Q. Zhang, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127615
non-canonical poly(A) RNA polymerase associated domains containing
proteins 5 and 7 (PAPD5 and PAPD7) were identified as the potential
protein targets by yeast three-hybrid (Y3H) system and validated by
oligonucleotide-mediated knockdown.12 PAPD5 and PAPD7 are re-
sponsible for cellular RNA homeostasis via PAPD5/Zinc Finger CCHC-
cells.13 While genetic studies showed that both the La binding sequence
(HPRE) are critical for the DHQ-1′s activity.11 Based on these, we be-
lieve DHQs might inhibit HBV virus and affect host cell with multiple
targets, which offers an unprecedented opportunity to advance novel
HBV therapeutics. Further mechanism of action studies of DHQs with
anti-HBV activities would help optimize DHQ structures to increase
HBV cure rates and decrease side effects.
activity evaluation of these two probes (7 and 8) showed that com-
pound 7 is very potent with ~5 nM EC50 against both HBV DNA and
HBsAg, while the activity of its enantiomer (8) decreases by ~50 folds.
In addition, both compounds did not exhibit cytotoxicity against
Hep2.2.15 cell line.
Photoaffinity labelling in whole proteome. All the obtained probes
were fully characterized prior to any biological experiments.
Subsequently, the probes were applied to the whole proteome photo-
affinity labelling (Fig. 1 a). Before applying the photoaffinity labeling,
into live cells, the recognition group of these cell-permeable probes
would drive the photoaffinity probe to the binding proteins. Then, after
photo-irradiation, a highly reactive intermediate carbene would be
generated and covalently incorporated into its binding proteins in a
distance-dependent manner. After the covalently linked protein-probe
complexes were formed, the following emendation could be realized by
clicking with different reporting groups for various subsequent bio-
chemical analysis, including tracking, enrichment and LC–MS/MS
identification. Next, we investigated the labelling profile of these pho-
toaffinity probes in HepAD38 in situ by fluorescence gel-based studies.
The labelling experiments were carried out as the concentration-de-
pendent and incubation time-dependent labelling. After incubating
with the probes, the labelled cells were subjected to UV irradiation and
then lysis. The probe-covalently labelled proteome would be subjected
to Cu(I)-catalysed click (CuAAC) reaction with rhodamine-azide (Rho-
damine-N3) and separated by SDS-PAGE. The labelling profile was
evaluated by in-gel fluorescence, while Coomassie blue staining was
used to indicate that an equal amount of all samples was loaded. The
fluorescence gel scanning showed that several proteins were labelled by
probe 7 with a significant contrast comparing with the negative control
9 and displayed concentration and time dependent manner. It turned
out that final concentration of 20 μM and at least 30 min’s incubation
time were the best for in situ labelling (Fig. 1c) and SI Fig. 1b). The
labelling effect comparison of the two chiral isomers is consistent with
the anti-HBV activities (SI Fig. 1c), meaning the S-isomer (7) performed
better than the R-isomer (8). The anti-HBV activities and labelling
suggested the direction of this side chain affects the binding of the
probe to the target protein. Moreover, the labelling profile of probe 7
exhibited obvious response to different concentrations of the compe-
titor (Fig. 1d), meaning the labelled bands are the binding proteins of
the parental scaffold. The labelling of 7 is much more sensitive to 6 than
1 (SI Fig. 1d), implying that the methyl ester group on 7 might guide
the recognition group to different binding proteins of RG7834. After
characterizing the labelling of 7 in situ, we believed that most of the 7
-labelled protein bands were specific for the recognition scaffold and we
would further identify the binding protein targets of 7 by affinity en-
richment and LC–MS/MS.
Activity/affinity-based proteome profiling (ABPP)14 is to use a bi-
functional probe to immediately capture the protein-probe interactions
in situ leading to subsequent protein enrichment and identification.
Photoaffinity labelling converts the noncovalent (weak) protein-probe
interactions into covalent cross-linking under native cellular conditions
by the photoaffinity probe containing a photoaffinity group, a reporting
group and the bio-active recognition scaffold.15 Biorthogonal chemical
handles could also be used as the reporting group and incorporated on
the main scaffold of the probe to facilitate the following detection and
enrichment. Integrating with quantitative proteomics, ABPP has been
In this report, we optimized the DHQ scaffold and screened the anti-
HBV activity to obtain the potent compound with incorporation of a
bifunctional photoaffinity tag. This DHQ-based probe was further ap-
plied to profile the potential cellular targets.
Structure-activity relationship of the DHQ scaffold against HBV. As
shown in previous studies,11 RG7834 (or DHQ-1) exhibited EC50 at
nanomolar level for viral particle, HBsAg and HBeAg respectively,
which are the designated three viral markers for anti-HBV agents. In
order to identify the suitable site for incorporation of photoaffinity tag
to RG7834 (or DHQ-1), we firstly investigated the influence of the
structural modifications on their anti-HBV effect. Compound 1 (race-
mate of RG7834) and compounds with different modification on R1/
R2/R3/R4 substituent were synthesized and tested for their anti-HBV
activities (Table 1). After analyzing the structure and anti-HBV activ-
minor modification to a methyl ester group and exhibited lower EC50 in
HBsAg (compound 1 vs 2). The bulky substituent on R1 was detrimental
to activity which was shown by compound 3. The lipid chain on R2
position is necessary for the anti-HBV activity as the activity of com-
pound 6 greatly decreased compared with compound 2. However,
when R2 was replaced by pyrrolidine, the activity of compound 4 was
comparable to compound 2, which indicated substituent with more li-
pophilicity on R2 was beneficial to the activity again. We also found the
carboxylic acid group on R4 cannot be modified either (compound 5).
According to the previously reported structure-activity relationship
(SAR) results9 and the SAR investigation in this study, we believe that
compound 2 could serve as the main scaffold for the development of
photoaffinity labelling probe.
LC–MS/MS assisted protein identification. After one hour’s incubation at
the final concentration of 20 μM and photo irradiation, the cells were
harvested and homogenized in HEPES lysis buffer, and then the lysates
were conjugated with biotin-Diazo-azide by cooper catalyzed click reac-
tion. Subsequently, large-scale affinity enrichment and LC–MS/MS ex-
periments were performed to identify the potential cellular targets of 7.
After enrichment, the labelled proteins were separated on SDS-PAGE and
identified by LC–MS/MS after in gel digestion (Fig. 2). Proteins labelled
by the negative probe would be ruled out to obtain the potential target
candidates of the recognition scaffold. In order to greatly decrease the
influence of non-specific targets of the photoaffinity probe or the affinity
resins, the LFQ intensity (positive/negative) ratio of all the identified
targets were ranked. After the MS result analysis, we screened several
potential protein targets, Hsp60, Aldehyde dehydrogenase X, Retinal
dehydrogenase, Isocitrate dehydrogenase (Idh1), HMOX2, Poly(rC)-
clinical index for liver fibrosis. PCBPs, RNA binding proteins, were im-
plicated to be involved in both cellular and viral mRNA stability by
Activity/affinity-based proteome profiling (ABPP) probe design, synth-
esis, and activities against HBV. Thus, we designed and synthesized bi-
functional diazirine-alkyne probes of two enantiomers (7 and 8)
(Scheme 1). The synthetic route to compound 7 starting from methyl 4-
formed by formylation of compound 10, followed by the Bischler-Na-
pieralski cyclization to produce 3,4-dihydroisoquinoline 12. Inter-
molecular cyclization of 12 with 13 yielded compound 14. After re-
moval of the benzyl group protection with Pd/C, alkylation of 15 with
16 afforded compound 17. The photoaffinity labelling probe 7 was
obtained by oxidation and deprotection of 17 in the presence of p-
chloranil. Probe 8 was synthesized following the similar route. The
2
Zhang, Qing
