Design and in vitro Characterization of Highly Sst₂-selective Somatostatin Antagonists Suitable for Radio-Targeting

Abstract

Radiolabeled sst₂ and sst₃ antagonists are better candidates for tumor targeting than agonists with comparable binding characteristics.¹ Because the majority of neuroendocrine tumors express sst₂, we used the known antagonists Acetyl-pNO₂Phe²-c[dCys³-Tyr⁷-dTrp⁸-Lys⁹-Thr¹⁰-Cys¹⁴]-dTyr¹⁵-NH₂ (1)^2,3 and H-Cpa²-c[dCys³-Tyr⁷-dTrp⁸-Lys⁹-Thr¹⁰-Cys¹⁴]-2Nal¹⁵-NH₂ (7)⁴ as leads for analogues with increased sst₂ binding affinity and selectivity. Among the 32 analogues reported here, DOTA-pNO₂Phe-c[dCys-Tyr-dAph(Cbm)-Lys-Thr-Cys]-dTyr-NH₂ (3) and DOTA-Cpa-c[dCys-Aph(Hor)-dAph(Cbm)-Lys-Thr-Cys]-dTyr-NH₂ (31) had the highest sst₂ binding affinity and selectivity. All of the analogues tested kept their sst₂ antagonistic properties (i.e. did not affect calcium release in vitro and competitively antagonized the agonistic effect of [Tyr³]-octreotide). Moreover, in an immunofluorescence-based internalization assay, the new analogues prevented sst₂ internalization induced by the sst₂ agonist [Tyr³]-octreotide, without being active by themselves. In conclusion, several analogues (in particular 3, 31 and 32) have outstanding sst₂ binding and functional antagonistic properties and, because of their DOTA moiety, are excellent candidates for in vivo targeting of sst₂-expressing cancers.

Introduction

The limitations of therapeutic applications of somatostatin (SRIF) due to its rapid proteolytic degradation and multiple targets have led to the search for peptide analogues with higher metabolic stability and improved selectivity with respect to the five SRIF receptor subtypes. Long-acting preparations of octreotide⁵ and lanreotide⁶ are now available for use in the treatment of acromegaly, neuroendocrine tumors, and various gastrointestinal disorders. Moreover, targeting neuroendocrine tumors expressing SRIF receptor subtypes with radiolabeled SRIF agonists is an established diagnostic and therapeutic approach in oncology. Somatostatin receptor scintigraphy with ¹¹¹In-DTPA-octreotide (¹¹¹In-diethylenetriaminepentaacetyl-octreotide) is the current imaging technique for the localization of neuroendocrine tumors⁷ whereas ¹⁷⁷Lu or ⁹⁰Y-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetyl-[Tyr³]-octreotate (⁹⁰Y-DOTA-TATE) are the radioligands used for tumor radiotherapy.^8,9 Radio-guided surgery for the in situ detection of neuroendocrine tumors during surgery also utilizes ¹¹¹In-DTPA-octreotide as a tool for tumor localization. The development of more selective analogues with high affinity and enhanced uptake by the SRIF receptor containing tumor cells is expected to open new and more sensitive avenues for radiotherapy and radio-guided surgery.^10-12 Classically, this can be achieved using established approaches used for drug design whereby the physico-chemical properties of the analogues are systematically modulated leading to stepwise improvements. For example, hydrophobicity, ionic charges, stabilization of secondary structures and, in the case of somatosatin analogues for radiotherapy, exhaustive modifications of the chelator moiety and of the radioactive metal have been reported.^13-19

While SRIF agonists readily internalize into tumor cells, permitting accumulation of radioactivity, radiolabeled antagonists do not, and therefore have not been considered for tumor targeting until recently.¹ We reported that the macrocyclic chelator DOTA-coupled sst₃ and sst₂-selective antagonists did not trigger sst₃ or sst₂ internalization, prevented agonist-stimulated internalization, yet were excellent in vivo tumor markers.^1,20 Potent agonists with strong binding and internalization properties showed a much lower and shorter-lasting uptake in SRIF receptor expressing tumors than the tested antagonists. The amount of uptake of the antagonist radioligand was particularly high in the tested tumors: 60% IA/g uptake has indeed never been achieved before by any radiolabeled SRIF receptor agonist, not even by those developed most recently.^11,12 Not only was the uptake at the peak time point very high, but also the long-lasting accumulation of the antagonist radioligand up to 72 h after injection was a remarkable result and represented a considerable advantage over radio-targeting with established agonists. We concluded that SRIF antagonist radiotracers are, therefore, preferable over agonists for the in vivo targeting of sst₃- or sst₂-expressing tumors. The use of potent radiolabeled antagonists for in vivo tumor targeting may considerably improve the sensitivity of diagnostic procedures, the staging of the disease, the detection of unexpected tumor sites and the efficacy of receptor-mediated radiotherapy and complementary procedures.^10,21-23

To generate pure sst₂ antagonists for therapeutic applications and because the great majority of neuroendocrine tumors express predominantly sst₂, we have focused the present study on the development of potent, highly sst₂-selective unlabeled and DOTA-labeled antagonists. This was achieved with the introduction of novel amino acid derivatives within the sequence of octreotide amide, and documented with binding assays to the five human sst_s and several functional assays such as internalization assays and calcium release.

Results and Discussion

All of the analogues shown in Table 1 were synthesized either manually or automatically on a MBHA resin using the Boc-strategy, diisopropylcarbodiimide (DIC)/HOBt (1-hydroxybenzotriazole) for amide bond formation and trifluoroacetic acid (TFA) for Boc removal. The peptide resins were treated with hydrogen fluoride (HF) in the presence of scavengers to liberate the fully deprotected crude linear peptides. Cyclization of the cysteines was mediated by iodine in an acidic milieu. Purification was carried out using multiple HPLC steps.²⁴ DOTA was coupled to the Lys(Fmoc)⁹ protected analogues in solution. The purity of the peptides was characterized by HPLC,²⁴ capillary zone electrophoresis²⁵ and mass spectrometry. The observed monoisotopic mass (M + H)⁺ values of each peptide correspond to the calculated mass (M) values. Results are shown in Table 1.

Table 1.

Physico-chemical Properties of Sst₂ Antagonists

	Structure of SRIF analogues Residues are numbered according to SRIF numbering H-Ala1-Gly2-c[Cys3-Lys4-Asn5-Phe6-Phe7-Trp8-Lys9-Thr10-Phe11-Thr12-Ser13-Cys14]-OH (SRIF)										Purity		MSc
											HPLCa	CZEb	Mcalc	M+Hobs
	Substitution in H-dPhe2-c[Cys3-Phe7-dTrp8-Lys9-Thr10-Cys14]-Thr15-ol (octreotide)
	N-terminus	2	3	7	8	9	10	14	15	C-terminus
12	Ac-	pNO2Phe-	dCys-	Tyr-	dTrp-	Lys-	Thr-	Cys-	dTyr-	NH2	99	99	1196.44	1197.59
2	DOTA-	pNO2Phe-	dCys-	Tyr-	dTrp-	Lys-	Thr-	Cys-	dTyr-	NH2	95	97	1540.61	1541.46
3	DOTA-	pNO2Phe-	dCys-	Tyr-	dAph(Cbm)-	Lys-	Thr-	Cys-	dTyr-	NH2	99	99	1559.63	1560.83
4	H2N -	pNO2Phe-	dCys-	Tyr-	dAph(Cbm)-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1207.47	1208.54
5	DOTA-	pNO2Phe-	dCys-	Tyr-	dAph(Cbm)-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1593.65	1594.17
6	H2N -	pNO2Phe-	dCys-	Aph(Hor)-	dAph(Cbm)-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1346.50	1347.59
74	H2N-	Cpa-	dCys-	Tyr-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1177.43	1178.43
827	H2N-	Cpa-	DCys-	Tyr-	DTrp-	NMeLys-	Thr-	Cys-	2Nal-	NH2	97	99	1191.45	1192.52
9	H2N-	Cpa-	dCys-	L-Agl(NMe, benzoyl)	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	95	98	1204.45	1205.51
10	H2N-	Cpa-	dCys-	D-Agl(NMe, benzoyl)	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	98	99	1204.45	1205.48
11	H2N -	Cpa-	dCys-	Leu-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1127.45	1128.46
12	H2N-	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	99	97	1219.45	1220.12
13	Cbm-	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	96	98	1262.46	1263.40
14	DOTA-	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	99	98	1605.64	1606.50
15	DOTA-β Ala-	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	99	98	1676.67	1677.67
16	DOTA-Peg-	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1809.62	1810.24
17	H2N-	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	-	NH2	99	99	1022.37	1023.49
18	H2N -	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	Cha-	NH2	99	96	1175.48	1176.36
19	H2N -	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	Aph(Hor)-	NH2	99	99	1324.47	1325.55
20	H2N -	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	dAph(Cbm)-	NH2	99	99	1227.45	1228.45
21	H2N -	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	Aph(Cbm)-	NH2	99	99	1227.45	1228.37
22	H2N -	Cpa-	dCys-	Aph(Cbm)-	dTrp-	Lys-	Thr-	Cys-	dAph(Cbm)-	Gly-OH	98	98	1285.46	1286.34
23	H2N -	Cpa-	dCys-	Aph(CONH-OCH3)-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1249.46	1250.56
24	H2N -	Cpa-	dCys-	Aph(CONH-OH)-	dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1235.44	1236.47
25	H2N -	Cpa-	dCys-	Aph(Cbm)-	5F-dTrp-	Lys-	Thr-	Cys-	2Nal-	NH2	96	96	1237.44	1238.44
26	H2N -	Cpa-	dCys-	Aph(Cbm)-	5F-Trp-	Lys-	Thr-	Cys-	2Nal-	NH2	96	85	1237.44	1238.24
27	H2N -	Cpa-	dCys-	Tyr-	dAph(Cbm)-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1196.43	1197.36
28	DOTA-	Cpa-	dCys-	Tyr-	dAph(Cbm)-	Lys-	Thr-	Cys-	2Nal-	NH2	99	99	1582.62	1583.72
29	H2N -	Cpa-	dCys-	Aph(Hor)-	dAph(Cbm)-	Lys-	Thr-	Cys-	2Nal-	NH2	91	95	1335.47	1336.44
30	DOTA-	Cpa-	dCys-	Aph(Hor)-	dAph(Cbm)-	Lys-	Thr-	Cys-	2Nal-	NH2	95	94	1721.65	1722.56
31	DOTA-	Cpa-	dCys-	Aph(Hor)	dAph(Cbm)-	Lys-	Thr-	Cys-	dTyr-	NH2	96	97	1687.64	1688.83
32	DOTA-	pNO2Phe-	dCys-	ITyr-	dTrp-	Lys-	Thr-	Cys-	dTyr-	NH2	99	99	1666.52	1667.74

^aPercent purity determined by HPLC using buffer system: A = TEAP (pH 2.5) and B = 60% CH₃CN/40% A with a gradient slope of 1% B/min, at flow rate of 0.2 mL/min on a Vydac C₁₈ column (0.21 × 15 cm, 5-μm particle size, 300 Å pore size). Detection at 214 nm.

^bCapillary zone electrophoresis (CZE) was done using a Beckman P/ACE System 2050 controlled by an IBM Personal System/2 Model 50Z and using a ChromJet integrator. Field strength of 15 kV at 30 °C, mobile phase: 100 mM sodium phosphate (85:15, H₂O:CH₃CN) pH 2.50, on a Supelco P175 capillary (363 μm OD × 75 μm ID × 50 cm length). Detection at 214 nm.

^cThe calculated m/z of the monoisotope compared with the observed [M + H]⁺ monoisotopic mass.

To investigate their sst_s binding properties, the peptides were tested for their ability to bind to cryostat sections from membrane pellets of cells expressing the five human sst_s (Table 2). For each of the tested compounds, complete displacement experiments with the universal SRIF radioligand [Leu⁸,DTrp²², ¹²⁵I-Tyr²⁵]SRIF-28. Results are shown in Table 2.

Table 2.

Sst_1-5 Binding Affinities (IC₅₀s, nM) and Function of Sst₂-Selective Analogues

	IC50 (nM)a					Functional Characterization
	sst1	sst2	sst3	sst4	sst5	sst2- internalization	Calcium assay
SRIF-28	2.7 ± 0.2	2.7 ± 0.2	3.3 ± 0.4	2.6 ± 0.4	2.4 ± 0.2	-	-
1	>1000	3.6 ± 0.4	>1000	349 ± 30	276 ± 119	Antagonist	-
2	>1000	1.5 ± 0.4	>1000	287 ± 27	>1000	Antagonist	Antagonist
3	>1000	0.75 ± 0.2	>1000	>1000	>1000	Antagonist	Antagonist
4	>1000	2.6 ± 0.7	384 ± 97	>1000	>1000	Antagonist	-
5	>1000	1.3 ± 0.2	>1000	>1000	>1000	Antagonist	Antagonist
6	>1000	2.7 ± 0.6	451 ± 80	>1000	>1000	-	-
7	>1000	5.7 ± 1.5	112 ± 32	296 ± 19	218 ± 63	Antagonist	Antagonist
8	>1000	10 ± 3.5	61 ± 14	715 ± 137	53 ± 19	Antagonist	Antagonist
9	>1000	17 ± 5	827 ± 244	>1000	442 ± 254	Antagonist	-
10	>1000	158 ± 37	102 ± 10	116 ± 47	728 ± 272	-	-
11	>1000	58 ± 21	340 ± 77	908 ± 138	657 ± 299	-	-
12	>1000	6.9 ± 0.7	155 ± 29	479 ± 8	149 ± 37	Antagonist	Antagonist
13	>1000	23 ± 4.3	54 ± 15	136 ± 7.5	111 ± 17	-	-
14	>1000	9.8 ± 1.2	972 ± 212	831 ± 82	>1000	Antagonist	Antagonist
15	>1000	46 ± 13	124 ± 53	>1000	>1000	Antagonist	Antagonist
16	>1000	40 ± 1.5	88 ± 13	728 ± 158	895 ± 294	Antagonist	-
17	>1000	5.9 ± 1.8	138 ± 52	>1000	461 ± 106	Antagonist	Antagonist
18	>1000	4.1 ± 0.9	255 ± 79	>1000	247 ± 66	-	-
19	>1000	27 ± 3.8	162 ± 19	>1000	320 ± 69	-	-
20	>1000	5.4 ± 1	328 ± 69	800 ± 295	191 ± 49	-	-
21	>1000	15 ± 3	336 ± 46	551 ± 151	560 ± 144	-	-
22	>1000	52 ± 4.7	661 ± 115	>1000	810 ± 200	-	-
23	>1000	9.3 ± 0.9	157 ± 49	883 ± 174	313 ± 35	-	-
24	>1000	9.3 ± 1.4	120 ± 45	813 ± 152	426 ± 189	-	-
25	>1000	4.9 ± 1.5	50 ± 5.8	287 ± 64	94 ± 34	Antagonist	Antagonist
26	>1000	23 ± 3.7	90 ± 11	905 ± 132	618 ± 248	-	-
27	>1000	3.7 ± 1.3	346 ± 81	>1000	>1000	-	-
28	>1000	1.4 ± 0.5	>1000	>1000	>1000	Antagonist	Antagonist
29	>1000	2.4 ± 0.6	83 ± 2.0	>1000	>1000	Antagonist	Antagonist
30	>1000	1.7 ± 0.2	>1000	>1000	>1000	Antagonist	Antagonist
31	>1000	0.7 ± 0.12	>1000	>1000	>1000	Antagonist	Antagonist
32	>1000	1.2 ± 0.4	>1000	455 ± 125	>1000	Antagonist	Antagonist

^aThe IC₅₀ values (nM) were derived from competitive radioligand displacement assays reflecting the affinities of the analogues for the cloned SRIF receptors using the non-selective [Leu⁸,DTrp²², ¹²⁵I-Tyr²⁵]SRIF-28, as the radioligand. Mean value ± SEM when n ≥ 3.

Inverting chirality at positions 2 and 3 in the octreotide scaffold (H-dPhe²-c[Cys³-Phe⁷-dTrp⁸-Lys⁹-Thr¹⁰-Cys¹⁴]-Thr¹⁵-ol, SRIF numbering) was reported to be the key structural modification converting an SRIF agonist into an antagonist.² Additional substitutions resulted in partially selective antagonists Acetyl-pNO₂Phe-c[dCys-Tyr-dTrp-Lys-Thr-Cys]-dTyr-NH₂² or H-Cpa-c[dCys-Tyr-dTrp-Lys-Thr-Cys]-2Nal-NH₂.⁴ These antagonists display preferentially high binding affinity for sst₂, and lower or no affinity to sst₃, sst₄ and sst₅. None of the analogues bind to sst₁. Using these lead compounds, we have designed SRIF antagonists that were more affine (> 3-fold) and more sst₂-selective than those reported so far. Guided by earlier observations whereby amide bond-rich moieties are favorably recognized by GPCR, most of the analogues reported have carbamoyl functionalities.²⁶ We like to believe that this empirically-based approach to drug design is to be distinguished from what is referred to as SAR studies. Indeed, a state-of-the-art and comprehensive SAR study will include conformational considerations”, whereas a “drug design” strategy may be based on intuitive, systematic and iterative substitutions, an understanding (although partial) of the fundamentals of peptide/protein interactions, deletions and changes in chirality among others, with the exclusion of structural requirements. On the other hand, it is only with the knowledge of the 3D NMR structures of the improved, structurally constrained and bioactive analogues resulting from such empirical strategies (applying a drug design approach) that a consensus pharmacophore can be determined and used for true SAR. This is exemplified in a manuscript to be submitted elsewhere (Grace et al., 2008).

This being said, analogues of antagonists like Acetyl-pNO₂Phe²-c[dCys³-Tyr⁷-dTrp⁸-Lys⁹-Thr¹⁰-Cys¹⁴]-dTyr¹⁵-NH₂² (1) and H-Cpa²-c[dCys³-Tyr⁷-dTrp⁸-Lys⁹-Thr¹⁰-Cys¹⁴]-2Nal¹⁵-NH₂⁴ (7) were synthesized to investigate the effect of different substitutions on binding affinity, receptor-subtype selectivity, overall hydrophilicity as well as agonism and antagonism.

The substitution of the N-terminal acetyl group by DOTA in 1 (IC₅₀ = 3.6 nM at sst₂) resulted in 2, which bound to sst₂ with an IC₅₀ = 1.5 nM, suggesting that the DOTA moiety, which is crucial for radiolabeling with ¹¹¹In, ⁹⁰Y, or ¹⁷⁷Lu for in vivo targeting, is well tolerated by sst₂ (Table 2). This conclusion is confirmed further with several additional examples. The introduction of dAph(Cbm)⁸ in place of dTrp⁸ in 2 yielded 3 (IC₅₀ = 0.75 nM). It is noteworthy that these two substitutions are cumulative, thus resulting in the most potent sst₂ antagonist in this series, with no measurable binding affinity to any of the other receptors. Further replacement of dTyr¹⁵ in 3 by 2Nal¹⁵ yielded 5 with a similar binding affinity for sst₂ (IC₅₀ = 1.3 nM). Analogue 4, a peptide with the same sequence as 5 but without DOTA at its N-terminus, still had excellent binding affinity for sst₂ (IC₅₀ = 2.6 nM) and also bound measurably to sst₃ (IC₅₀ = 384 nM). Substitution of Tyr in position 7 by Aph(Hor) resulting in 6 had no effect on sst₂ binding affinity and selectivity when compared with the parent 4 (IC₅₀ = 2.6 and 2.7 nM at sst₂ and 384 nM and 451 nM at sst₃, respectively, and no binding affinity at the other three receptors) (Table 2).

We also used H-Cpa²-c[dCys³-Tyr⁷-dTrp⁸-Lys⁹-Thr¹⁰-Cys¹⁴]-2Nal¹⁵-NH₂ (7) published by Hocart et al.⁴ as a second lead for sst₂-selective antagonists. This antagonist has IC₅₀ values in our binding assay equal 5.7, 112 and 218 nM at sst_2/3/5, respectively, as compared to the reported values of 26, 93 and 48 nM. In our assays, 7 is more potent than reported at sst₂ by a factor of five and less potent at sst₅ by the same factor. This points to the danger of comparing results from one laboratory to another when engaged in SAR studies.

Whereas N^α-methylation of Lys⁹ in 7⁴ to yield 8²⁷ increased binding affinity by a modest 5-fold in the assay used by Hocart et al.⁴ (K_i = 26 nM and 5.51 nM, respectively) with no improvement at sst₃ or sst₅, (K_i = ca. 50-100 nM), our in vitro binding assay could not confirm this improvement at sst₂ and, as a result, we did not pursue the use of this substitution in the design of additional sst₂-selective analogues. Instead, we synthesized 9 with an L-Agl(NMe,benzoyl)⁷ in an attempt to constrain the orientation of the side chain at position 7. The use of such aminoglycine derivatives (betides)^28,29 had been taken advantage of in the design of an sst₃-selective antagonist.^30,31 While 9 lost some binding affinity for sst₂ (3-fold) as compared to that of 7, it also lost comparable binding affinity for sst₃ and sst₅. This observation further suggests that position 7 is critical for all three sst_2,3,5. In fact, 10 with the d-Agl(NMe,benzoyl)⁷ lost binding affinity at sst₂ while retaining similar binding affinities as 7 at sst_3/4/5, thus accomplishing one of our goals of identifying the or those residues/conformations responsible for binding to any particular receptor (i.e., sst₂ in this case).

Whereas substituting Phe by Leu at position 7 in 7 yielded 11 that lost 10-fold binding affinity for sst₂ and selectivity, substitution by Aph(Cbm) yielded 12 which exhibited similar binding affinity and selectivity as 7 at the five sst_s. N-terminal carbamoylation of 12 to yield 13 improved binding affinity slightly at sst_3/4 with some loss of binding affinity for sst₂ as compared to 12. Addition of DOTA to 12 resulted in 14 whose sst₂ binding affinity is similar to that of 12 and increased selectivity for sst₂.

Interestingly, addition of a spacer in 14 between DOTA and the octapeptide such as βAla in 15 and Peg in 16 was unexpectedly detrimental^32,33 in terms of sst₂ binding affinity, yet favorable for sst₃ and neutral at sst_1/4/5.

From our observation that 2Nal¹⁵ may contribute to the sst₃, sst₄, and sst₅ binding pocket, 17 (missing this residue) was synthesized and found to have similar binding affinities when compared to the parent 12. Substitution of 2Nal¹⁵ in 12 by different other residues such as Cha in 18, Aph(Hor) in 19, dAph(Cbm) in 20 and Aph(Cbm) in 21 did not markedly influence affinity at sst₂ or selectivity. This is noteworthy in that there is only a three-fold difference in binding affinity at sst₂ for 20 (D-configuration and IC₅₀ = 5.4 nM) and 21 (L-configuration and IC₅₀ = 15 nM) where the C-terminal amino acid is of the d or l configuration, respectively. This supports the earlier observation that dTyr (as in 1 and 2) or 2Nal (as in 4 and 5) are both equally accepted. On the other hand, extension of the sequence of 20 by Glycine-OH as in 22, leads to significant loss of affinity at all receptors.

In order to modulate the overall hydrophilicity of 7 (with Tyr at position 7), we introduced the following carbamates (Aph(Cbm)⁷) in 12, (Aph(CONH-OCH₃)⁷) in 23 and (Aph(CONH-OH)⁷) in 24 at position 7. Whereas binding affinities for these analogues are not different from that of the parent 7, the order of elution of these analogues on HPLC at neutral pH suggests that 24 (RT = 31.6 min) may be more hydrophilic than 7 (RT = 34.8 min), 12 (RT = 31.9 min) and 23 (RT = 34.2 min). Since hydrophilicity may be a critical criterion for a clinically relevant radioligand, subtle differences in structure may favor one of these analogues when selecting a clinical candidate. The fact that 12, 23 and 24 are not superior to 7 in terms of sst₂ binding affinity and selectivity supports our previous finding that residue 7 is not an essential contributor to the sst₂ pharmacophore.³⁴

We then investigated the effect of substitutions at position 8. There is literature precedent suggesting that 5F-Trp is a favorable substitution for Trp⁸.³⁵ When introduced in 12, to yield 25 and 26, we observed a slight improvement in binding affinity for the three sst_2/3/5 as expected for the 5F-dTrp-containing 25 and less so for the corresponding l-isomer-containing 26. No increase in selectivity however, was seen for either analogue.

It was therefore very rewarding to find out that substitution of dTrp⁸ in 7 by dAph(Cbm)⁸ yielding 27 was clearly superior in terms of sst₂ selectivity with improved binding affinity. Further derivatization with the addition of DOTA at the N-terminus yielded 28 with an additional increase in binding affinity to sst₂ and greater than 500-fold selectivity at all other receptors.

Substitution of Tyr⁷ in 27 and 28 with Aph(Hor) yielded 29 and 30. Whereas 29 retained high binding affinity at sst₂ it also exhibited moderate binding affinity for sst₃; the binding affinity at sst₃ was lost upon the introduction of DOTA (30). Substitution of Tyr⁷ in 2 with ITyr yielded 32, the binding affinity of which was similar to that of 2 at sst₂.

We then substituted 2Nal¹⁵ in 30 by dTyr¹⁵ to yield 31. Of all analogues presented here, 31 (because of its hydrophilicity, RT = 13.2 min) may be the preferred candidate for bio-distribution and ultimately clinical investigation over 3 (RT = 13.6 min), 5 (RT = 26.1 min), 28 (RT = 26.7 min), 29 (RT = 27.7 min) or 32 (RT = 25.0 min) that are equally potent and selective in the binding assay. It is remarkable that the dipeptide sequence -Aph(Hor)-dAph(Cbm)- found in 29-31 is identical to that found in degarelix (Fe-200486),²⁶ a gonadotropin releasing hormone antagonist where it played a critical role in stabilizing a turn and in extending duration of action.

Put in perspective, the most affine DOTA-containing antagonists presented here (3 and 31) have binding affinities three to four-fold higher than that of SRIF-28 with no detectable binding affinity at any of the other four sst_s and are therefore potential candidates for clinical use.

All of the analogues tested here are antagonists in the calcium release assay in HEK293 cells stably expressing the human sst₂. Testing them alone, they do not affect calcium release up to 10 μM. However, the agonistic effect of the sst₂ agonist [Tyr³]-octreotide can be competitively antagonized with a 100-fold excess of each of the analogues applied individually. Figure 1 illustrates the antagonistic properties of some of the sst₂ antagonists using the calcium release assay.

Figure 1.

The SRIF analogues 3, 31 and Coy-14 behave like antagonists when tested in the calcium release assay. The calcium release assay was performed as described in Materials and Methods. HEK-sst₂ cells were treated either with 1 nmol/L, 10 nmol/L, 100 nmol/L, 1 μmol/L, and 10 μmol/L [Tyr³]-octreotide (●) alone, or with 10 nmol/L, 100 nmol/L, 1 μmol/L, and 10 μmol/L [Tyr³]-octreotide in the presence of an increasing concentration (10 nmol/L, 100 nmol/L, 1 μmol/L, and 10 μmol/L) of 3 (■) or 31 (▲), or the specific sst₂-antagonist Coy-14 (◆). 3, 31 and Coy-14 are antagonists since they shift the dose-response curve of [Tyr³]-octreotide to a higher molar range. Tested alone at 1 μmol/L and 10 μmol/L 3 (×), 31 (○) and Coy-14 (◇) have no effect on calcium release in HEK-sst₂ cells.

The antagonistic property of the analogues 3, 31 and 32 was also confirmed in an immunofluorescence-based internalization assay²⁰ with HEK293 cells stably expressing the human sst₂. Figure 2 illustrates that although the control agonist [Tyr³]-octreotide can induce sst₂ internalization, the tested sst₂-selective antagonists have no effect when given alone, even at a concentration of 10 μM. Moreover, they prevent sst₂ internalization induced by [Tyr³]-octreotide. Figure 3 shows the antagonistic properties of another analogue (32) in the ELISA internalization assay.

Figure 2.

Sst₂ internalization induced by [Tyr³]-octreotide is efficiently antagonized by the three SRIF analogues 32, 3, and 31. The immunofluorescence-based internalization assay was performed as described in Materials and Methods. HEK-sst₂ cells were treated for 30 min either with vehicle (no peptide, a), or with 100 nmol/L [Tyr³]-octreotide (b), a concentration inducing a sub-maximal internalization effect. Panels (d, f, h) show HEK-sst₂ cells treated with 100 nmol/L [Tyr³]-octreotide in the presence of 10 μmol/L of 32, 3, and 31. The effect of 32, 3, and 31 alone at a concentration of 10 μmol/L is shown in panels (c, e, g). As controls, panel (i) shows cells treated with 100 nmol/L [Tyr³]-octreotide in the presence of 10 μmol/L of the specific sst₂-antagonist Coy-14, and panel (j) shows cells treated with 100 nmol/L [Tyr³]-octreotide in the presence of 10 μmol/L of the sst₃-antagonist sst₃-ODN-8. A clear punctate perinuclear staining is detectable for [Tyr³]-octreotide. This punctate staining is efficiently abolished by an excess of the analogues 32, 3, 31, as well as the established antagonist Coy-14. However, 32, 3, and 31 given alone have no effect on sst₂ internalization. Sst₃-ODN-8, as negative control, is not able to antagonize the [Tyr³]-octreotide effect.

Figure 3.

The SRIF analogues 32 and Coy-14 behave like antagonists when tested in the ELISA-based internalization assay. HEK-sst₂ cells were pre-incubated with the mouse monoclonal HA epitope antibody (1:1,000) at room temperature for 2 h and then processed for ELISA as described in Material and Methods. HEK-sst₂ cells were treated either with 0.1 nmol/L, 1 nmol/L, 10 nmol/L, 100 nmol/L, 1 μmol/L, and 10 μmol/L [Tyr³]-octreotide (●) alone, or with 1 nmol/L, 10 nmol/L, 100 nmol/L, 1 μmol/L, and 10 μmol/L [Tyr³]-octreotide in the presence of an increasing concentration (1 nmol/L, 10 nmol/L, 100 nmol/L, 1 μmol/L, and 10 μmol/L) of 32 (■) or the specific sst₂-antagonist Coy-14 (▲). 32 and Coy-14 are antagonists since they shift the dose dependent internalization curve of [Tyr³]-octreotide to a higher molar range. Tested alone at 100 nmol/L, 1 μmol/L and 10 μmol/L 32 (□) and Coy-14 (△) do not stimulate sst₂ internalization in HEK-sst₂ cells.

To conclude, a great majority of the analogues reported here have a high affinity binding in the nanomolar range for sst₂ and often a high selectivity for sst₂ as well. The best compounds were 3 and 31 (with IC₅₀ values below 1 nM) followed by 32, 5, 28, 2 and 29. All of these antagonists are of particular interest, since they all include a DOTA moiety, making them candidates for in vivo tumor targeting. The labeling properties of the analogues and the specific in vivo uptake in tumors will be discussed in a subsequent publication (Maecke et al., in preparation).

Experimental Section

Starting Materials

MBHA resin with a capacity of 0.3-0.4 mequiv/g was used in the solid phase syntheses. All Boc-N^α-protected amino acids with side chain protection: Cys(Mob), Lys(ε-2Cl-Z), Lys(Fmoc), Thr(Bzl), Tyr(2Br-Z) and ITyr(3Br-Bzl) are commercially available (Bachem Inc., Torrance, CA; Chem Impex, Wood Dale, IL; Reanal, Budapest, Hungary) except Boc-Aph(Cbm)-OH, Boc-dAph(Cbm)-OH, Boc-Aph(Cbm-OCH₃)-OH, Boc-Aph(Cbm-OH)-OH, Boc-Aph(Hor)-OH,²⁶ Fmoc-d/l-Agl(NMe,Boc)-OH,²⁸ Fmoc-d-Agl(Boc)-OH,³⁶ Boc-5F-Trp-OH, Boc-5F-dTrp-OH which were synthesized in our laboratory. 1,4,7,10-Tetraazacyclododecane-1,4,7,10-tetraacetic acid mono(N-hydroxysuccinimide).ester.3CH₃COOH.HPF₆ (DOTA-NHS) was purchased from Macrocyclics Inc. (Dallas, TX). All reagents and solvents were ACS grade and were used without further purification.

Peptide Synthesis

Peptides were synthesized by the solid-phase approach either manually or on a CS-Bio Peptide Synthesizer Model CS536.³⁷ A 3-equiv excess of Boc-amino acid (1.2 mmol) based on the original substitution of the resin was used for each coupling. Peptide couplings were mediated for 1 h by DIC/HOBt (1.2 mmol/1.8 mmol) in dimethylformamide (DMF) and monitored by the qualitative ninhydrin test.³⁸ Boc removal was achieved with trifluoroacetic acid (TFA) (60% in CH₂Cl₂, 1-2% ethanedithiol or m-cresol) for 20 min. An isopropyl alcohol (1% m-cresol) wash followed TFA treatment and then successive washes with triethylamine (TEA) solution (10% in CH₂Cl₂), methanol, triethylamine solution, methanol and CH₂Cl₂ completed the neutralization sequence. The ureido group (Cbm) at the N-terminus of 13 was introduced on the resin. The N-terminal Boc group of the fully assembled peptide was deprotected with TFA in the usual manner,²⁶ after neutralization, the carbamoylation proceeded with NaOCN (100 mg, 0.65 mmol) in N-methylpirrolidinone (NMP) (4 mL) and glacial acetic acid 3 mL per gram of initial resin. The mixture was agitated at room temperature for 30 min and ninhydrin test indicated a complete reaction. The completed peptide was then unprotected and cleaved from the resin by HF containing the scavengers anisole (10% v/v) and methyl sulfide (5% v/v) for 60 min at 0 °C. The diethyl ether precipitated crude peptides were cyclized in 75% acetic acid (200 mL) by addition of iodine (10% solution in methanol) until the appearance of a stable orange color. Forty minutes later, ascorbic acid was added to quench the excess of iodine.

For the synthesis of 9, we used unresolved Fmoc-d/l-Agl(NMe,Boc)-OH and the two diastereomers were separated readily during the standard HPLC purification steps.^24,39 The optical configuration of the two diastereomers was tentatively inferred from a comparison of the HPLC elution behavior with analogue synthesized separately as diastereomers of known optical configuration. In short: after coupling Fmoc-dAgl(Boc)-OH in position 7, the side chain protecting Boc group was removed with 60% TFA, washed, neutralized and to the 0.9 g peptide resin (0.36 mmol/g) swollen in dichloromethane, Dod-Cl (130 mg; 0.5 mmol) was added along with DIEPA (500 μL). The mixture was shaken for an hour to complete the alkylation. The resin was washed, and shaken after the addition of formaldehyde (2 mL, 37% solution) in NMP (18 mL) and acetic acid (100 μL). After 5 min, sodium cyanoborohydride (300 mg) was added and the mixture was shaken for 60 min. After the removal of the Dod group with TFA (60%) for 30 min, benzoyl chloride (500 μL) was used to acylate the free secondary amino group of the side chain.⁴⁰ Removal of the N^α-Fmoc protecting group with 20% piperidine in NMP in two successive 5 and 15 min treatments was followed by the standard elongation protocol until completion of the peptide. The peptide was cleaved, deprotected and cyclized as described above. On HPLC, this d configuration diastereomer coeluted with the ealier eluting diastereomer from the synthesis performed with the unresolved amino acid, therefore the slower eluting peptide (9) was tentatively identified as the L-Agl(NMe, benzoyl)⁷ containing analogue.

Generally, for the synthesis of the DOTA-peptide-conjugates, the side chain of Lys⁹ was protected with an Fmoc protecting group that stays on after HF cleavage. To a solution of the RP-HPLC purified [Lys(Fmoc)⁹]-sst₂-antagonist (~20 μM) in dry DMF (800 μL) was added a solution of DOTA-NHS-ester (38 mg, 48 μM) in DMF (160 μL) and N,N’-Diisopropylethylamine (DIPEA) (40 μL, 24 μM). The mixture was stirred at room temperature for 5 hours. The progress of the reaction was followed by analytical HPLC. After completion of the reaction, a preparative RP-HPLC purification was performed yielding the pure DOTA-[Lys(Fmoc)]⁹-sst₂-antagonist. Removal of the Fmoc protecting group from the Lys side chain was achieved with 20% piperidine/DMF solution resulting in the DOTA-sst₂-antagonist, which was further purified by preparative RP-HPLC.

Purification of Peptides

The crude, lyophilized peptides were purified by preparative RP-HPLC²⁴ on a 5 cm × 30 cm cartridge, packed in the laboratory with reversed-phase 300Å Vydac C₁₈ silica (15-20 μm particle size). The peptides eluted with a flow rate of 100 mL/min using a linear gradient of 1% B per 3 min increase from the baseline % B (eluent A = 0.25 N TEAP pH 2.25, eluent B = 60% CH₃CN, 40% A). All peptides were subjected to a second purification step carried out with eluents A = 0.1% TFA in water and B = 60% CH₃CN/40% A on the same cartridge using a linear gradient of 1% B per min increase from the baseline % B. Analytical HPLC screening of the purification was performed on a Vydac C₁₈ column (0.46 × 25 cm, 5 μm particle size, 300 Å pore size) connected to a Rheodyne injector, two Waters pumps Model 501, System Controller Programmer, Kratos 750 UV detector, and Houston Instruments D-5000 strip chart recorder. The fractions containing the product were pooled and subjected to lyophilization.

Characterization of SRIF Analogues (Table 1)

The purity of the final peptides was determined by analytical RP-HPLC performed with a linear gradient using 0.1 M TEAP pH 2.5 as eluent A and 60% CH₃CN/40% A as eluent B on a Hewlett-Packard Series II 1090 Liquid Chromatograph connected to a Vydac C₁₈ column (0.21 × 15 cm, 5 μm particle size, 300 Å pore size), Controller Model 362 and a Think Jet printer. Capillary zone electrophoresis (CZE) analysis was performed as described earlier.²⁵ Each peptide was found to have a purity of >95% by HPLC and CZE. Mass spectra (MALDI-MS) were measured on an ABI-Perseptive DE-STR instrument. The instrument employs a nitrogen laser (337 nm) at a repetition rate of 20 Hz. The applied accelerating voltage was 20 kV. Spectra were recorded in delayed extraction mode (300 ns delay). All spectra were recorded in the positive reflector mode. Spectra were sums of 100 laser shots. Matrix α-cyano-4-hydroxycinnamic acid was prepared as saturated solutions in 0.3% trifluoroacetic acid and 50% acetonitrile. The observed monoisotopic (M + H)⁺ values of each peptide corresponded with the calculated (M + H)⁺ values.

Reagents

All reagents were of the best grade available and were purchased from common suppliers. [Tyr³]-octreotide⁴¹ was from Novartis Inc. (Basel, Switzerland). All other peptides, including Coy-14²⁷ were synthesized at the Salk Institute. The sst_2A-specific antibody R2-88 antibody to the sst_2A receptor was generated as previously described and has been extensively characterized.⁴² The secondary antibody Alexa Fluor 488 goat anti-rabbit IgG (H+L) was from Molecular Probes, Inc. (Eugene, OR), the monoclonal anti-T7 antibody from Novagen (Madison, WI), the goat anti-mouse IgG horseradish peroxidase conjugate from Bio-Rad Laboratories, Inc. (Hercules, OR); the Fluo-4NW Calcium Assay kit was from Molecular Probes, Inc. (Eugene, OR), substrate mix for horseradish peroxidase (ABTS) was from Bio-Rad Laboratories, Inc. (Hercules, OR), lactalbumin hydrolysate was from HyClone (Logan, UT).

Cell lines

CHO-K1, CCL39 cells stably expressing the cloned five human sst_s and the HEK293 cell line expressing the T7-epitope tagged human sst_2A (HEK-sst₂) were grown as described previously.^8,20 All culture reagents were from Gibco BRL, Life Technologies, (Grand Island, NY).

Receptor autoradiography

Cell membrane pellets were prepared as previously described⁸ and stored at -80 °C. Receptor autoradiography was performed on 20-μm thick cryostat (Microm HM 500, Walldorf, Germany) sections of the membrane pellets, mounted on microscope slides, and then stored at -20 °C. For each of the tested compounds, complete displacement experiments with the universal SRIF radioligand [Leu⁸, D-Trp²², ¹²⁵I-Tyr²⁵]-SRIF-28 (¹²⁵I-[LTT]-SRIF-28) (2,000 Ci/mmol; Anawa, Wangen, Switzerland) using 15,000 cpm/100 μl and increasing concentrations of the unlabelled peptide ranging from 0.1 - 1000 nM were performed. As control, unlabelled SRIF-28 was run in parallel using the same increasing concentrations. The sections were incubated with ¹²⁵I-[LTT]-SRIF-28 for 2 hours at room temperature in 170 mmol/L Tris-HCl buffer (pH 8.2), containing 1% BSA, 40 mg/L bacitracin, and 10 mmol/L MgCl₂ to inhibit endogenous proteases. The incubated sections were washed twice for 5 min in cold 170 mmol/L Tris-HCl (pH 8.2) containing 0.25% BSA. After a brief dip in distilled water to remove excess salts, the sections were dried quickly and exposed for 1 week to Kodak BioMax MR film. IC₅₀ values were calculated after quantification of the data using a computer-assisted image processing system as described previously.⁴³ Tissue standards (Autoradiographic [¹²⁵I] microscales, GE Healthcare; Little Chalfont, UK) that contain known amounts of isotope, cross-calibrated to tissue-equivalent ligand concentrations were used for quantification.⁴⁴

Immunofluorescence-based sst₂ Internalization Assay

Immunofluorescence microscopy-based internalization assay for sst₂ was performed with HEK-sst₂ using the sst₂-specific antibody R2-88 as described earlier.²⁰ HEK-sst₂ cells were treated either with vehicle alone, the sst₂ agonist [Tyr³]-octreotide at a concentration of 100 nM, [Tyr³]-octreotide at a concentration of 100 nM in the presence of an excess of the SRIF analogues to be tested (100 times the concentration of [Tyr³]-octreotide), or with the SRIF analogues to be tested alone at a concentration of 10 μM, and then processed for immunofluorescence microscopy as described previously.²⁰

Quantitative assay for sst₂ internalization (ELISA)

Receptor internalization was determined using an ELISA to quantitate T7-epitope-tagged human sst₂ on the cell surface. HEK-sst₂ cells were seeded on poly-D-lysine (20 μg/mL) coated 24-well plates (250′000 cells per well) in growth medium and cultured for 1 day at 37 °C and 5% CO₂. On the day of the assay, cells were incubated with the monoclonal anti-T7 antibody at a dilution of 1:3000 for 2 h at room temperature in DMEM containing 5 g/L lactalbumin hydrolysate + 20 mM HEPES, pH 7.4 (DMEM-LH) to label cell surface receptors. After washing with DMEM-LH to remove unbound antibody, cells were incubated for 30 min at 37 °C and 5% CO₂ either without or with the SRIF analogues to be tested, added at the concentrations indicated. Incubations were terminated by placing the plates in an ice bath. Cells were then washed twice with cold PBS and fixed for 10 min at room temperature with 3% paraformaldehyde in PBS (pH 7.4). Nonspecific binding sites were blocked by incubating the cells for 60 min at room temperature with PBS containing 1% bovine serum albumin (BSA; Fraction V; SERVA, Heidelberg, Germany). Cells were then incubated for 60 min at room temperature with goat anti-mouse IgG horseradish peroxidase conjugate (1:1000) in PBS containing 1% BSA. After 3 additional washes with PBS, antibody binding was measured by adding 0.3 mL substrate mix for horseradish peroxidase (ABTS). The OD₄₀₅ was measured after an approximately 30 min incubation at room temperature. The amount of sst₂ remaining at the cell surface after ligand treatment was calculated as the absorbance measured in treated cells expressed as a percentage of the absorbance in untreated cells. Nonspecific absorbance was determined in experiments in which HEK293-sst₂ cells were incubated without the anti-T7 antibody. Each data point represents the mean ± SEM of three experiments performed in duplicate.

Calcium release assay

Intracellular calcium release was measured in HEK-sst₂ using the Fluo-4NW Calcium Assay kit as described previously.^45,46 In brief, HEK-sst₂ cells were seeded (25,000 cells per well) in poly-D-lysine (20 μg/mL) coated 96 well plates and cultured for 1 day at 37 °C and 5% CO₂ in culture medium. On the day of the experiment, the cells were washed with assay buffer (1 × HBSS, 20 mM HEPES) containing 2.5 mM probenecid, and then incubated with 100 μL/well Fluo-4NW dye in assay buffer containing 2.5 mM probenecid for 30 min at 37 °C and 5% CO₂ and an additional 30 min at room temperature. To measure the intracellular calcium mobilization after stimulation with the SRIF analogues to be tested, the dye-loaded cells were transferred to a SpectraMax M2^e (Molecular Devices, Sunnyvale, CA). Intracellular calcium mobilization was recorded in a kinetic for 60 sec at room temperature monitoring fluorescence emission at 520 nm (with λ_ex = 485 nm) in the presence of the analogues at the concentrations indicated. Maximum fluorescence (F-max) was measured after the addition of 25 μM ionomycin. Baseline (F-baseline) measurements were taken for dye-loaded, untreated cells. Data are shown as percentage of maximum calcium response (F-max - F-baseline = 100 % of maximum calcium response) as reported previously.^45,46 All experiments were repeated at least three times in triplicate.

Abbreviations

The abbreviations for the common amino acids are in accordance with the recommendations of the IUPAC-IUB Joint Commission on Biochemical Nomenclature (Eur. J. Biochem. 1984, 138:9-37). The symbols represent the L-isomer except when indicated otherwise. Additional abbreviations:

Agl

aminoglycine

Aph

4-aminophenylalanine

3-Br-Bzl

3-bromobenzyl

Boc

t-Butoxycarbonyl

BSA

bovine serum albumin

Bzl

benzyl

Cbm

carbamoyl

Cha

cyclohexylalanine

Cpa

4-Cl-phenylalanine

CZE

capillary zone electrophoresis

DIC

N,N’-diisopropylcarbodiimide

DIPEA

N,N’-Diisopropylethylamine

DMEM-LH

Dulbecco’s modified Eagle’s medium-lactalbumine hydrolysate/HEPES

DMF

dimethylformamide

Dod

4,4′-dimethoxydityl

DOTA

1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid

DTPA

diethylenetriaminepentaacetic acid

ELISA

enzyme-linked immunosorbent assay

Fmoc

9-Fluorenylmethoxycarbonyl

HBSS

Hank’s balanced salt solution

HEPES

4-(2-hydroxyethyl)-1-piperazineethansulfonic acid

HOBt

1-hydroxybenzotriazole

MBHA

methyl-benzhydryl resin

Mob

4-methoxybenzyl

2Nal

3-(2-naphthyl)-alanine

NMP

N-methylpirrolidinone

NMR

nuclear magnetic resonance

3D

three-dimensional

Peg

12-amino-4,7,10-trioxadodecanoic acid

PBS

Phosphate buffered saline

pNO₂-Phe

para-nitrophenylalanine

SAR

structure activity relationships

SRIF

somatostatin-14

sst_s

SRIF receptors

SRIF-28

somatostatin-28

TATE

[Tyr³,Thr⁸]-octreotide = [Tyr³]-octreotate

TEA

triethylamine

TEAP

triethylammonium phosphate

TFA

trifluoroacetic acid

TOC

[Tyr³]-octreotide

Z(2Br)

2-bromobenzyloxycarbonyl

Z(2Cl)

2-chlorobenzyloxycarbonyl