Asymmetric total synthesis of tetrahydroprotoberberine derivatives and evaluation of their binding affinities at dopamine receptors

Article abstract of DOI:10.1016/j.bmcl.2017.01.090

Cocaine addiction remains a serious challenge for clinical and medical research because there is no effective pharmacological treatment. L-THP, a natural product isolated from Corydalis yanhusuo W.T. Wang, is one of the most frequently used traditional herbs to treat drug addiction in China. Our laboratory first reported that its demethylated metabolites L-ICP, L-CD, and L-CP had high affinity at dopamine D1, D2, and D5 receptors. Here we report the chemical synthesis of these metabolites and other derivatives and their binding affinities at dopamine receptors. The synthesis of these bioactive metabolites will allow further in vivo study of their potential in treating cocaine addiction.

Full text of DOI:10.1016/j.bmcl.2017.01.090

Accepted Manuscript
Asymmetric Total Synthesis of Tetrahydroprotoberberine Derivatives and Eval-
uation of Their Binding Affinities at Dopamine Receptors
David Y.W. Lee, Jing Liu, Shuzhen Zhang, Peng Huang, Lee-Yuan Liu-Chen
PII:
S0960-894X(17)30114-2
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.01.090
BMCL 24666
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
9 December 2016
25 January 2017
27 January 2017
Please cite this article as: Lee, D.Y.W., Liu, J., Zhang, S., Huang, P., Liu-Chen, L-Y., Asymmetric Total Synthesis
of Tetrahydroprotoberberine Derivatives and Evaluation of Their Binding Affinities at Dopamine Receptors,
Bioorganic & Medicinal Chemistry Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.01.090
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Asymmetric Total Synthesis of Tetrahydroprotoberberine Derivatives and Evaluation of
Their Binding Affinities at Dopamine Receptors
David Y.W. Lee,a,* Jing Liu,a Shuzhen Zhang,a Peng Huang,b and Lee-Yuan Liu-Chen,b
a
Bio-Organic and Natural Products Laboratory, McLean Hospital, Harvard Medical School, 115
Mill Street, Belmont, MA 02478, USA
b
Department of Pharmacology and Center for Substance Abuse Research, School of Medicine,
Temple University, 3420 N. Broad Street, Philadelphia, PA 19140, USA
*Corresponding author. Tel.: +1 617 855 2038; fax: +1 617 855 2040
E-mail address: DLee@mclean.harvard.edu (D.Y.W. Lee)
Abstract: Cocaine addiction remains a serious challenge for clinical and medical research because
there is no effective pharmacological treatment. L-THP, a natural product isolated from Corydalis
yanhusuo W.T. Wang, is one of the most frequently used traditional herbs to treat drug addiction in
China. Our laboratory first reported that its demethylated metabolites L-ICP, L-CD, and L-CP had
high affinity at dopamine D1, D2, and D5 receptors. Here we report the chemical synthesis of
these metabolites and other derivatives and their binding affinities at dopamine receptors. The
synthesis of these bioactive metabolites will allow further in vivo study of their potential in
treating cocaine addiction.
Keywords: tetrahydroprotoberberine; cocaine; dopamine receptor
Cocaine is a highly addictive psychostimulant associated with intense craving during periods of
abstinence. In the US, an estimated 1.4 million individuals aged 12 years or older were current
(past month) users in 2013 (SAMHSA, 2014). Cocaine use disorder (CUD) is likely to continue,
and there is no FDA-approved pharmacologic treatment1-3, so it remains a major challenge for
clinical and medical research4.
L-THP is a natural product isolated from Corydalis yanhusuo W.T. Wang, one of the most
frequently used traditional herbs to treat drug addiction in China5. It has been used in clinical
practice in China for >40 years for its analgesic, sedating, and hypnotic effects (commercial name:
Rotundine™)6, 7. Structurally, it resembles two dopamine molecules fused into a tetracyclic
alkaloid (Figure 1). It has low toxicity and a high therapeutic index. Extensive studies, carried out
mostly by Chinese scientists, show that it has no affinity for opioid receptors, does not change
levels of prostaglandins, binds with moderate affinity to dopamine D1, D2, and D3 receptors8 and
has dual pharmacological properties of D1 partial agonism and D2 antagonism, with some affinity
for α-adrenergic and serotonin receptors9.
Figure 1. Structure of dopamine and tetrahydroprotoberberine

Recently we identified the monodemethylated metabolites L-corypalmine (L-CP), L-corydalmine
(L-CD), and L-isocorypalmine (L-ICP) in the serum of mice and rats 30 min and 60 min after
administration of L-THP 20 mg/kg. It is noteworthy that the partial structures (A, B ring and C, D
ring) of these metabolites resemble the structure of dopamine. We have found that the
C2-demethylated metabolite L-ICP binds to D1 and D5 receptors with high affinity and to D2, D3,
and D4 with moderate affinity, and that it is functionally a partial agonist at D1 and D5 and an
antagonist at D2, D3, and D4. As L-CP and L-CD are structurally analogous to L-ICP, they are
likely to have similar novel pharmacological profiles. We thus hypothesize that L-THP is a
pro-drug and exerts its behavioral effects via demethylated metabolites.
In a previous study, we synthesized L-ICP and L-SPD by treating L-THP with MsOH and
methionine (1.2 eq) at room temperature10. But L-CP and L-CD were the minor products in this
reaction. So total synthesis is the best method to get the tetrahydroprotoberberine derivatives
(Table 1).
Table 1. Structure of tetrahydroprotoberberine derivatives
R1
R2
CH3O
OH
CH3O
CH3O
OH
R3
CH3O
CH3O
CH3O
OH
L-THP
CH3O
CH3O
OH
CH3O
CH3O
L-ICP (14a)
L-CP (14b)
L-CD (14c)
L-SPD (14d)
L-THB (14e)
L-DTHB (14f)
OH
OCH2O
OCH2O
CH3O
OH
Total synthesis requires the two key intermediates 3 and 7, which were synthesized using the route
described in Scheme 1 below. Phenylethylamine 3 was readily synthesized from vanillin or
isovanillin; the overall yield was 50-52%. To prepare the intermediates 7, bromination of
2-(4-hydroxyphenyl) acetate with bromine was followed by protection with benzyl bromide or
methyl iodide in acetone to give the intermediate 4. Phenols 5 were synthesized from halogen
benzene 4 with (bis(8-quinolinolate)copper(II)) catalyst with a chemical yield of 90%.
Intermediate 5 was subjected to hydroxymethylation with phenylboric acid and paraformaldehyde
to afford the lactone 6. Finally, the lactone 6 was methylated with iodomethane to provide the
intermediates 7.
The method for the preparation of Intermediate 5 has been reported by Weller et al.11, but the
reaction conditions require concentrated sodium hydroxide as base and CuSO4 as a catalyst,
reacted at 150℃ for 36 hours in stainless steel cannula. The change was made to simplify the
reaction conditions by choosing bis(8-quinolinolate)copper(II) as the catalyst. The reaction was
done at 120°C for 6 hours without nitrogen protection in sealed stainless steel cannula, and the
yield was increased to 85–90%. At the same time, the bis(8-quinolinolate)copper (II) catalyst was
recovered by filtration and washing.
Intermediates 8 was prepared by the condensation of lactone 7 with phenethylamine 3 in ethanol.
The benzyl alcohol group was converted to the corresponding acetate 9 with acetic
anhydride/triethylamine. The Bischler-Napieralski reaction was done in the presence of
phosphorus oxychloride in acetonitrile, and the imine 10 was obtained in approximately
quantitative yield. Freshly prepared imine 10 was used directly for the next reduction without

further purification. Asymmetric transfer hydrogenation was done under argon in DMF, in the
presence of Noyori’s catalyst and formic acid/triethylamine (v/v = 5/2) as the hydrogen source.
Intermediates 13 was prepared by the chlorination of the benzyl alcohol group in 12 with thionyl
chloride in methylene chloride at 0℃. The ring closure reaction was accomplished by basifying the
methylene chloride solution with saturated sodium bicarbonate. Deprotection was achieved by
refluxing 14 in concentrated hydrochloric acid, with ee values up to 99.6%.
Scheme 1. Preparation of the amine and isochroman-3-one. Reagents and conditions: (a) CH3NO2,
CH3COOH, CH3COONH4, 100 ℃, 68–78%; (b) THF, LiAlH4, reflux, 70–75%; (c) 30% NaOH，
bis(8-quinolinolate)copper(II), 89–90%; (d) C6H5B(OH)2, toluene, 110 ℃, 1 h; (HCHO)n, 20 h,
then H2O, reflux, 2 h, 46–50%; (e) CH3I, acetone, K2CO3, reflux, 84–88%.
Scheme 2. Synthesis of (-)-tetrahydroprotoberberine derivatives. Reagents and conditions: (f)
C2H5OH, reflux, 46–57%; (g) CH3COCl, pyridine, CH2Cl2, 62–70%; (h) POCl3, CH3CN, reflux; (i)
(Ru[(R,R)-Tsdpen](g6-p-cymene), DMF, HCOOH/TEA, room temperature, 32–40%; (j) NaOH,
EtOH/H2O, 86–92%; (k) SOCl2, CH2Cl2; (l) saturated NaHCO3, 56–65%; (m) concentrated HCl,
C2H5OH, reflux, 68–85%.
Dopamine receptor binding was performed according to published procedures12-13. For D1 and D5
receptors, [3H]SCH23390 was used as the radiolabeled ligand and fluphenazine (10 μM) was used
to define nonspecific binding. For D2, D3, and D4 receptors, [3H]methylspiperone and

(+)-butaclamol (4 μM) were used, respectively. Membranes were prepared from transfected
HEK293 or CHO cells as described previously14. Saturation binding of [3H]SCH23390 or
[3H]methylspiperone to the D1 and D5 receptors or D2, D3, and D4 receptors, respectively, was
done with at least 8 concentrations of [3H]SCH23390 (ranging from 20 pM to 20 nM) or
[3H]methylspiperone (ranging from 20 pM to 10 nM). Binding was done in 50 mM Tris-HCl
buffer containing 120 mM NaCl, 5 mM KCl, 2 mM CaCl2, and 1 mM MgCl2 (pH 7.4) at room
temperature for 1 h in duplicate in a volume of 250 μL with 10-200 μg membrane protein
depending on receptor expression level. Incubations were terminated by filtration through
Whatman GF/B filters under vacuum, and radioactivity on filters was measured. Competitive
inhibition of [3H]SCH23390 (2 nM) binding to the D1 and D5 receptors or [3H]methylspiperone
(1 nM) binding to the D2, D3, and D4 receptors by tetrahydroprotoberberine derivatives was
performed with various concentrations (10-11 M to 10-5 M). Binding data were analyzed with the
Prism program (GraphPad, San Diego, CA) and Kd, Bmax, and Ki values were determined.
Table 2: Ki values (nM) of tetrahydroprotoberberine derivatives binding to the human dopamine
D1, D4, and D5 or D2 and D3 receptors stably expressed in HEK293 cells or CHO cells.
Competitive inhibition by three compounds of [3H]SCH23,390 binding to D1 and D5 receptors
and [3H]N-methylspiperone binding to D2, D3, and D4 receptors was done and the Ki values were
determined. Each value represents mean ± s.e.m. of three experiments performed in duplicate.
D1
153 ± 9
D2
D3
D4
˃1000
77.4 ± 17.5
>1000
D5
L-THP*
1125 ± 97
41.8 ± 1.8
42.4 ± 3.9
533 ± 86
80.2 ± 9.6
49.2 ± 3.7
1371 ± 212
37.3 ± 3.4
262 ± 24
305 ± 65
9.5 ± 1.4*
313 ± 50
242 ± 33
L-ICP (14a)*
L-CP (14b)
L-CD (14c)
L-THB (14e)
L-DTHB (14f)
6.2 ± 0.1
199 ± 19
107 ± 16
48.6 ± 4.4
6.5 ± 0.7
1713 ± 545
>1000
433 ± 45
82.4 ± 0.5
486 ± 49
1978 ± 70
62.1 ± 6.0
15.1 ± 0.9
* Data from 13
Table 2 shows competitive inhibition by L-THP and its demethylated analogs of radioligand
binding to dopamine receptors. L-THP has a low affinity for D1 and D5 (Ki of 140 nM and 305
nM) receptors and does not appear to bind to D2, D3, or D4 receptors (Ki >1,000 nM). L-THB,
and L-DTHB showed high to moderate affinity for D1, D2 and D5, low affinity for D3 and D4.
L-ICP showed high to moderate affinity for D1, D2, D3, D4 and D5. L-CP and L-CD have similar
agonists activity with L-THP.
A great deal of research has focused on candidate medications to modulate dopaminergic
mechanisms through which psychomotor stimulants produce addictive effects, i.e., drugs that
expressly target the DA transporter, D1-like receptor, or D2-like receptors15-17. However, effective
dopaminergic-based pharmacotherapies to manage the widespread use of psychomotor stimulant
addiction have not yet been successful, suggesting that explicitly targeting the DA transporter or
individual DA receptor subtypes may be insufficient to fully blunt the addiction-related effects of
cocaine15-20. In this study, a series of metabolites of L-tetrahydroprotoberberine were
sterospecifically synthesized following the procedure reported in the literature with minor
modifications. The results of dopamine receptor binding revealed that all have stronger affinity
than the parent L-THP. Further functional assay indicated that L-ICP, L-THB, and L-DTHB are
partial antagonists at D1, antagonists at D2, and agonists at D5. In view of the prominent role of
dopaminergic mechanisms in cocaine’s abuse-related effects, an alternative, and perhaps more
attractive approach for treating cocaine addiction may lie in the use of pharmacological agents that
concurrently target both DA D1- and D2-like receptor subtypes. Their effects on cocaine
self-administration and reinstatement of cocaine seeking following extinction are being

investigated.
Acknowledgements
We thank the National Institutes of Health for financial support under grant
R01-AT-006899-4.
Reference
1. Mariani JJ, Levin FR. Psychiatr Clin North Am. 2012, 35, 425.
2. Haile CN, Mahoney JJ, Newton TF, De La Garza R. Pharmacol Ther, 2012, 134, 260-277.
3. Shorter D, Domingo CB, Kosten TR. Expert Opin Emerg Drugs, 2015, 20:15.
4. Xi, Z. X. & Gardner, E. L. Curr Drug Abuse Rev. 2008, 1, 303.
5. Wang JB, Mantsch JR. Future Med Chem, 2012, 4, 177.
6. Jin, G. Z. l-Tetrahydropalmatine and its analogues as new dopamine receptor antagonists
Trends Pharmacol Sci, 1987, 8, 81-82.
7. Yang, Z., Shao YC, Li SJ, Fan M, Zhang MJ, Hao W, Jin GZ. Acta Pharmacol Sin, 2008, 29,
781.
8. Chu H, Jin G, Friedman E, Zhen X. Cell Mol Neurobiol, 2008, 28:491.
9. Huang KX, Jin GZ. Sci China B, 1992, 35, 688.
10. Ma ZZ, Xu W, Jensen NH, Roth BL, Liu-Chen LY, Lee DY. Molecules, 2008, 25, 13, 2303.
11. Weller D.D., Stirchak E.P., Yokoyama A. J Org Chem, 1984, 49, 2061.
12. Barton AC, Black LE, Sibley DR. Mol Pharmacol. 1991, 39, 650.
13. Xu W, Wang Y, Ma Z, Chiu YT, Huang P, Rasakham K, Unterwald E, Lee DY, Liu-Chen LY.
Drug Alcohol Depend. 2013, 133, 693.
14. Li S, Zhu J, Chen C, Chen Y.W, DeRiel JK, Ashby B, Liu-Chen L-Y. Biochem J, 1993, 295
(Pt. 3), 629.
15. Platt DM, Rowlett JK, Spealman RD. Psychopharmacology. 2002, 163, 265.
16. Gianluigi Tanda, Amy Hauck Newman, Aaron L. Ebbs, Valeria Tronci, Jennifer L. Green,
Ronald J. Tallarida, and Jonathan L. Katz. J Pharmacol Exp Ther, 2009, 330, 802.
17. Newman AH, Blaylock BL, Nader MA, Bergman J, Sibley DR, Skolnick P. Biochem
Pharmacol, 2012, 84, 882.
18. Vocci FJ, Acri J, Elkashef A. Am J Psychiatry, 2005, 162, 1432.
19. Howell LL, Kimmel HL. Biochem Pharmacol, 2008, 75, 196.
20. Negus SS, Mello NK, Blough BE, Baumann MH, Rothman RB. J Pharmacol Exp Ther, 2007,
320, 627.

Graphical abstract