Cloning and Expression of the Human Delta Opioid Receptor

Richard J. Knapp

Aventis Pharmaceuticals, Bridgewater, New Jersey, U.S.A. Ewa Malatynska

Johnson & Johnson, Spring House, Pennsylvania, U.S.A.

Eva V. Varga, William R. Roeske, and Henry I. Yamamura

University of Arizona, Tucson, Arizona, U.S.A.

Considerable evidence had been accumulated for the existence of multiple opioid receptor subtypes prior to the cloning of the cDNAs encoding the three major opioid receptor types—the mu, delta, and kappa opioid receptors [1,2]. Since our laboratory was working on the development of selective delta receptor agonists, including [D-Pen2, D-Pen4] enkephalin (DPDPE) and its halogenated analogues, we were primarily interested in the characterization of the delta opioid receptor and the identification of putative delta opioid receptor subtypes.

At least three hypotheses were proposed to explain pharmacological data suggesting delta receptor heterogeneity. The more conservative hypothesis was the existence of at least two distinct delta receptor proteins, encoded by two distinct mRNA species. A more imaginative hypothesis, proposed by

Rothman and Westfall [3,4], divided the delta receptor population into two components: one coupled to mu opioid receptors, and the other acting alone. The third hypothesis combined these ideas, by suggesting that the delta receptors that interact with the mu opioid receptors are different molecular entities from those that act alone [5,6]. A definitive test of the above hypotheses would be to identify two or more delta receptor proteins by molecular cloning. The cloning of the human delta opioid receptor [7] was the fortunate outcome of a larger project in our laboratory to clone of the cDNAs encoding putative delta opioid receptor subtypes.

Although the opioid receptors were perhaps the first neurotransmitter receptors to be characterized as molecular entities, the molecular cloning of the first opioid receptor cDNAs was elusive. The idea of a ''receptive substance'' was developed in the late 19th century by J.N. Langley, to explain the selectivity of drugs in certain tissues. While the concept of drug receptors was supported by pharmacological analyses in isolated tissue preparations, the molecular identity of the receptors was not known for a long time. The existence of opioid receptor molecules was confirmed only in the 1970s, by the work of Snyder [8], Simon [9], and Terenius [10], who independently used the technique of receptor radioligand binding, first suggested by Goldstein [11], to demonstrate the existence of a distinct, stereoselective opiate binding molecule in brain. These studies provided the first direct evidence that drugs could recognize a specific cellular component, with high affinity and a distinct tissue distribution. Further studies using this technique showed that receptor recognition was a property of many, though not all, drugs, and provided an essential tool to characterize their biochemical properties. Opioid receptors proved unsuitable for receptor isolation and biochemical characterization, largely because of their low density in neural tissue.

Our initial approach for the molecular cloning of delta receptor subtypes was to use the expression cloning technique, where cDNAs isolated from a suitable tissue are expressed in mammalian cells, and the presence of the desired receptor is detected by a selective radioligand. By that time, we had developed and characterized a novel radioiodinated ligand, [125I-Phe3]-DPDPE [12]. The high selectivity and specific activity, and low nonspecific binding of this radioligand made it ideal for expression cloning. However, before we could complete our work, two independent groups (Evans et al. [13] and Kieffer et al. [14]) applied this technique using other radioligands, to clone the mouse delta opioid receptor.

Both groups used expression cloning to screen cDNA libraries obtained from NG 108-15 mouse neuroblastoma-rat glioma hybridoma cells. These cells were a logical choice for delta opioid receptor cloning efforts, since they express high density of delta receptors and can be produced in great quantities in cell culture.

The delta opioid receptor protein described by Evans et al. [13] and Kieffer et al. [14] consists of 372 amino acids. Hydropathy analysis indicates seven putative a-helical transmembrane, domains in the cloned protein. The cloning of the mouse delta opioid receptor opened a new path to the molecular cloning of other opioid receptors by hybridization screening. Accordingly, the mouse mu [15] and kappa [16] opioid receptor types were cloned in rapid succession. Interestingly, one of the first reported kappa opioid receptor clones did not originate from homology screening using a cloned delta receptor probe, but instead resulted from efforts to clone somatostatin receptor types [17]. One of the cDNA clones identified in this work had low homology to other somatostatin receptors, but exhibited a relatively high homology to the just described mouse delta opioid receptor. Subsequently, this clone was tested for opioid ligand binding and was identified as a kappa opioid receptor.

The cloned mouse delta receptor sequence also facilitated the cloning of opioid receptor cDNAs from other animal species. Our cloning efforts were initially directed at finding a delta opioid receptor subtype, different from the cloned mouse delta opioid receptor type. We began by generating a cDNA probe to use for homology sceening. We usedmRNA from NG 108-15 cells as template, and the primer pair: 5'-GGGTCTTGGCTTCAGGTGTCG-3' (sense) and 5' -GCAGCGCTTGAAGTTCTCGTC-3' (antisense) in a reverse transcriptase polymerase chain reaction (RT-PCR) to amplify a 467-bp fragment of the mouse delta opioid receptor cDNA. The PCR product was ligated into a pCR II cloning vector, amplified in E. coli cells, and labeled with [a-32P]dCTP and random primers, to produce the desired radioactive hybridization probe. Since much of the work on delta receptor heterogeneity was done in mouse tissues, we subsequently used this probe to screen mouse brain cDNA libraries, to hopefully identify mouse delta opioid receptor subtypes. However, while we isolated several clones, none of the sequences were different from the previously published mouse delta opioid receptor. During this time we also started screening a human striatal cDNA library, and identified a single cDNA clone, designated 44-11. Since the human receptors are the ultimate pharmacological targets for drug development, and no human delta opioid receptor sequences had been described at that time, we decided to shift our attention to the cloning of the human delta opioid receptor.

A 0.7-kb human delta opioid receptor cDNA probe, corresponding to the 3' end of the expected delta receptor sequence, was produced from the 1.6kb cDNA 44-11 clone by digestion with EcoRI and Not I restriction enzymes. This probe was used to screen a human temporal cortex cDNA library. Three additional cDNA clones were isolated from this library. Each of these clones was ~1.0-kb long, and contained only partial fragments of the predicted receptor sequence. The breakthrough was the finding that one of the cloned fragments (designated 78-4) contained the elusive 5' initiation codon, and that most of its 3' sequence overlapped with that of the 44-11 clone (Fig. 1). However, an 84-bp stretch at the 3' end of the 78-4 clone did not overlap with the sequence of the 44-11 cDNA clone. The 84-bp sequence was further analyzed, and was identified as a receptor fragment that was complementary to nucleotide positions 382-465 in the open reading frame of the cloned delta receptor, in the inverted orientation. This indicated that the 78-bp fragment in the 78-4 clone was an artifact of the library construction process. With this information, we turned to the problem of reassembling the two fragments to restore the full open reading frame.

We were able to reconstitute the full open reading frame by the identification of a unique, HincII restriction site, within the overlapping region of the two clones (Fig. 2). Furthermore, the presence of an Apal restriction site in the untranslated 3' sequence of the 44-11 clone allowed ligation of the HincII-ApaI double digestion product of the 44-11 clone into the corresponding sites of the pBluescript 78-4 clone, to assemble the complete open reading frame in the correct orientation. Restriction analysis with HincII and ApaI demonstrated that both sites were preserved in the ligation product.

The reconstituted 78x44 clone was then transferred into pREP10 and pcDNA3 expression vectors for expression in mammalian (COS-7, CHO, and others) cells. For initial characterization, the recombinant hDOR/pcDNA vector was transiently transfected into COS-7 cells. The presence of the human delta opioid receptor in the transfected cells was confirmed by radio-ligand binding studies, using [3H] naltrindole, [3H] [4'-Cl-Phe4] DPDPE and [3H][D-Ala2, Glu4]deltorphin as radioligands. Each of these delta-selective radioligands exhibited high affinity specific binding to transfected COS-7 cell membrane preparations. In contrast, very low specific binding was detected with either of the radiololigands in sham-transfected (empty pcDNA3 vector)

Figure 1 Positions of cloned cDNA sequences relative to the mouse delta opioid receptor open reading frame. The figure shows an alignment of the insert sequences for the four cDNA clones obtained from human brain cDNA libraries. Together the 44-11 and 78-4 cDNA inserts cover the entire open reading frame and include 3' and 5' flanking sequences.

Figure 1 Positions of cloned cDNA sequences relative to the mouse delta opioid receptor open reading frame. The figure shows an alignment of the insert sequences for the four cDNA clones obtained from human brain cDNA libraries. Together the 44-11 and 78-4 cDNA inserts cover the entire open reading frame and include 3' and 5' flanking sequences.

Figure 2 Assembly of human delta opioid receptor open reading frame. The figure illustrates how the human delta opioid receptor open reading frame was assembled from the 44-11 and 78-4 cDNA fragments using the HincII and Apal restriction sites.

cell membranes. In addition, competition binding studies were also used to demonstrate that the expressed receptor exhibited the expected selectivity profile, with high affinity for delta-, but significantly lower affinities for mu-and kappa-selective opioid ligands (Table 1).

An initial controversy was encountered at this step, since the Kd and K; values of the ligands tended to be higher in cells expressing the cloned opioid receptors than the previously determined values in brain membrane preparations, leading to an ambiguity regarding the identity of the cloned opioid receptors. Thus, the Kd values we obtained for [3H]naltrindole binding to membranes from COS-7 cells transiently transfected with the human delta receptor clone ranged from 55 to 560 pM. Importantly, the obtained Kd values showed a strong correlation with the actual receptor concentrations in the particular batch of transfected cells used for the experiment (18-430 pM, depending on the transfection [31]). Ligand depletion, resulting in reduced

TABLE 1 Binding Affinities for Selective Opioid Receptor Ligands to Membrane Preparations from COS-7 Cells, Transiently Transfected with the Cloned Human Delta Opioid Receptor

Liganda

Preferred receptor

Kib (nM + SEM)

0 0

Post a comment