Steep
Darifenacin -8.3 (-9.3) 87.8 (-7.3) 11.4 87.8 - 11.4 = 76.4% Shallow
MT-3 -6.0 (-7.0) 90.9 (-5.0) 9.1 90.9 - 9.1 = 81.8% Steep
S-secoverine -6.9 (-7.9) 82.2 (-5.9) 12.7 82.2 - 12.7 = 69.5% Shallow
Ligand C biphasic with point of inflection at 40% Specific binding

Visual inspection indicates that two competition binding curves are shallow sigmoidal (darifenacin and S-secoverine) and one is biphasic (Ligand C), providing strong evidence that there is a heterogeneous population of receptors (more than one subtype of M receptor). The position of the point of inflection at the 40% Specific Binding level suggests that one receptor subtype makes up 40% of the population and another the remaining 60% of the receptor population.

IC50 / Ki value

As a very low concentration of radioligand was used in the study (compared to the Kd of the radioligand), the –logIC50  -logKi. For the steep sigmoidal curves of pirenzepine, methoctramine and MT-3 the –logIC50 = -logKi, whereas for the shallow sigmoidal curves (darifenacin, S-secoverine), the single –logIC50 cannot be used as a single –logKivalues as it is a composite value of the two different –logKi values at the two different receptors.

Elimination / Summation

It is helpful to firstly analyse the –logKi values of the steep sigmoidal curves, with the possibility of perhaps eliminating some receptor subtypes. In this example, there are 3 ligands that generated steep sigmoidal curves – pirenzepine, methoctramine and MT-3.

Competing ligand Visually-determined -logKi value Known -logKi values
M1 M2 M3 M4 M5
Pirenzepine 7.1 8.2 6.5 6.9 7.4 7.2
Methoctramine 6.2 6.7 7.7 6.0 7.0 6.3
MT-3 6.0 6.7 5.9 6.0 8.1 6.0

Pirenzepine eliminates the likelihood of M1 receptor being present in the cell/tissue
– visually-determined –logKi value of 7.1 is ≥ 1 log unit different than that expected for pirenzepine binding to the M1 receptor (8.2)

Methoctramine eliminates the likelihood of M2 receptors being present in the cell/tissue
– visually-determined –logKi value of 6.2 is ≥ 1 log unit different than that expected for methoctramine binding to the M2 receptor (7.7)

MT-3 eliminates the likelihood of M4 receptors being present in the cell/tissue
– visually-determined –logKi value of 6.0 is ≥ 1 log unit different than that expected for MT-3 binding to the M4 receptor (8.1)

Thus the cell/tissue does not contain significant populations of either M1, M2 or M4 receptors, indicating that the 2 receptor subtypes that are present are likely M3 and M5 receptors – but which one accounts for 40% of the population and which one accounts for the remaining 60%?

Answering this question requires analysing the –logKi values of the two separate phases of the biphasic curve of Ligand C.

Competing ligand Visually-determined -logKi value Known -logKi values
M1 M2 M3 M4 M5
Ligand C (1st phase) 10.0 7.0 7.1 6.9 10.0 10.1
Ligand C (2nd phase) 6.9 7.0 7.1 6.9 10.1 10.0

The first phase of the Ligand C competition binding curve had a –logKi value of 10.0 – so we can eliminate the likelihood of M1, M2 or M3 receptors, leaving just the possibility of M4 or M5 receptors. However, MT-3 has previously eliminated M4 receptors, so the only reasonable conclusion (based on this process of elimination is the first phase of the Ligand C competition binding curve reflects Ligand C binding to high affinity M5 receptors, which accounts for 60% of all M receptors.

Likewise, the second phase of the Ligand C competition binding curve had a –logKi value of 6.9 – so we can eliminate the likelihood of M4 or M5 receptors, leaving just the possibility of M1, M2 or M3 receptors. However, pirenzepine and methoctramine have previously eliminated M1 and M2 receptors, respectively, so the only reasonable conclusion (based on this process of elimination) is the second phase of the Ligand C competition binding curve reflects Ligand C binding to lower affinity M3 receptors, which accounts for 40% of all M receptors.

Thus, visual inspection of the presented competition binding curves indicates that the cell/tissue contains a mixture of 40%M3 and 60%M5 receptors.

Note: As expected, the ‘composite –logKi values’ obtained for the shallow sigmoidal curve of darifenacin (8.3) lay between the known –logKi values of darifenacin for M3 (8.8) and M5 (8.0) receptors, and the ‘composite –logKi values’ obtained for the shallow sigmoidal curve of S-secoverine (6.9) lay between the known –logKi values of S-secoverine for M3 (7.7) and M5 (6.5) receptors).

Consider an example where a competition binding study was conducted using the 5 established ligands pirenzepine, methoctramine, darifenacin, MT-3 and S-secoverine. Apply the SPIKES approach to the following competition binding curves to determine which M cholinoceptor subtype(s) is (are) present in this cell/tissue.Note that the competition binding curve for methoctramine (RED) cannot be clearly seen because it lays behind the curve for MT-3 (GREEN).

Shape / Position

Competing Ligand Visually-determined logIC50 value> Specific Binding (%) at logIC50-1 Specific Binding (%) at logIC50+1 Difference Shape
Pirenzepine -7.4 (-8.4) 82.7 (-6.4) 15.0 82.7 - 15.0 = 67.7% Shallow
Methoctramine -6.3 (-7.3) 89.1 (-5.3) 11.5 89.1 - 11.5 = 77.6% Shallow
Darifenacin -8.4 (-9.4) 87.0 (-7.4) 13.7 87.0 - 13.7 = 73.3% Shallow
MT-3 -6.3 (-7.3) 89.1 (-5.3) 11.5 89.1 - 11.5 = 77.6% Shallow
S-secoverine -7.8 (-8.8) 90.1 (-6.8) 9.1 90.1 - 9.1 = 81.0% Steep

As there are four competition binding curves that are shallow sigmoidal (pirenzepine, methoctramine, darifenacin and MT-3), it is clear that there is a heterogeneous population of receptors (more than one subtype of M receptor). There are no biphasic curves thus the relative densities of the receptor subtypes present cannot be determined visually – determination of the relative densities would require the fitting of the data to a two-site (or multiple-site) binding model, and solving using computer-assisted analysis.

IC50 / Ki value

As a very low concentration of radioligand was used in the study (compared to the Kd of the radioligand), the –logIC50 ~ -logKi . For the steep sigmoidal curve of S-secoverine the –logIC50 = -logKi, whereas for the shallow sigmoidal curves (pirenzepine, methoctramine, darifenacin and MT-3), the single –logIC50 cannot be used as a single –logKi values as it is a composite value of the two (or more) different –logKi values at the two (or more) different receptors.

Elimination / Summation

It is helpful to firstly analyse the –logKi values of the steep sigmoidal curves, with the possibility of perhaps eliminating some receptor subtypes. In this example, there is only 1 ligand that generated steep sigmoidal curves – S-secoverine.

Competing ligand Visually-determined -logKi value Known -logKi values
M1 M2 M3 M4 M5
S-secoverine 7.8 8.0 7.9 7.7 7.7 6.5

S-secoverine eliminates the likelihood of the M5 receptor being present in the cell/tissue
– visually-determined –logKi value of 7.8 is ≥ 1 log unit different than that expected for S-secoverine binding to the M5 receptor (6.5)

Thus the cell/tissue does not contain M5 receptors, but we can’t definitively determine whether it contains M1 and/or M2 and/or M3 and/or M4 receptors. Some indication of which of M1 or M2 or M3 or M4 receptors might be present can be obtained from further analysis of the shallow sigmoidal competition binding curves.

Competing ligand Visually-determined -logIC50 value Known -logKi values
M1 M2 M3 M4 M5*
Pirenzepine 7.4 8.2 6.5 6.9 7.4 -
Methoctramine 6.3 6.7 7.7 6.0 7.0 -
Darifenacin 8.4 7.8 7.0 8.8 7.7 -
MT-3 6.3 6.7 5.9 6.0 8.1 -

*M5 values not considered because this receptor subtype has already been eliminated.

It is unlikely that a significant population of M2 receptors is present in the cell/tissue because it would expected that the visually-determined –logIC50 value (6.3) for methoctramine would be much closer to the known –logKi value of methoctramine for M2 receptors (7.7).

Similarly, it is unlikely that a significant population of M4 receptors is present in the cell/tissue because it would expected that the visually-determined –logIC50 value (6.3) for MT-3 would be much closer to the known –logKi value of MT-3 for M4 receptors (8.1).

This might indicate that the receptor subtypes present are a mixture of M1 and M3 receptors. If this were the case, we would expect that the visually-determined –logIC50 values for all the competing ligands would lay between the –logKi values of the ligands for M1 and M3 receptors – and this is the case as can be seen in the Table below.

Competing ligand Visually-determined -logIC50 value Known -logKi values
M1 M2 M3 M4 M5*
Pirenzepine 7.4 8.2 - 6.9 - -
Methoctramine 6.3 6.7 - 6.0 - -
Darifenacin 8.4 7.8 - 8.8 - -
MT-3 6.3 6.7 - 6.0 - -
S-secoverine 7.8 8.0 - 7.7 - -

Thus, visual inspection of the presented competition binding curves indicate the cell/tissue:

1. does not contain a significant population of M5 receptors (eliminated by S-secoverine)

2. is unlikely to contain a significant population of M2 or M4 receptors (based on methoctramine and MT-3 data – BUT both these curves were shallow sigmoidal, so conclusions based on visual inspection are not definitive)

3. likely contains a mixture of M1 and M3 receptors, but of unknown relative density as no competition binding curve was biphasic. Further computer-assisted analysis of the competition binding curves using a two-site binding model is required to confirm these conclusions.

Note 1: This example has many caveats in the conclusion compared to previous examples #1 - #5.

Note 2: Neither of the other available established ligands (solifenacin and DAU-5884) would have been useful in distinguishing between M1 and M3 receptors in the example because both ligands have very similar affinities (-logKi values) for M1 and M3 receptors (see –logKi reference Table). On the other hand, either of the hypothetical ligands Ligand A or Ligand B would have been particularly useful for determining the M receptor subtypes present in this cell/tissue because both ligands have markedly different affinities for M1 and M3 receptors (1,000- and 1,000,000-fold, respectively, as per –logKi Reference Table).

Consider an example where a competition binding study was conducted using the 4 established ligands (S-secoverine, MT-3, pirenzepine and methoctramine) and 2 hypothetical ligands (Ligand A and Ligand B). Apply the SPIKES approach to the following competition binding curves to determine which M cholinoceptor subtype(s) is (are) present in this cell/tissue.

Shape / Position

Competing Ligand Visually-determined logIC50 value> Specific Binding (%) at logIC50-1 Specific Binding (%) at logIC50+1 Difference Shape
S-secoverine -7.9 (-8.9) 91.0 (-6.9) 9.8 91.0 - 9.8 = 81.2% Steep
MT-3 -6.2 (-7.2) 87.2 (-5.2) 10.3 87.2 - 10.3 = 78.9% Shallow
Pirenzepine -7.2 (-8.2) 78.0 (-6.2) 15.4 78.0 - 15.4 = 62.6% Shallow
Methoctramine -6.8 (-7.8) 83.8 (-5.8) 16.5 83.3 - 16.5 = 66.8% Shallow
Ligand A Biphasic with point of inflection at 30% Specific binding
Ligand B Triphasic with POIs at 60% and 30% Specific binding

Visual inspection indicates there is one steep sigmoidal curve (S-secoverine), three shallow sigmoidal curves (MT-3, pirenzepine and methoctramine), onebiphasic curve (Ligand A) and one triphasic curve (Ligand B), providing strong evidence that there is a heterogeneous population of receptors (at least 3 subtypes of M receptor). The position of the POIs at the 60% and 30% Specific Binding level for Ligand B suggests that one receptor subtype makes up 40% of the population and the other two make up 30% each (making a total of 100%).

IC50 / Ki value

As a very low concentration of radioligand was used in the study (compared to the Kd of the radioligand), the –logIC50 ~ -logKi . For the steep sigmoidal curve of S-secoverine the –logIC50 = -logKi, whereas for the shallow sigmoidal curves (MT-3, pirenzepine and methoctramine), the single –logIC50 cannot be used as a single –logKi values as it is a composite value of the three different –logKi values at the three different receptors.

Elimination / Summation

It is helpful to firstly analyse the –logKi values of the steep sigmoidal curve, with the possibility of perhaps eliminating some receptor subtypes. In this example, there is only one ligand that generated a steep sigmoidal curve – S-secoverine.

Competing ligand Visually-determined -logKi value Known -logKi values
M1 M2 M3 M4 M5
S-secoverine 7.9 8.0 7.9 7.7 7.7 6.5

S-secoverine eliminates the likelihood of the M5 receptor being present in the cell/tissue
– visually-determined –logKi value of 7.9 is ≥ 1 log unit different than that expected for S-secoverine binding to the M5 receptor (6.5)

Thus the cell/tissue does not contain a significant population of M5 receptors. Determining which of M1 and/or M2 and/or M3 and/or M4 receptors are present and their relative densities requires analysing the –logKi values of the different phases of the multiphasic curves. Let’s start with Ligand B because that is triphasic and may provide some key information.

Competing ligand Visually-determined -logKi value Known -logKi values
M1 M2 M3 M4 M5
Ligand B (1st phase) 8.0 8.0 5.0 2.0 5.0 8.0
Ligand B (2nd phase) 5.0 8.0 5.0 2.0 5.0 8.0
Ligand B (3rd phase) 2.0 8.0 5.0 2.0 5.0 8.0

The first phase of the Ligand B competition binding curve has a –logKi value of 8.0 – so we can eliminate the likelihood of M2, M3 or M4 receptors, leaving just the possibility of M1 or M5 receptors being responsible for this specific phase. However, S-secoverine has previously eliminated M5 receptors, so the only reasonable conclusion (based on this process of elimination) is the first phase of the Ligand B competition binding curve reflects Ligand B binding to high affinity M1 receptors, which accounts for 40% of all M receptors.

The second phase of the Ligand B competition binding curve has a –logKi value of 5.0 – so we can eliminate the likelihood of M1, M3 or M5 receptors, leaving just the possibility of M2 and/or M4 receptors being responsible for this particular phase.

The third phase of the Ligand B competition binding curve has a –logKi value of 2.0 – so we can eliminate the likelihood of M1, M2, M4 or M5 receptors, leaving just the possibility of M3 receptors being responsible for this specific phase. Thus, the only reasonable conclusion (based on this process of elimination) is the third phase of the Ligand B competition binding curve reflects Ligand B binding to low affinity M3 receptors, which accounts for 30% of all M receptors.

So with Ligand B, we have confirmed that M1 and M3 receptors are present in the relative proportions of 40% and 30% respectively. Thus, the remaining population of 30% M receptors must be M2 and/or M4 receptors. Can the Ligand A competition binding curve shed any light?

Competing ligand Visually-determined -logKi value Known -logKi values
M1 M2 M3 M4 M5
Ligand A (1st phase) 9.0 9.0 9.0 6.0 6.0 2.0
Ligand A (2nd phase) 6.0 9.0 9.0 6.0 6.0 2.0

The first phase of the Ligand A competition binding curve had a –logKi value of 9.0 – so we can eliminate the likelihood of M3, M4 or M5 receptors, leaving just the possibility of M1 and/or M2 receptors being responsible for this specific phase. This first phase of Ligand A binding accounts for 70% of Specific binding and we know from the analysis of Ligand B above that the cell/tissue contains 40%M1 receptors, so the remaining 30% of the binding of Ligand A in this phase must be to M2 receptors.

The second phase of the Ligand A competition binding curve had a –logKi value of 6.0 – so we can eliminate the likelihood of M1, M2 or M5 receptors being responsible for this phase, leaving just the possibility of M3 and/or M4 receptors being responsible for this particular phase. As this phase accounts for 30% of Specific binding and we have already established that the M3 subtype accounts for 30% of total M receptors present, we can eliminate the possibility that a significant population of M4 receptors being present in the cell/tissue.

Thus, visual inspection of the presented competition binding curves indicates that the cell/tissue contains a mixture of 40%M1, 30%M2 and 30%M3 receptors.

Note 1: As expected, the ‘composite –logKi values’ obtained for the shallow sigmoidal curve of MT-3 (6.2), pirenzepine (7.2) and methoctramine (6.8) lay between the known –logKi values of these ligands for M1, M2 and M3 receptors.

Note 2: As demonstrated in the Examples of competition binding studies presented above, the process of determining which receptors are present (and their relative densities) in a particular cell/tissue using the SPIKES approach is best conducted in a systematic way.

STEP 1. Identify the competition binding curves that are steep sigmoidal and analyse these first – this may lead to the elimination of particular receptor subtypes, making further analyses simpler.

STEP 2. Identify the competition binding curves that are clearly multiphasic – this will often provide important information regarding the relative densities of M receptor subtypes.

STEP 3. Analyse the shallow sigmoidal competition binding curves last, as visual inspection of these curves is likely to provide indicative (e.g. composite –logIC50 and –logKi values), rather than definitive information (true –logIC50 and –logKi values). Shallow sigmoidal competition binding curves are best analysed by fitting the data to more complex binding models (2-site, 3-site, etc.) using computer-assisted approaches. This approach enables the estimation of true –logKi values of the competing ligand to each of the multiple sites (rather than relying on the composite –logKi value obtained by visual inspection).

  • About
  • ⋅
  • Instructions
  • ⋅
  • Contact

Creative Commons 2018

Instructions

The Development Tool can be used to visualise how the shape and position of competition binding curves can be influenced by:

1.  the relative densities of receptor subtypes expressed by cells or tissues, and

2.  the affinities of ligands for the receptor subtypes.

The following instructions demonstrate how these changes in the shape and position of competition binding curves can be visualised using the Development Tool.

Firstly, you select which receptor subtypes are present in the cell or tissue of interest.

Within the Receptor window, select up to three receptor subtype combinations by clicking on the appropriate buttons below the name of the subtype. Once a particular subtype has been selected, the Relative Density (%) box appears immediately below.

If only one receptor subtype is selected, then the Relative Density (%) value will be 100.

If two receptor subtypes are selected, then the sum of their Relative Density (%) values must be 100, i.e. increasing the value of one will decrease the value of the other to maintain a total value of 100. The Relative Density (%) values can be changed by either by using the arrows for increments of 1%, or by typing in the desired value (0-100).

If three receptor subtypes are selected, then the combined value must again be 100. In this case, enter the receptor density values in order from left to right, i.e. start with the receptor subtype with the lowest number selected (e.g. M1) to the receptor subtype with the highest number (e.g. M5).

Once the first two receptor densities have been selected, the final density will automatically be calculated.

For example, if you wished to use a cell that expresses a combination of M2, M4 and M5 receptors in the relative proportions 20:30:50, respectively, then select the M2, M4 and M5 subtypes. In order, enter the values of 20% for M2 and 30% for M4. M5 will automatically be assigned a 50% density.

Secondly, you select which ligands you wish to use in the competition binding studies.

Within the Ligands window, use the drop down menu to select up to 6 ligands from a list of 8 Established ligands and 4 Hypothetical ligands.

Established ligands are ligands that have been used extensively in the past, and for which there are known -log Ki values (e.g. Atropine).

Hypothetical ligands, as their name suggests do not exist in the real world, but have been included in SPIKESbind to increase applicability.

There are 2 types of Hypothetical ligands - one type includes Ligand A and Ligand B, which have marked (but defined) differences in affinities for receptor subtypes.

The second type of Hypothetical ligand includes Ligand C and Ligand D, and these are ligands where you can assign (and change) a –log Ki value (affinity) for each receptor subtype. The assigned –log Ki values must be between 3 and 10 and may be changed by either using the arrows (increments of 0.1), or by typing in the desired value (3-10).

The ligands included in the drop down menu and their corresponding -log Ki values can be viewed in the '-log Ki Table' tab.

Lastly, once receptor subtypes and ligands have been selected, a plot of the specific binding % is displayed in real time within the Competition Binding Curves window.

By inserting the appropriate -log Ki values into the one-, two- or three-site binding models, accompanied by the selected subtype receptors, SPIKESbind will calculate the % specific binding values over the ligand range 10-12 - 10-2 M and will generate the Competition Binding Curves for viewing.

A feature of SPIKESbind is that any changes to the receptor subtype or the ligands selected will cause the Competition Binding Curves window to be dynamically updated in real time, to reflect the new values. For example, you can directly observe (in real time) changes in the shape and position of Competition Binding Curves caused by (1) altering the Relative Densities of receptor subtypes or (2) altering ligand affinity (use Ligands C (or D) to observe this effect).

In the Self-Assessment Tool, you choose up to 6 ligands (from the list of 12 Established and Hypothetical ligands) that you wish to use in a competition binding study. The Competition Binding Curves for these selected ligands are then displayed in the Competition Binding Curves window based on a receptor profile (the receptor subtypes and their relative densities) that has been randomly generated by SPIKESbind.

The task for you is to analyse the competition binding curves using the SPIKES approach, and deduce which receptor subtypes are present and their relative densities. The actual receptor profile can be revealed at any time by clicking the Reveal Subtype button.

Firstly, you select which ligands you wish to use in the competition binding studies.

From within the Ligands window, select up to 6 Ligands from a list of 8 Established ligands and 4 Hypothetical ligands. As indicated above, Established ligands are ligands that have been used extensively in the past, whereas Hypothetical ligands, as their name suggests, do not exist in the real world, and have been included in SPIKESbind to increase applicability.

There are 2 types of Hypothetical ligands - one type includes Ligand A and Ligand B, which have marked (but defined) differences in affinities for receptor subtypes.

The second type of Hypothetical ligand includes Ligand C and Ligand D, and these are ligands where you can assign (and change) a –log Ki value (affinity) for each receptor subtype. The assigned –log Ki values must be between 3 and 10 and may be changed by either using the arrows (increments of 0.1), or by typing in the desired value (3-10).

The ligands included in the drop down menu and their corresponding -log Ki values can be viewed in the '-log Ki Table' tab.

Secondly, the competition binding curves will appear for the selected ligands, showing the specific binding % over the ligand range 10-12 - 10-2 M.

The shape and position of the competition binding curves generated depends on which specific receptor subtypes are present and their relative densities (e.g. 30% M1 and 70% M2). The receptor profile is randomly selected by SPIKESbind from a defined bank of approximately 75 options.

The bank of 75 options is based on the premise (1) that there are 10 possible combinations involving two receptor subtypes (M1/M2, M1/M3, M1/M4, M1/M5, M2/M3, M2/M4, M2/M5, M3/M4, M3/M5 and M4/M5), and (2) that for each receptor combination there are 7 possible receptor density combinations (20/80, 30/70, 40/60/ 50/50, 60/40, 70/30 and 80/20) – giving a total of 70 options for two-receptor subtype combinations. This 70 options, together with another 5 options for pure receptor subtypes (100%), gives a grand total of 75 options from with SPIKESbind can choose to present. All 90%/10% and small incremental combinations have been omitted for your benefit as they cannot be reliably interpreted by eye.

Using the SPIKES approach to deduce which receptor subtypes are present and their relative densities. Your answer should be one of the 75 options.

Lastly, after you have worked out your answer, you may reveal the solution to validate your work.

You can reveal which receptor subtype population exists and their relative densities by clicking the Reveal Subtype button, and the answer will appear.

After viewing the solution, the ligands may be modified in the ligand table to view other competition binding curves, including the custom -log Ki values for ligands C and D, and the graph will dynamically update.

To generate a new question in the Self-Assessment Tool, click the Reset button and a new receptor subtype combination option will be randomly generated by SPIKESbind.

The Quiz Tool works off the same premise as the Self-Assessment tool, where any one question is randomly selected from a list of 75 options (i.e. up to two receptor subtypes with relative densities between 20-80%, using 10% increments).

However, the Quiz Tool is different from the Self-Assessment tool as the ligands used in the Competition Binding studies are pre-selected by SPIKESbind. In the Quiz Tool, SPIKESbind selects 5 ligands – 4 of which will be randomly selected from the list of 7 selective Established ligands (Pirenzepine, Methoctramine, Darifenacin, MT-3, S-Secoverine, Solifenacin and DAU-5884), and one Hypothetical ligand, either Ligand A or Ligand B. This provides the opportunity for over 500 different questions to be posed by the Quiz Tool – sufficient to gain expert proficiency in interpreting competition binding data.

Analyse the Competition Binding Curves using the SPIKES approach to solve which receptor subtypes are present, and what their densities are. Note that not all questions within the Quiz Tool can be answered by eye – for these questions you should be able to indicate which additional ligand(s) would allow the question to be answered unequivocally (and why).

Prior to commencing the quizzes, you may wish to read through the SPIKES examples to gain an insight into how to systematically approach solving competition binding data using the SPIKES approach.

Follow the instructions below to undertake the Quiz.

From the Quiz homepage, click the Start Quiz button, and the quiz will commence with the question appearing.

At the top of the page, you will notice the timer appear, counting up from 0 seconds until your answer is submitted. This allows for you to record the time spent evaluating each question, and over time you should be able to see your times improve. Once the quiz is complete you are taken to the answer review page.

The graph in question will appear on the right hand size, showing the 5 pre-selected ligands plotted with the randomised mixed receptor subtype combination.

Use this information to work out your solution (i.e. which receptor subtypes are present and their relative densities). Provided is a ‘SPIKES Approach’ table for you to help determine the answer. In this table you are able to view the list of ligands displayed in the graph, along with their corresponding -log Ki values, a drop down menu to determine the shape of the curve, a position textbox to enter information for the -logIC50 value, and checkboxes to eliminate receptor subtypes (i.e. apply the SPIKES approach).

Once you have completed your analysis, enter which receptor subtype(s) you believe to be present and their relative densities, using the checkboxes and drop down menus provided.

After you are happy with your answer, click the Submit button to head to the answer 'Review' page.

At the ‘Review’ page, "Correct!" or "Incorrect" will appear, as will the competition binding curve again. If you answered incorrectly, another graph will be displayed to view what competition binding curves your selection would have created. You have the option to Amend your Answers or to start a New Quiz.

About

SPIKESbind is an educational application to boost students' proficiency in interpreting competition binding data by providing self-development, self-assessment and quiz-me options.

Most drugs work by binding to specific receptors, which are large proteins typically expressed on the surface of cells.

The process of drug binding requires the existence of affinity (chemical forces of attraction) between the drug and receptor, and can be readily described by relatively simple mathematical relationships.

As drug binding is essential for drug action, pharmacology students need to have an advanced level of understanding of how changes in the affinity of a drug for a receptor can influence binding, and be able to readily interpret drug binding data.

SPIKESbind was created by Madeline King, Jason Gan, Robert Fernandez and Timothy Mennell; students of CITS3200: Professional Computing at the University of Western Australia 2017, under the guidance of Associate Professor Peter Henry.

Cheng Y. and Prusoff W.H. (1973) Relationship between the inhibition constant (Ki) and the concentration of an inhibitor that causes a 50% inhibition (I50) of an enzymatic reaction. Biochem. Pharmacol. 22: 3099–3108.

D'Agostino G, Condino AM, Gioglio L, Zonta F, Tonini M, Barbieri A (2008). Isolated porcine bronchi provide a reliable model for development of bronchodilator anti-muscarinic agents for human use. Br J Pharmacol. 154: 1611-8.

DeLean A, Hancock AA, Lefkowitz RJ (1982). Validation and statistical analysis of the computer modeling method for quantitative analysis of radioligand binding data for mixtures of pharmacological subtypes. Mol. Pharmacol. 21: 5–16.

Eglen RM, Nahorski SR (2000). The muscarinic M(5) receptor: a silent or emerging subtype? Br J Pharmacol. 130:13-21.

Eglen RM, Choppin A, Watson N (2001). Therapeutic opportunities from muscarinic receptor research. Trends Pharmacol Sci. 22:409-14.

Flanagan CA (2016). GPCR-radioligand binding assays. Methods Cell Biol. 132:191-215.

Hein P., Michel MC, Leineweber K, Wieland T, Wettschureck N, Offermanns S (2005). Receptor and Binding Studies. In: Dhein S., Mohr F.W., Delmar M. (eds) Practical Methods in Cardiovascular Research. Springer, Berlin, Heidelberg.

Hulme EC, Trevethick MA. (2010) Ligand binding assays at equilibrium: validation and interpretation. Br J Pharmacol. 161:1219-37.

Kenakin TP (2004) A Pharmacology Primer: Theory, Application, and Methods. First Edn. Elsevier Academic Press, San Diego, CA, USA.

Limbird LE, Motulsky H (2011). Receptor Identification and Charcterisation. Suppl. 20. Handbook of Physiology, The Endocrine System, Cellular Endocrinology. https://doi.org/10.1002/cphy.cp070104

Liu H, Hofmann J, Fish I, et al. (2018) Structure-guided development of selective M3 muscarinic acetylcholine receptor antagonists. Proc Natl Acad Sci U S A. 115:12046-12050.

Mansfield KJ, Liu L, Mitchelson FJ, Moore KH, Millard RJ, Burcher E (2005). Muscarinic receptor subtypes in human bladder detrusor and mucosa, studied by radioligand binding and quantitative competitive RT-PCR: changes in ageing. Br J Pharmacol. 144:1089-99.

Neubig RR, Spedding M, Kenakin T, Christopoulos A, International Union of Pharmacology Committee on Receptor Nomenclature and Drug Classification (2003). International Union of Pharmacology Committee on Receptor Nomenclature and Drug Classification. XXXVIII. Update on terms and symbols in quantitative pharmacology. Pharmacol Rev. 55:597-606.

Ohtake A, Saitoh C, Yuyama H, Ukai M, Okutsu H, Noguchi Y, Hatanaka T, Suzuki M, Sato S, Sasamata M, Miyata K (2007). Pharmacological characterization of a new antimuscarinic agent, solifenacin succinate, in comparison with other antimuscarinic agents. Biol Pharm Bull. 30:54-8.

Contact

Found a bug in our application?
Or simply wanted to give some feedback?
Please don't hesitate to let us know!

Peter Henry
Associate Professor
The University of Western Australia

Room 1.34, M Block,
School of Biomedical Sciences,
QEII Medical Centre

peter.henry@uwa.edu.au