1
Molecular analysis of some camel cytochrome P450 enzymes reveals lower
2
evolution and drug binding properties
3
4
5
Yousef Altaher1, Mahmoud Kandeel2,3, *
1
Faculty of Veterinary Medicine and animal resources, King Faisal University, Alhofuf,
6
Alahsa, Saudi Arabia;2Department of Physiology, Biochemistry and Pharmacology,
7
Faculty of Veterinary Medicine and animal resources, King Faisal University, Alhofuf,
8
Alahsa, Saudi Arabia; 3Department of Pharmacology, Faculty of Veterinary Medicine,
9
Kafrelshikh University, Kafrelshikh 33516, Egypt,
10
11
12
Short title: camel Cytochrome P450
13
14
15
16
17
18
1
1
Abstract
2
Camels bear unique genotypes and phenotypes for adaptation of their harsh
3
environment. They have unique visual systems, sniffing, water metabolism, heat control
4
mechanisms that are different from other creatures. The recent announcement for the
5
complete sequence of camel genome will allow for discovery of many secrets of camel
6
life. In this context, the genetic bases of camel drug metabolizing enzymes are still
7
unknown. Furthermore, the genomic content of camel that rendered it highly susceptible
8
to some drug (as monensin and salinomycin) and became easily intoxicated needs to be
9
investigated. The objectives of this work are annotation of camel genome and retrieval
10
of camel for cytochrome P450 1A1, 2C and 3A enzymes. This is followed by
11
comprehensive phyolgentic, evolution, molecular modeling and docking studies. In
12
comparison with the human enzymes, camel CYPs showed lower evolution rate
13
especially
14
alfanaphthoflavone, felodepine and ritonavir was weaker in camel enzymes.
15
Interestingly, rerank score indicated instable binding of monensin and salinomycin with
16
camel CYP1A1 as well as salinomycin with camel CYP2C. The results of this work
17
suggest that camels are more susceptible to toxicity with compounds undergoing
18
metabolic oxidation. This conclusion was based on lower evolution rate and lower
19
binding potency of camel s compared with the human enzymes.
CYP1A1.
Furthermore,
the
binding
of
monensin,
20
21
Keywords
22
Camel, cytochrome P450, phylogenetics, docking, molecular modeling
23
2
salinomycin,
1
Introduction
2
Camels are reared in arid and semi-arid harsh environment. In order to adapt such
3
habitat, camels developed unique molecular pathways including rapid evolution or
4
regression of some genes, new genotypes and phenotypes. In this regard, camels
5
developed unique pathways to cope with their surrounding environment. For instance,
6
camels can withstand deprivation of water, heat and dryness for long time. This is
7
accompanied by complex regulation processes, which control water metabolism and
8
whole body metabolic processes. Thermographic tracing of camels body temperature
9
reveals daily changes in body temperature across the day and night in a temperature
10
range from 34-41°C. The maximal and minimal temperatures are within the afternoon
11
and early morning, respectively. Therefore, the flow of heat to the body of camels from
12
the surrounding desert is minimized and water loss by perspiration is reduced.
13
Interestingly, camels has high sweating threshold, in which camels start sweating when
14
their body temperature exceeds 42°C (Kohler-Rollefson, 1993). Camels mitochondrial
15
enzymes are characterized by high evolution rate and contribute to the adaptation of
16
camels to live in different environments (Di Rocco et al., 2006; Di Rocco et al., 2009).
17
Camels are raised for production of wool, milk, leather and meat. Additionally, it is used
18
in transportation and management of agricultural crops in arid and semiarid regions.
3
1
Increased interest in camels rearing is increasing nowadays for human nutrition and
2
production of modern therapeutics. Camels nanobodies derived from the unique camel
3
single domain antibodies are new interesting class of therapeutic and diagnostic
4
reagents (Badr, 2013; Fraile et al., 2013; Korish, 2014; Unciti-Broceta et al., 2013).
5
Camels are intoxicated with several drugs, which seems to be used safely in other
6
animals (Ali, 1988; Alquarawi and Ali, 2000; el Bahri et al., 1999; Mousa and Elsheikh,
7
1992). For instance, camels cannot tolerate ionophores as monensin, and salinomycin
8
(Abu Damir et al., 2013; Miller et al., 1990; Mousa and Elsheikh, 1992). In comparison,
9
these compounds (Fig. 1) are well tolerated in other animals as poultry, sheep and cattle.
10
These ionophores are used in veterinary practice for prophylaxis against protozoal
11
diseases, growth promotion and antibacterial effect. For comparison, monensin at dose
12
rate of 0.6 mg/kg is lethal for camels, while chicken can resist up to 50 mg/kg. In
13
another example, diminazine aceturate was found to be more toxic in camels than other
14
species (Homeida et al., 1981).
15
The full sequence of camels genome has been published (Wu et al., 2014). The
16
discovery of camel genome sequence will help in resolving the secrets of camel’s
17
molecular processes and mechanisms of adaptation. In this context, the nature of drug
18
metabolism and its associated camel genetic background is not investigated. It is not
4
1
known whether camel cytochrome P450 enzymes are specifically evoluted or adapted
2
for desert life. This study investigates three camel cytochrome P450 enzymes (CYP),
3
CYP1A1, CYP2C and CYP3A. This work will be the first investigation of camel
4
genome sequence and its pharmacogenomic implications. Cytochrome P450 enzymes
5
are the major drug metabolizing enzymes by performing more than 80% of the
6
metabolic functions. The metabolism of monensin was found to be mainly by CYP
7
oxidation (Nebbia et al., 1999; Nebbia et al., 2001). In this study, we show the lower
8
evolution rate and suboptimal drug binding capacity of camel CYPs compared with the
9
human enzymes. This might suggest the reason beyond camel toxicity with these
10
compounds.
11
12
Methods
13
1- Annotation of camel genome
14
Genomic
15
(http://tritrypdb.org/tritrypdb/),
16
(http://www.ncbi.nlm.nih.gov)
17
(http://www.camel.kacst.edu.sa). Inspection and analysis of ESTs was carried out by
18
CLC genomics workbench version 7.5.1. About 17155 contigs were retrieved and kept
19
as local BLAST database.
data
are
extracted
from
the
protein
and
Kinetoplastom
and
the
20
5
Arabian
genome
camel
genome
resources
databases
genome
at
project
1
2- Retrieval of camel CYP1A1, CYP2C and CYP3A enzymes
2
The sequences of genes encoding CYP1A1, CYP2C and CYP3A were searched in
3
nucleotide databases and the human, Camelus ferus and Camelus bactrianus sequences
4
are saved. The retrieved sequences are used to BLAST the local database of camel ESTs
5
library to retrieve the hits of highest similarity. The retrieved E values were as from
6
1.16 E-130 to zero. The obtained contigs were retrieved and subjected for further
7
molecular modeling and phylogenetic studies.
8
We tried to do the above mentioned procedure for CYP4A enzyme. However, we were
9
unable to find any hits with sufficient E value. So that we excluded CYP4A from
10
further investigations. We ran another EST assembly run by using CLC genomics
11
workbench De Novo assembly from which we retrieved new 2466 contigs.
12
Unfortunately, BLAST search with human CYP4 sequence did not came back with
13
significant similarities.
14
15
3- BLAST search and sequence alignment
16
The retrieved camel sequences were used to BLAST non-redundant nucleotides
17
databases at NCBI. All organisms are searched with expect value of 10. The top 100
18
hits with the lowest E value were downloaded and kept in one batch file for sequence
19
alignment. Sequence alignment was set to very accurate.
20
21
4- Phylogenetic and evolutionary analysis
22
In order to evaluate the evolution rate of camel enzymes we used Bayesian evolutionary
23
analysis under relaxed phylogenetic models by using BEAST (Bayesian Evolutionary
24
Analysis Sampling Tree) software package version 2.1.3 (Drummond and Rambaut,
6
1
2007; Drummond et al., 2012). The nucleotide sequences from different hits were
2
exported in NEXUS format by using geneious 7.1.7 software package. The BEAUti
3
software was used to convert NEXUS file to xml format for Bayesian evolutionary
4
analysis. Substitution rate was set for 1, gamma category count was set to 4 and the
5
JC69 was used as a substitution model. Timing data were not provided and the branch
6
lengths represented the substitutions per year with an assumed average of 1. Relaxed
7
clock log normal was set for clock model at clock rate of 1 and uniform birth rate.
8
Marcov Chain Monte Carlo (MCMC) mathematical model is used by BEAST, the
9
length of chain was set to 1000000 and the log file is saved for tree annotation by
10
TreeAnnotator software. The produced tree is visualized and annotated by FigTree
11
software. Examination of MCMC results were viewed by Tracer program.
12
13
5- Molecular modeling studies
14
Requests for building molecular models for the camel enzymes were submitted to
15
SWISS Model server. Models of camel CYPs were built on their corresponding human
16
enzymes templates in automated mode of SWISS Model server. The used human CYPs
17
templates were PDB ID no. 4I8V, 2NNJ and 4NXU. The obtained structure models
18
were checked, minimized, water removed and prepared for docking studies (Fig. 2).
19
20
6- Molecular docking studies
21
Molecular docking studies were carried out by Molegro virtual docker. Monensin and
22
salinomycin were selected for comparative docking studies (Fig. 1). Furthermore, each
23
derived human PDB template contains a drug bound to its active site. These drugs are
24
alfa naphthflavone, felodepine and ritonavir for 4I8V, 2NNJ and 4NXU, respectively.
7
1
These compounds were also used for docking of both camel and human enzymes. The
2
accuracy of docking run was evaluated by comparison of the docked poses with the
3
conformation of compounds in their sites of crystal structures.
4
For preparation of compounds for docking, SDF files were exported from PubChem
5
database. The energy was minimized and conformationally optimized by Chem3D Pro,
6
Cambridge soft and saved as MDL MOL files. Docking site was based on template
7
docking after concluding the forces of interaction. For camel enzymes, the docking site
8
was assigned from the respective human interactions. The docking session was
9
performed for 10 runs for each compound under 1500 maximal iterations per
10
compound.
11
12
Results
13
Sequence alignment
14
Sequence alignments of human and camel CYP1A1, CYP2C and CYP3A are shown in
15
Fig. 2. The alignment results indicate high similarity during within the same CYP
16
family and low similarity among different families. The similarity of human and camel
17
enzymes within CYP1A1, CYP2C and CYP3A were 82.7, 68.7 and 75%, respectively.
18
In contrast, comparison of camel CYP enzymes with each other or human CYP enzymes
19
with each other gives low similarity, range from 20-31%.
20
21
Docking studies
8
1
The selected docking output included hydrogen bonding, MolDock score and rerank
2
score. MolDock score is the default output of docking energy. Rerank score is used to
3
assess the stability of binding. Following docking run, rerank of the output poses is
4
performed. Rerank score is a modification of docking score with considerations of steric
5
hindrance. Thus, rerank score is more computationally valuable than the docking score.
6
For CYP1A1, zero value of hydrogen bonding of naphthoflavone indicates the general
7
capacity of CYP1A1 to bind and metabolize substrates. The negative value of MolDock
8
score indicates the potential binding of all drugs to human and camel CYP1A1.
9
However, rerank score values indicates instability of binding salinomycin with human
10
CYP1A1 and instable binding of monensin and salinomycin with camel CYP1A1
11
(Table 1).
12
For CYP2C, all compounds showed favourable and stable binding with both human and
13
camel enzymes except salinomycin which is expected to form unstable complexes with
14
the camel enzyme (Table 2). All compounds were found to bind to CYP3A with
15
potential stability.
16
For all analyzed enzymes and compounds, results from camel enzymes showed inferior
17
binding potency by showing lower docking and rerank scores in all compounds (Tables
18
1, 2 and 3).
9
1
2
Phylogenetic analysis and gene evolution
3
Phylogenetic analysis of CYP1A1, CYP2C and CYP3A are shown in figures 3, 4 and 5,
4
respectively. The selected phylogenetic parameters are represented in Table 4. For easier
5
identification, several markers are applied for conclusions about gene evolution rates.
6
These include the colours of branches and species, the size of the node or the numbers
7
wrote for each branch. The scale of rate is provided in the top left corner of each figure.
8
Red colour denotes the highest rate. Similarly, the thicker branch line indicates higher
9
rate. The estimated rate of evolution was lower in camel enzymes compared to human
10
enzymes (Table 4). The values of camel CYP2C and CYP3A were almost comparable
11
to average rate determined by Bayesian analysis. Adversely, Camel CYP1A1 showed
12
lower evolution rate.
13
14
Discussion
15
Camels are thought to be evolved several million years ago. They gradually adapted to
16
their surrounding environment by genetic adaptation and phenotype variation. Camels
17
were domesticated about 3000-6000 years ago (Burger and Palmieri, 2014). During
18
domestication, adaptation to the new environment elicited further variations in the
10
1
genetic composition. By the advent of new genome technologies, it became easier to
2
map genetic variations and changes in genes by comparison of domesticated and wild
3
animals. Furthermore, phylogenetic and other computational tools improved the
4
understanding of biological functions, disease control and discovery of drugs (Kandeel
5
et al., 2014; Kandeel and Kitade, 2013a, b). The selection pressure for rapid evolution,
6
slow evolution or loss of certain gene can be analyzed and compared between the
7
Arabian camel and wild camels. Understanding these changes highlight the
8
physiological status of animals, their response to drugs and potential success or failure
9
of treatment.
10
Clinical and experimental practice with camel diseases revealed that camels are resistant
11
to toxicity of some compounds and were also very susceptible to toxicity of other drugs.
12
In this work, some drug metabolizing enzymes (CYP1A1, CYP2C and CYP3A) in
13
camels are investigated to reveal their role in susceptibility or resistance of camels to
14
drug toxicity. Furthermore, phylogenic approaches are undertaken to assess the
15
evolution rates of camel enzymes.
16
Phylogenetic analysis revealed a lower rate of evolution of camel enzymes compared
17
with that of human cytochromes. Camel CYP2C and CYP3A did not show major
18
changes compared with the tested dataset. In contrast, CYP1A1 showed considerably
11
1
low rate of evolution. In this regards, camel CYP1A1 docking data revealed instable or
2
potential lack of binding with monensin and salinomycin. In parallel with the
3
phylogenetics analysis, docking data also confirmed lower efficiency of camel
4
cytochrome enzymes in binding different drugs compared with the human enzymes.
5
Lower rate of evolution of camel enzymes, lower binding with different drugs and
6
instability of binding of some drugs highlights the potential susceptibility of camels to
7
toxicity with drugs. Monensin and salinomycin, which are highly fatal to camels might
8
have lower binding capacity to the camel enzymes allowing for toxicity. Both monensin
9
and salinomycin were bound effectively by camel CYP3A but not with CYP1A1 and
10
CYP2C indicating that CYP3A might be the major metabolizing enzyme for these drugs
11
in camel. Similar experimental finding was previously described in cattle, chicken and
12
rats (Nebbia et al., 1999). As described above the binding potency of monensin with
13
camel enzymes is lower than that of human indicating more potential toxicity to camels
14
due to slower metabolism.
15
16
Acknowledgements
17
This work is supported by grants for undergraduate students projects to Mahmoud
18
Kandeel and Yousef Altaher (grant number 165033) from the deanship of scientific
12
1
research, King Faisal University, Hofuf, Alahsa, Saudi Arabia.
2
3
References
4
Abu Damir, H., Ali, M.A., Abbas, T., Omer, E., Al Fihail, A., 2013. Narasin poisoning
in the dromedary camel (Camelus dromedarius). Comparative Clinical Pathology, 1-7.
Ali, B.H., 1988. A survey of some drugs commonly used in the camel. Vet Res Commun
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
12, 67-75.
Alquarawi, A.A., Ali, B.H., 2000. A survey of the literature (1995-1999) on the kinetics
of drugs in camels (Camelus dromedarius). Vet Res Commun 24, 245-260.
Badr, G., 2013. Camel whey protein enhances diabetic wound healing in a
streptozotocin-induced diabetic mouse model: the critical role of beta-Defensin-1, -2
and -3. Lipids Health Dis 12, 46.
Burger, P.A., Palmieri, N., 2014. Estimating the Population Mutation Rate from a de
novo Assembled Bactrian Camel Genome and Cross-Species Comparison with
Dromedary ESTs. J Hered 105, 839-846.
Di Rocco, F., Parisi, G., Zambelli, A., Vida-Rioja, L., 2006. Rapid evolution of
cytochrome c oxidase subunit II in camelids (Tylopoda, Camelidae). J Bioenerg
Biomembr 38, 293-297.
Di Rocco, F., Zambelli, A.D., Vidal Rioja, L.B., 2009. Identification of camelid specific
residues in mitochondrial ATP synthase subunits. J Bioenerg Biomembr 41, 223-228.
Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by
sampling trees. BMC evolutionary biology 7, 214.
Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian phylogenetics
with BEAUti and the BEAST 1.7. Mol Biol Evol 29, 1969-1973.
el Bahri, L., Souilem, O., Djegham, M., Belguith, J., 1999. Toxicity and adverse
reactions to some drugs in dromedary (Camelus dromedarius). Vet Hum Toxicol 41,
35-38.
Fraile, S., Jimenez, J.I., Gutierrez, C., de Lorenzo, V., 2013. NanoPad: an integrated
platform for bacterial production of camel nanobodies aimed at detecting environmental
biomarkers. Proteomics 13, 2766-2775.
Homeida, A.M., El Amin, E.A., Adam, S.E., Mahmoud, M.M., 1981. Toxicity of
diminazene aceturate (Berenil) to camels. J Comp Pathol 91, 355-360.
Kandeel, M., Al-Taher, A., Nakashima, R., Sakaguchi, T., Kandeel, A., Nagaya, Y.,
13
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Kitamura, Y., Kitade, Y., 2014. Bioenergetics and Gene Silencing Approaches for
Unraveling Nucleotide Recognition by the Human EIF2C2/Ago2 PAZ Domain. PLoS
One 9, e94538.
Kandeel, M., Kitade, Y., 2013a. Computational analysis of siRNA recognition by the
Ago2 PAZ domain and identification of the determinants of RNA-induced gene
silencing. PLoS One 8, e57140.
Kandeel, M., Kitade, Y., 2013b. In silico molecular docking analysis of the human
Argonaute 2 PAZ domain reveals insights into RNA interference. Journal of
computer-aided molecular design 27, 605-614.
Kohler-Rollefson, I., 1993. About camel breeds: A reevaluation of current classification
systems. J Anim Breed Genet 110, 66-73.
Korish, A.A., 2014. The antidiabetic action of camel milk in experimental type 2
diabetes mellitus: an overview on the changes in incretin hormones, insulin resistance,
and inflammatory cytokines. Horm Metab Res 46, 404-411.
Miller, R.E., Boever, W.J., Junge, R.E., Thornburg, L.P., Raisbeck, M.F., 1990. Acute
monensin toxicosis in Stone sheep (Ovis dalli stonei), blesbok (Damaliscus dorcus
phillipsi), and a Bactrian camel (Camelus bactrianus). J Am Vet Med Assoc 196,
131-134.
Mousa, H.M., Elsheikh, H.A., 1992. Monensin poisoning in dromedary camels. Dtsch
Tierarztl Wochenschr 99, 464.
Nebbia, C., Ceppa, L., Dacasto, M., Carletti, M., Nachtmann, C., 1999. Oxidative
metabolism of monensin in rat liver microsomes and interactions with tiamulin and
other chemotherapeutic agents: evidence for the involvement of cytochrome P-450 3A
subfamily. Drug Metab Dispos 27, 1039-1044.
Nebbia, C., Ceppa, L., Dacasto, M., Nachtmann, C., Carletti, M., 2001. Oxidative
monensin metabolism and cytochrome P450 3A content and functions in liver
microsomes from horses, pigs, broiler chicks, cattle and rats. J Vet Pharmacol Ther 24,
399-403.
Unciti-Broceta, J.D., Del Castillo, T., Soriano, M., Magez, S., Garcia-Salcedo, J.A.,
2013. Novel therapy based on camelid nanobodies. Ther Deliv 4, 1321-1336.
Wu, H., Guang, X., Al-Fageeh, M.B., Cao, J., Pan, S., Zhou, H., Zhang, L., Abutarboush,
M.H., Xing, Y., Xie, Z., Alshanqeeti, A.S., Zhang, Y., Yao, Q., Al-Shomrani, B.M.,
Zhang, D., Li, J., Manee, M.M., Yang, Z., Yang, L., Liu, Y., Zhang, J., Altammami,
M.A., Wang, S., Yu, L., Zhang, W., Liu, S., Ba, L., Liu, C., Yang, X., Meng, F., Li, L.,
Li, E., Li, X., Wu, K., Zhang, S., Wang, J., Yin, Y., Yang, H., Al-Swailem, A.M., 2014.
Camelid genomes reveal evolution and adaptation to desert environments. Nat Commun
14
1
5, 5188.
2
3
4
5
Table 1: The docking results of docking human and camel CYP1A1 with alfa
naphthoflavone, monensin and salinomycin.
6
7
human
camel
Hydrogen
MolDock
Rerank
bond
score
score
272.29
0
-106.7
-91.5
monensin
670.87
-6.2
-185.5
-71.6
salinomycin
750.9
-11.5
-152.6
69.25
Alfa
naphthflavone
272.29
0
-100
-76.7
monensin
670.87
-10
-160
74.1
salinomycin
750.9
-3.7
-121
106.8
enzyme
Compounds
MW
Cytochrome
Alfa
naphthoflavone
1
Cytochrome
1
8
9
10
11
15
1
2
3
4
5
Table 2: The docking results of docking human and camel CYP2C with felodepine,
monensin and salinomycin.
6
Enzyme
human
camel
Cytochrome
2
Cytochrome
2
Hydrogen MolDock Rerank
bond
score
score
Compounds
MW
felodepine
384
0
-61.2
-17.5
monensin
670.87
-19.1
-134.7
.71.1
salinomycin
750.9
-8.2
-91.6
-91.6
felodepine
384
0
-62.4
-46.2
monensin
670.87
-7.1
-134.6
-27.9
salinomycin
750.9
-2.1
-91.4
61.2
7
8
9
10
11
16
1
2
3
4
Table 3: The docking results of docking human and camel CYP3A with ritonavir,
monensin and salinomycin.
5
Enzyme
human
camel
Cytochrome
3
Cytochrome
3
Hydrogen MolDock Rerank
bond
score
score
Compounds
MW
ritonavir
720
-1.5
-154.5
-92.1
monensin
670.87
-2.5
-119.3
-79.8
salinomycin
750.9
-2.3
-126.8
-89.7
ritonavir
720
0
-152.6
-23.4
monensin
670.87
-18.2
-115.5
-35.4
salinomycin
750.9
-1.1
-107
-81.5
6
7
8
9
10
17
1
2
3
Table 4. The parameters of rate obtained by BEAST analysis and Tracer program for
CYP1A1, CYP2C and CYP3A
4
Rate mean
Rate variance
Tree likelihood
Camel rate
Human rate
CYP1A1
0.94
0.5
-7.4 E-4
0.73
1.18
CYP2C
0.97
0.18
-3.1 E-4
1.09
1.62
CYP3A
0.98
0.17
-2.2 E-4
1.04
1.65
5
6
7
8
9
10
11
12
13
14
15
16
18
1
Figure legends
2
Fig. 1. Chemical structure of compounds used in docking studies
3
Fig. 2. Molecular models of camel CYP1A1 bound with alfanaphthoflavone (A), camel
4
CYP2C bound with felodepine (B) and camel CYP3A bound with ritonavir (C).
5
Fig. 3. Sequence alignment of human and camel CYP1A1, CYP2C and CYP3A. The
6
alignment, annotation, and figure was created by Geneious 7.1 software package.
7
Fig. 4. Phylogenetic tree of CYP1A1
8
Fig. 5. Phylogenetic tree of CYP2C
9
Fig. 6. Phylogenetic tree of CYP3A
10
11
12
13
19