Essay: Uncovering Kinetic Screen of CCR2 Intracellular Allosteric Antagonists



Kinetic screen of intracellular allosteric antagonists for CCR2 .

‡Ussama I. Choudhry, ‡Tom M. Schubert, ‡Michiel B. Otte, Lisa S. den Hollander*

Division of Medicinal Chemistry, Leiden Academic Center for Drug Research, Leiden University, Post Office Box 9502, 2300 RA Leiden, Netherlands

 

INTRODUCTION

Chemokines are small molecules with chemoattracting properties[1][2]. These chemokines are used for guiding our immune system to places of injury or inflammation through a changing concentration gradient until the leukocyte (or other immune cells) reaches the site. When the expression of chemokines is dysregulated, the immune system is not working as it is supposed to. Keeping that in mind, this may be the reason as to how various human diseases like immunodeficiency, autoimmune disease and cancer could develop. These days, there are approximately 20 receptors and 50 chemokines recognized, what makes it so hard to research them is that one chemokine may bind to multiple chemokine receptors and also the other way around, one receptor may interact with multiple chemokines. Inflammation is mainly guided by 2 subfamilies of chemokines, CXC and CC. The CXC chemokines get induced by inflammatory stimuli, including bacterial membranes, tumour necrosis factor cytokines, and interleukin cytokines. The CC chemokines seem to be involved in mobilising monocytes into inflamed tissues[1]. With exception to CCR5, there has been no breakthrough in finding drugs relating to chemokine receptors[3]. The reason why it is so hard to develop new therapeutics is that the receptor is redundant, so there is almost always a back-up plan in the cells. Also the animal models are unpredictable. The first problem can be solved if there is a synthesised drug which works on multiple receptors, but for this, a lot of research needs to be done[3].

For this research project we looked at the allosteric binding site for the receptor. On the orthosteric binding site the chemokine ligand, and the newly synthesised drug, would be continuously in competition with each other. Since the main goal is to dose the drug in a human body, the competition with the chemokine ligand would make it difficult to dose the drug properly, because the drug would probably be in a disadvantageous concentration compared to the chemokine concentration. This is not the case for the allosteric site. The compounds which bind to the allosteric site have the consequence of altering the binding and/or potency of the receptor.

Figure 1. Structure of LUF7591 (top) and CCR2-RA-[R] (bottom)

Table 1. Triazolo pyrimidine scaffold, R-group on the scaffold, mean KRI, the KRI values, and Kinetic Parameters (RT, koff, kon) for 12 different compounds.

Compound R1 Mean KRI KRI values ckon (nM-1min-1)

ckoff (min-1) RT (min)

CCR2-RA-[R]a – 0,7122 0,030595 (SEM)b 6,33•109  7,63•1010 0,6412  0,0356 1,6  0,09

LUF7591 – 1,1720 0,0554077 (SEM)b d1,09•10-7 d0,01345 d74,3

LUF7621 2-Cl 0,7127 0,74338/0,68204 – – –

LUF7657 3-Br 0,7492 0,80154/0,69695 – – –

LUF7521 3-Cl 0,7790 0,76122/0,79687 – – –

LUF7690 3-F 0,8856 0,87659/0,894526 4,54•106  6,24•1019 0,12029  ,02261

8,3  1,56

LUF7677 3-I 0,7848 0,74684/0,82282 – – –

LUF7645 3-Me 0,7124 0,71960/0,70524 – – –

LUF7624 3-Ome 0,7680 0,74790/0,78805 – – –

LUF7689 4-Br 0,8257 0,85633/0,79499 – – –

LUF7642 4-Cl 0,7641 0,73346/0,79482 – – –

LUF7688 4-F 0,7171 0,67023/0,76403 – – –

aControl compound. bThe SEM of 2 individual experiments performed in duplicate. cMean from 2 individual experiments  SEM. dThe experiment was done once

RESULTS AND DISCUSSION

Dual point competition association assay of triazolo pyrimidines to [3H]CCR2-RA-[R]

A dual point competition association assay was done, where [3H]CCR2-RA-[R] was in competition with 12 different compounds. With this assay, the KRI values in table 1 were calculated using formula (3).

From table 1, the compounds with a 3-halogen show relatively higher KRI values. The compounds also seem to show an increasing KRI with increasing electronegativity. Looking at the compounds with a 4-halogen it seems as if bigger and/or more electronegative atoms have higher KRI values. Not much can be said about the compound with the 2-Cl, because it is the only compound with a substituent at the 2-position.

When looking at the affinity[4], the compound with a 3-F R-group shows smaller affinity, but a high KRI. It is not clear as to why, and thus more research needs to be done on this.

The only compound which shows significantly higher KRI compared to the control compound CCR2-RA-[R] is LUF 7690. What this means is that LUF7690 has a longer residence time, which is useful in drug effectiveness.

However, the KRI value of LUF 7591 is still higher than that of LUF 7690. This means that the residence time of LUF7690 is shorter than LUF 7591.

The compound used to further analyse in the competition association assay was LUF7690 since this showed the highest mean KRI amongst the new compounds

competition association assay of LUF7690 to [3H]CCR2-RA-[R]

A competition association assay was done, where [3H]CCR2-RA-[R] was in competition with LUF 7690. Nevertheless, the results calculated were not reliable when using the Motulsky and Mahan 1984 model.

The curve seemed to show a biphasic behaviour instead of a monophasic behaviour, which is what the model is based on. At the end of the curve it appears to still increase as time would go on. In a monophasic curve, the line would eventually reach a plateau.

This would also mean that the KRI values measured and calculated are higher than they are supposed to be, because the curve was still climbing and thus the KRI was probably decreasing after the measurement was taken.

However, to approximate the kinetic parameters, the final timepoint (240 min.) was left out to try and fit the non-linear regression curve. This worked out successfully, but the final results are still not completely reliable.

Figure 2. Competition association assay. [3H]CCR2-RA-[R] in competition with LUF7690 at 4x IC50 concentration. Both curves represent an N=2 experiment executed in duplicate.

The residence time of 2 individual experiments was calculated using formula (4) and the mean of the values  standard deviation is shown in table 1. The residence time seems to supports the idea of LUF7690 having a higher KRI and thus a longer residence time. It also proves that LUF7591 is a better compound than LUF7690, due to the longer residence time of LUF7591

The kon was also calculated using formula (5) and documented in the same manner as the koff and residence time, in table 1. From the kon values, it seems that CCR2-RA-[R] and LUF7690 have a substantially higher value, what would mean that there are numerous associations per M*min, and so there is a smaller concentration of the compound needed to get the same association as LUF7591, which is a property since the dosage of the compound would be much smaller when using LUF7690 than LUF7690.

The competition association assay was previously done with a pH of 7.4 and pH 6.5, using the compound LUF7591. In these assay, the effect of pH was shown. The pH 6.5 assay showed that there was an overshoot in the first part of the curve which translates to a KRI >1. When using a pH of 7.4, there was no noticeable overshoot. This is why for the research done, a pH of 6.5 was used, even though this does not resemble the pH of human blood. In the appendix are 2 figures where the overshoot and the absence of overshoot are displayed.

CONCLUSION

The compound LUF7690 showed the most promising KRI and also a slightly longer residence time, compared to CCR2-RA-[R]. Additionally, the competition association assay showed a biphasic trend in the control compound, instead of a monophasic trend, which makes the results gathered not reliable to analyse with the Motulsky and Mahan 1984 model, and means that a different radioligand needs to be used to improve the assay.

EXPERIMENTAL SECTION

Membrane Cultivation

TANGOTM U2OS-CCR2_bla cells were used in the assays for this article since they were modified to stably express several proteins of which the most important one, the CCR2 receptor (expressed on the membranes). The cells were cultivated on McCoy’s 5A medium.

Membrane Harvest

The cells were harvested by aspirating the medium and adding PBS to the plates and also rinsed with PBS after collecting the initially added PBS. The cells were harvested on 4 December 2018 and stored at -80 C. This batch will be referred as Batch B since there was another batch, also to be used, made on 29 October 2018 which will be referred as Batch A. All the next procedures were done using both batches.

Membrane Preparation

In the first step, cold 50mM Tris-HCL based buffer with pH 6.5 was added to resuspend and afterwards also rinsed with the same buffer. The suspension was homogenized with the Ultra Turrax homogenizer and centrifuged at 31000 rpm and 4C using the ultracentrifuge Optima LE-80K. Afterwards the pellet was resuspended and centrifuged again in a smaller volume. The final volume was separated in aliquots of 250 L and 100 L. These aliquots were then stored at -80 C.

The protein concentration of the membranes was done using a BCA protein determination assay[5]. Measured with Wallac EnVisionTM.

Dual point competition association assay of triazolo pyrimidines to [3H]CCR2-RA-[R]

[3H]CCR2-RA-[R] was incubated in a 96-wells plate with the IC50 of 10 different compounds, Assay Buffer (AB) (cold 50mM Tris-HCL pH 6.5, 5mM MgCl2 0.1% CHAPS) and CCR2 membranes. The binding of the radioligand was measured at 2 timepoints (40 min. and 240 min.) and the association started the moment the membrane proteins were added. At 0 minutes, the plate was harvested using the Perkin Elmer Filtermate-harvester for 96-wells format onto GF/B filters pre-wetted with washing buffer (cold 50mM Tris-HCL pH 6.5, 5mM MgCl2 0.05% CHAPS). The radiation was counted using the MicroBeta plate counter. The concentration of the radioligand was determined using the formulas:

SA(t)=SA(0)∙(〖1/2)〗^(t/t_(1/2) ) (1)

Concentration (nM)= mean TC/(222∙SA) (2)

Where SA(t) is the specific activity at time = t, SA(0) is the specific activity at production (59.6 Ci/mmol), t is the time since production, t1/2 is the half-life of tritium and TC is the total counts.

This assay was done to find out what the KRI values are of the different compounds used. This was done using the formula:

KRI=t_1/t_2 (3)

Where t1 is binding at timepoint 40 minutes and t2 is binding at timepoint 240 minutes.

Competition association assay of triazolo pyrimidines to [3H]CCR2-RA-[R]

This assay was done to make full specific binding curves of the compounds which showed the highest KRI values. Instead of having 2 timepoints, there were now 11 timepoints where the first timepoint was removed. The timepoints used, in minutes, were as follows: 245, 240, 180, 120, 75, 40, 30, 20, 14, 6, 2 and 0 for harvesting. With this assay, the kon and koff of the compounds was determined and used to calculate the retention time with the formula:

RT=1/k_off (4)

Where RT is the retention time and koff is the disassociation rate of the compound.

The kon was calculated using the following formula:

k_obs=k_off  〖+ k〗_on∙[L] (5)

Where kobs is the association rate determined with the non-linear regression analysis, koff is the disassociation rate determined with the non-linear regression analysis and [L] is the radioligand concentration.

Data analysis

The data was analysed using the program GraphPad Prism 5. In there, the function non-linear regression analysis based on the Motulsky and Mahan 1984 model was used to plot lines through the data.

REFERENCES

[1] A. Viola, A.D. Luster, Chemokines and Their Receptors: Drug Targets in Immunity and Inflammation, Annu. Rev. Pharmacol. Toxicol. 48 (2008) 171–197. doi:10.1146/annurev.pharmtox.48.121806.154841.

[2] Y. Zheng, L. Qin, N.V.O. Zacarías, H. de Vries, G.W. Han, M. Gustavsson, M. Dabros, C. Zhao, R.J. Cherney, P. Carter, D. Stamos, R. Abagyan, V. Cherezov, R.C. Stevens, A.P. IJzerman, L.H. Heitman, A. Tebben, I. Kufareva, T.M. Handel, Structure of CC chemokine receptor 2 with orthosteric and allosteric antagonists., Nature. 540 (2016) 458–461. doi:10.1038/nature20605.

[3] R. Horuk, Chemokine receptor antagonists: Overcoming developmental hurdles, Nat. Rev. Drug Discov. (2009). doi:10.1038/nrd2734.

[4] N. V Ortiz Zacarías, J.P.D. Van Veldhoven, L. Portner, E. Van Spronsen, S. Ullo, M. Veenhuizen, W.J.C. Van Der Velden, A.J.M. Zweemer, R.M. Kreekel, K. Oenema, E.B. Lenselink, L.H. Heitman, A.P. Ijzerman, Pyrrolone Derivatives as Intracellular Allosteric Modulators for Chemokine Receptors: Selective and Dual-Targeting Inhibitors of CC Chemokine Receptors 1 and 2, (2018). doi:10.1021/acs.jmedchem.8b00605.

[5] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J. Olson, D.C. Klenk, Measurement of protein using bicinchoninic acid, Anal. Biochem. (1985). doi:10.1016/0003-2697(85)90442-7.

Appendix

ACKNOWLEDGMENT

Written by Ussama Choudhry s2397129

Special thanks to our supervisor Lisa den Hollander for being there emotionally and intellectually.

Thanks to the department of Medicinal Chemistry at the Leiden Academic Center for Drug Research for letting us use the laboratoria.

The experiments were done together with Tom Schubert and Michiel Otte.