Reactivity of the nitrogen-centered tryptophanyl radical in the catalysis by the radical SAM enzyme NosL

Article abstract of DOI:10.1039/C6CC08869D

The radical SAM tryptophan (Trp) lyase NosL involved in nosiheptide biosynthesis catalyzes two parallel reactions, converting l-Trp to 3-methyl-2-indolic acid (MIA) and to dehydroglycine and 3-methylindole, respectively. The two parallel reactions diverge from a nitrogen-centered tryptophanyl radical intermediate. Here we report an investigation on the intrinsic reactivity of the tryptophanyl radical using a chemical model study and DFT calculations. The kinetics of the formation and fragmentation of this nitrogen-centered radical in NosL catalysis were also studied in detail. Our analysis explains the intriguing catalytic promiscuity of NosL and highlights the remarkable role this enzyme plays in achieving an energetically highly unfavorable transformation.

Full text of DOI:10.1039/C6CC08869D

View Article Online
View Journal
ChemComm
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: H. Qianzhu, W. Ji,
X. Ji, L. Chu, C. Guo, W. Lu, W. Ding, J. Gao and Q. Zhang, Chem. Commun., 2016, DOI:
10.1039/C6CC08869D.
This is an Accepted Manuscript, which has been through the
Volume 52 Number
1 4 January 2016 Pages 1–216
Royal Society of Chemistry peer review process and has been
accepted for publication.
ChemComm
Accepted Manuscripts are published online shortly after
Chemical Communications
www.rsc.org/chemcomm
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1359-7345
COMMUNICATION
Marilyn M. Olmstead, Alan L. Balch, Josep M. Poblet, Luis Echegoyenet al.
Reactivity differences of Sc3N@C2n (2n 68 and 80). Synthesis of the
first methanofullerene derivatives of Sc N@D5h-C80
=
3
rsc.li/chemcomm

Page 1 of 4
ChemComm
View Article Online
DOI: 10.1039/C6CC08869D
COMMUNICATION
Reactivity of the nitrogen-centered tryptophanyl radical in the
catalysis of the radical SAM enzyme NosL
Received 00th January 20xx,
Accepted 00th January 20xx
Haocheng Qianzhu,‡ab Wenjuan Ji,‡b Xinjian Ji,b Leixia Chu,a Chuchu Guo,b Wei Lu,b Wei Ding,bc
Jiangtao Gao,*a and Qi Zhang*b
DOI: 10.1039/x0xx00000x
www.rsc.org/
The radical SAM tryptophan (Trp) lyase NosL invovled in centered radicals are also produced by other radical SAM
nosiheptide biosynthesis catalyzes two parallel reactions, enzymes, including the hydrogenase-maturating enzyme
converting L-Trp to 3-methyl-2-indolic acid (MIA) and to HydG,12-15 thiamine biosynthesis protein ThiH,16, 17 and the F420
dehydroglycine and 3-methylindole, respectively. The two parallel biosynthesis protein CofH.18, 19 However, all these enzymes
reactions diverge from a nitrogen-centered tryptophanyl radical cleave the Cα-Cβ bond of L-Tyr, not the Cα-C bond as found for
intermediate. Here we report investigation on the intrinsic NosL. Interestingly, NosL also cleaves the Cα-Cβ bond of L-Trp
reactivity of the tryptophanyl radical by chemical model study and with an efficiency comparable to that of the Cα-C bond
by DFT calculations. The kinetics in the formation and cleavage, resulting in the production of dehydroglycine (DG)
fragmentation of this nitrogen-centered radical in NosL catalysis and 3-methylindole (MI) as shunt-pathway products (Fig. 1).7, 8
were also studied in detail. Our analysis explains the intriguing The fact that a wild type enzyme catalyzes two parallel
catalytic promiscuity of NosL and highlights the remarkable role reactions on its natural substrate is
a highly unusual
this enzyme plays in achieving a energetically highly unfavorable observation in enzymology,20 raising intriguing questions
transformation.
regarding the reactivity of
enzyme plays in the catalysis.
1
and the energetic role that
The tryptophan (Trp) lyase NosL cleaves the Cα-C bond of L-Trp
and converts it to 3-methyl-2-indolic acid (MIA) (Fig. 1), which
is an intermediate involved in the biosynthesis of a clinically
interesting thiopeptide antibiotic nosiheptide.1, 2 NosL belongs
to the radical S-adenosylmethionine (SAM) superfamily, a large
and rapidly growing enzyme superfamily currently containing
over 165, 000 predicted members.3-5 Enzymes of the radical
SAM superfamily share a common mechanism in radical
generation, utilizing a [4Fe-4S] cluster to bind SAM and
reductively cleave its carbon-sulfur bond to produce a highly
reactive 5′-deoxyadenosyl (dAdo) radical, which then initiates
remarkably diverse reactions.3-5 Structural and biochemical
studies have demonstrated that the NosL-catalyzed reaction is
initiated by the dAdo radical-mediated hydrogen abstraction
from the L-Trp amino group,6-10 which results in a nitrogen-
NH
H
O-
NH
COO-
NH
COO-
+
O
N
N
2
H
H
L-Trp
dAdo
N
HC=NH
H
O-
O
O-
O
C
-C
cleavage
dAdoH
N
N
H
NH
H
MIA
O-
C
-C
NH
COO-
cleavage
[H]
+
O
H
N
1
N
N
DG
H
3
H
H
MI
Fig. 1. Two parallel reactions catalyzed by NosL. The nitrogen-
centered tryptophanyl radical is shown in a yellow eclipse,
1
and the two fragmentation patterns of
and purple.
1 are shown in green
centered tryptophanyl radical (
then proceeds via cleavage of the Cα-C bond of
subsequent attachment of the carboxylate radical onto the
1) (Fig. 1). MIA biosynthesis
1
and
To investigate the intrinsic reactivity of
1, we used a
indole C2 to produce an indolyl radical 2 11
.
Upon Cα-Cβ
method similar to that reported by Lin and Li in studying the
reactivity of thymine radicals.21 In the crystal structure of NosL,
the guanidinium moiety of Arg323 binds the carboxyl group of
L-Trp and compensates its negative charge.6 Because the
protonation state of a carboxylate could vary and complicate
cleavage and a further isomerization process,
the final product MIA (Fig. 1).
2
is converted to
Similar to the NosL-catalyzed reaction, the nitrogen-
the analysis,
a chemical model for investigation. Indeed, density functional
theory (DFT) calculations showed that and exhibit similar
energetic profiles in fragmentation (Fig.S1, ESI† and vide infra).
We therefore synthesized compound and cleaved its N-S
5 (the methyl ester form of 1, Fig. 2a) was used as
5
1
4
bond by photolysis (Fig. 2a). Liquid chromatography (LC) with
high-resolution (HR) mass spectrometry (MS) analysis of the
reaction mixture showed the production of two compounds X1
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
ChemComm
Page 2 of 4
View Article Online
DOI: 10.1039/C6CC08869D
COMMUNICATION
and X2, which exhibit the same protonated molecular ions at product of the Cα-C cleavage of
Journal Name
5
(Fig. 2a), whereas production
m/z = 501.160 (Fig. 2b), and UV-vis spectra that are very of X1 and X2 requires successful addition of
similar to that of compound (Fig.S2, ESI†). X1 and X2 also subsequent oxidative aromatization process (Fig. S5, ESI†), and
exhibit the same set of fragment ions in HR-MS/MS analysis, can also be converted to other products such as 3-
including several ions that are also observed in the MS/MS methylindole (Fig. S8, ESI†). As a result, the ratio of Cα-C to Cα-
spectrum of , and a unique fragment ions at m/z = 283.054 Cβ fragmentation of is lower than 5%, which is in contrast to
(Fig. S3-S4, ESI†). These results suggest that X1 and X2 are two the ratios of fragmentation in NosL catalysis (ranging from
regioisomers resulting from addition of the 3-methylindole 30% to 75% in different reaction conditions).2, 8 This analysis
radical ( ) to the nitrophenyl moiety of , and apparently, is provides direct evidence that, although both Cα-C and Cα-Cβ
resulted from homolytic cleavage of the Cα-Cβ bond of the bond cleavages of the nitrogen-centered radical are possible,
nitrogen-centered radical (Fig. 2a). DFT calculations and the latter process is more favorable than the former.
orbital analysis showed that addition of to occurs To further investigate the reactivity of the nitrogen-
preferentially at the two positions ortho to the nitro group centered tryptophanyl radical , we carried out DFT calculation
3 to 4 and a
4
3
4
5
1
3
4
3
5
3
4
1
(Fig. S5, ESI†), suggesting that the 3-methylindole moiety in X1 on the possible reaction pathways. We performed calculations
(major product) and X2 (minor product) are at the nitro ortho- using both B3LYP and M06 hybrid functionals in parallel and
and meta-position, respectively (Fig. 2a). We also observed a the SMD solvation model22 in both water and ethanol, and all
series of peaks (Y1, Y2, and Y3) that exhibit the same calculations show similar results (Fig. 3 and Fig. S9, ESI†).
protonated molecular ions at m/z = 525.090, which are Because the L-Trp carboxylate interacts with Arg323,6 it was
resulted from addition of the thyil radical
However, distinct from the methylindole radical
electron-rich radical, is much more electrophilic and the relatively low activation barrier (36.3 kJ/mol, hereafter values
radical addition reactions preferentially occur on the indole are from calculations at the B3LYP/6-311+G(2d,p)/SMD(water)
ring of
, as revealed by DFT calculation and orbital analysis level) (Fig. 3), which is similar to that reported by Nicolet et al.6
(Fig. S6, ESI†); this proposal is supported by comparative HR- Cleavage of the Cα-C bond of , however, is much more
6
to
4
(Fig. 2). treated as a protonated carboxyl group in our calculation.
3
that is an These analyses show that cleavage of Cα-Cβ bond has a
1
6
4
1
MS/MS analysis (Fig. S7, ESI†).
demanding, with an activation barrier of 86.8 kJ/mol, 52.5
kJ/mol higher than that of the Cα-Cβ bond cleavage (Fig. 3).
These results are consistent with the chemical model study
discussed above and indicate that the Cα-Cβ bond of
intrinsically prone to homolytic cleavage, whereas the Cα-C
bond cleavage of is not a favorable process.
1 is
1
Further DFT calculations showed that the addition of the
carboxyl radical to the indole C2 is a strong exothermic process
(ΔG = –64.3 kJ/mol), with an activation barrier of only 30.2
kJ/mol. Remarkably, cleavage of the Cα-Cβ bond of the radical
intermediate
activation barrier of 162.0 kJ/mol. Such a high energetic
barrier suggests that is a stable intermediate in catalysis, and
this is consistent with the work by Nicolet et al. in
identification of by electron paramagnetic resonance (EPR)
2 is energetically highly demanding, with an
2
2
analysis.11 Together, our analysis demonstrates the MIA
production is a highly unfavorable process, indicating that
enzyme should play a remarkable role in tuning the reactivity
of the radical intermediates to achieve MIA production.
Fig. 2. Model study on the intrinsic reactivity of the nitrogen-
centered radical. (a) Photoreaction of the compound
precursor of the nitrogen-centered radical . (b) LC-HRMS
analysis of the photoreaction of . The reaction was carried
out under 254 nm UV light for 60 min; the unreacted solution
of was also analyzed for comparison.
4, a
Previous biochemical studies on NosL have shown that
multiple deuterium was incorporated into dAdoH when the
reactions were performed in D2O, suggesting that the dAdo
5
4
radical-mediated hydrogen abstraction that produce
reversible process.7, 8, 23 To study the rates for
1 is a
4
1
formation and
cleavage, we ran NosL reaction in highly enriched D2O buffer
(D2O > 98%, pD 8.0) and carried out a time course analysis. LC-
HRMS analysis of the reaction mixture showed that the sum of
di- and tri-deuterated dAdoH accounts for more than 75% of
the resulting dAdoH, and the ratios of multiply deuterated
dAdoH species remain roughly constant at different time
points (Fig. 4a). We noted that even for the reaction that was
rapidly quenched within 3 seconds, the resulting dAdoH is still
mainly tri-deuterated (Fig. 4a).
In contrast to the products originated from the Cα-Cβ bond
cleavage of , 3-indoleacetaldehyde (IA), the product of the
Cα-C cleavage of , was barely observable in HPLC analysis.
5
5
The reaction mixture was treated with an excess amount of
NaBH4 to reduce any IA, and the resulting 3-indoleethanol was
concentrated and carefully quantified with the standard. This
analysis showed that the yield of IA is 15-fold lower than the
total amount of X1 and X2 (Fig. S8, ESI†). IA is the direct
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
ChemComm
Page 3 of 4
View Article Online
DOI: 10.1039/C6CC08869D
Journal Name
COMMUNICATION
Fig.
3. DFT-calculated potential energy profile in NosL catalysis. Energies were calculated at the B3LYP/6-
311+G(2d,p)/SMD(water) level of theory with geometry optimized at the B3LYP/6-31+G(d,p)/SMD(water) level. The potential
energy profiles calculated by using M06 hybrid functionals and SMD solvation model in water and ethanol are shown in Fig S9,
ESI†. All the transition states (shown in brackets) have been validated by intrinsic reaction coordinate (IRC) calculation in both
forward and reverse directions (Fig. S10, ESI†). ΔGsol, free energy of activation; ΔHsol, enthalpy of activation.
We proposed a reaction model for the NosL-catalyzed generally 10～12 folds higher than those of deuterium transfer
hydrogen abstraction in D2O (Fig. 4b, a detailed model is (Table S1, ESI†), and this is consistent with a recent stop flow
shown in Fig. S11, ESI†). Briefly, the dAdo radical abstracts a study on the adenosylcobalamin-dependent ethanolamine
deuterium from the L-Trp amino group to produce
could be converted to the reaction products upon C-C bond the generation and fragmentation of
cleavage (kfrag) (Fig. 4b). On the other hand, can abstract a (Fig. 4b, see Fig. S11, ESI† and the legend for a detailed
1
, which ammonia lyase.28 Based on these results, the reaction rates for
can be roughly inferred
1
1
protium or deuterium from dAdoH to produce dAdo radical discussion).
and L-Trp again, and protium abstraction is kinetically more
favorable (see below). The resulting protium on the L-Trp
amino group is then replaced by a deuterium via a rapid
solvent exchange process (ksol), and thereby each hydrogen
abstraction/rebound cycle leads to accumulation of
a
deuterium into dAdoH (Fig. 4b). The observation that the
multiply deuterated dAdoH accounts for more than 75% of the
total dAdoH suggests that kfrag is significantly smaller than k2.
We next estimated the rate constants of the dAdo radical-
mediated hydrogen abstraction by DFT and statistical
mechanics analysis using the canonical transition state theory
with Wigner tunneling correction (TSTW),24 which takes
quantum tunneling effect25, 26 into consideration. Because the
transition state for
1 formation involves hydrogen transfer
between two radical-like centers, it appears to be insensitive
to the changes away from the reaction center.27 Therefore, in
our analysis dAdo radical is modeled as an ethyl radical and L-
Trp is modeled as methylamine, in a way similar to a recent
study on the radical SAM enzyme HydE.27 By using such small
quantum models, we were able to carry out high accuracy
calculation with large basis set and avoid local minima in the
potential energy surface, which could have severe
consequences on the energetics. Calculations were performed
for all the model hydrogen abstraction reactions to address
the effect of H/D exchange (Fig. S12, ESI†), showing that the
rate constants for the dAdo radical-based protium transfer is
Fig 4. Kinetic study of NosL catalysis. (A) Summed MS spectra
of dAdoH produced in highly enriched (98%) D2O buffer at
different time points. (B) A model for NosL reaction in D2O. For
a detailed model and discussion on NosL kinetics, see Fig. S11,
ESI†. The tryptophanyl radical
1 is shown in red, and the
estimated rate constants are shown in s-1. (C) Summed MS
spectrum of dAdoH produced by the NosL Y90F mutant in 98%
D2O at 5 min.
The observation that a significant proportion of dAdoH was
triply deuteriated in the rapid quench experiment suggests
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
ChemComm
Page 4 of 4
View Article Online
DOI: 10.1039/C6CC08869D
COMMUNICATION
Journal Name
that the L-Trp amino group undergoes a rapid solvent 4. J. B. Broderick, B. R. Duffus, K. S. Duschene and E. M.
exchange during catalysis. In NosL crystal structure, the L-Trp Shepard, Chem Rev, 2014, 114, 4229-4317.
amino group forms a hydrogen bond with Tyr90, which could 5. B. J. Landgraf, E. L. McCarthy and S. J. Booker, Annu Rev
possibly mediate hydrogen exchange between solvent and the Biochem, 2016.
L-Trp amino (Fig. S13, ESI†). To test this hypothesis, we 6. Y. Nicolet, L. Zeppieri, P. Amara and J. C. Fontecilla-Camps,
replaced Tyr90 with a Phe, and the resulting Y90F mutant was Angew Chem Int Ed Engl, 2014, 53, 11840-11844.
used to run a reaction in highly enriched D2O buffer same to 7. X. Ji, Y. Li, W. Ding and Q. Zhang, Angew Chem Int Ed Engl,
that used for the wild type enzyme. LC-HRMS analysis of the 2015, 54, 9021-9024.
reaction mixture showed that, in stark contrast to the wild 8. D. M. Bhandari, H. Xu, Y. Nicolet, J. C. Fontecilla-Camps and
type NosL (Fig. 4a), dAdoH produced by the mutant enzyme T. P. Begley, Biochemistry, 2015, 54, 4767-4769.
was predominately singly deuterated (Fig. 4c). The catalytic 9. X. J. Ji, Y. Z. Li, Y. L. Jia, W. Ding and Q. Zhang, Angew Chem
efficiency of the Y90F mutant is only about 2-fold lower than Int Edit, 2016, 55, 3334-3337.
that of the wild type enzyme (Fig. S14, ESI†), suggesꢀng that 10. X. Ji, Y. Li, L. Xie, H. Lu, W. Ding and Q. Zhang, Angew Chem
the mutant folds in the same way as the wild type enzyme. Int Ed Engl, 2016, 55, 11845-11848.
These results demonstrate that the solvent exchange 11. G. Sicoli, J. M. Mouesca, L. Zeppieri, P. Amara, L. Martin, A. L.
mechanism is significantly, if not completely, abolished in the
Y90F mutant, supporting the essential role of Tyr90 in
Barra, J. C. Fontecilla-Camps, S. Gambarelli and Y. Nicolet,
Science, 2016, 351, 1320-1323.
mediating the hydrogen exchange between solvent and the L- 12. J. M. Kuchenreuther, W. K. Myers, T. A. Stich, S. J. George, Y.
Trp amino group. In NosL active site, no water molecule was
found around the phenol moiety of Tyr90 and L-Trp amino
Nejatyjahromy, J. R. Swartz and R. D. Britt, Science, 2013,
342, 472-475.
group.6 Our analysis thus suggests that generation of
1
13. B. R. Duffus, S. Ghose, J. W. Peters and J. B. Broderick, J Am
Chem Soc, 2014, 136, 13086-13089.
involves multiple rounds of solvent molecules move in and out
of the enzyme active site. It remains to be seen whether the 14. P. Dinis, D. L. Suess, S. J. Fox, J. E. Harmer, R. C. Driesener, L.
dynamic solvent exchange process contributes to the reaction
thermodynamics.
De La Paz, J. R. Swartz, J. W. Essex, R. D. Britt and P. L. Roach,
Proc Natl Acad Sci U S A, 2015, 112, 1362-1367.
In summary, by using a combination of model chemical 15. A. Pagnier, L. Martin, L. Zeppieri, Y. Nicolet and J. C.
study and DFT calculations, this investigation clearly
demonstrates that cleavage of the Cα-Cβ of the nitrogen-
Fontecilla-Camps, Proc Natl Acad Sci U S A, 2016, 113, 104-
109.
centered tryptophanyl radical
favorable than that of the Cα-C cleavage. We also estimated
the rate constants for the formation and fragmentation of
NosL catalysis, and showed that a key Tyr residue mediates
1 is thermodynamically more 16. N. C. Martinez-Gomez, M. Robers and D. M. Downs, J Biol
Chem, 2004, 279, 40505-40510.
in 17. M. R. Challand, F. T. Martins and P. L. Roach, J Biol Chem,
2010, 285, 5240-5248.
1
rapid hydrogen exchange between solvent and the L-Trp 18. L. Decamps, B. Philmus, A. Benjdia, R. White, T. P. Begley and
amino group. Our study demonstrates the remarkable role O. Berteau, J Am Chem Soc, 2012, 134, 18173-18176.
that NosL plays in modulating the reactivity of radical 19. B. Philmus, L. Decamps, O. Berteau and T. P. Begley, J Am
intermediates, indicating that the unusual catalytic Chem Soc, 2015, 137, 5406-5413.
promiscuity of NosL is likely a compromise to achieve an 20. W. Ding, X. Ji, Y. Li and Q. Zhang, Front Chem, 2016,
energetically highly demanding reaction. Similar scenarios of 21. G. Lin and L. Li, Angew Chem Int Ed Engl, 2013, 52, 5594-
4, 27.
catalytic promiscuity likely apply to other radical-mediated
enzymatic reactions.
5598.
22. A. V. Marenich, C. J. Cramer and D. G. Truhlar, J Phys Chem B,
2009, 113, 6378-6396.
We thank Prof. Tadhg P. Begley (Texas A&M University) for
helpful discussions on NosL kinetics. This work was supported 23. X. Ji, W. Q. Liu, S. Yuan, Y. Yin, W. Ding and Q. Zhang, Chem
by grants from Fujian Agriculture and Forestry University and Commun (Camb), 2016, 52, 10555-10558.
from Fudan University, from National Natural Science 24. a. D G Truhlar and R. E. Wyatt, Annu Rev Phys Chem, 1976,
Foundation of China (1500028 to Q.Z.), and from the National 27, 1-43.
Key Research and Development Program (2016 Y F A0501302 25. A. M. Kuznetsov and J. Ulstrup, Can J Chem, 1999, 77, 1085-
to Q.Z.).
1096.
26. Z. D. Nagel and J. P. Klinman, Nat Chem Biol, 2009, 5, 543-
Notes and references
550.
27. R. Rohac, P. Amara, A. Benjdia, L. Martin, P. Ruffie, A. Favier,
O. Berteau, J. M. Mouesca, J. C. Fontecilla-Camps and Y.
1. Y. Yu, L. Duan, Q. Zhang, R. Liao, Y. Ding, H. Pan, E. Wendt-
Pienkowski, G. Tang, B. Shen and W. Liu, ACS Chem. Biol.,
Nicolet, Nat Chem, 2016, 8, 491-500.
2009,
2. Q. Zhang, Y. Li, D. Chen, Y. Yu, L. Duan, B. Shen and W. Liu,
Nat. Chem. Biol., 2011, , 154-160.
4, 855-864.
28. A. R. Jones, J. Rentergent, N. S. Scrutton and S. Hay,
Chemistry, 2015, 21, 8826-8831.
7
3. P. A. Frey, A. D. Hegeman and F. J. Ruzicka, Crit. Rev.
Biochem. Mol. Biol., 2008, 43, 63-88.
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins