IJSRK International Journal of Scientif ic Research in Knowledge
www.i jsrpub.com
Feb 2014
Volume 2, Issue 2
Pages 57 – 115
Table of Contents
Article Author(s) page
The Prediction of Disulphide Bonding in HIV and other lenti-viruses by Machine Learning Techniques
Anubha Dubey 57
Bioinformatics Prediction of Interaction Silver Nanoparticles on the Disulfide Bonds of HIV-1 Gp120 Protein
Shahin Gavanji, Hassan Mohabatkar, Hojjat Baghshahi and Ali Zarrabi
67
A Comparative Evaluation of Drug Release and Permeability of Ethylcellulose, Cellulose Acetate and Eudragit RS100 Microspheres
Prakash Katakam, Saousen R. Diaf, Baishakhi Dey, Shanta K. Adiki, Babu R. Chandu and K.P.R. Chowdary
75
Investigation of a Proposed Four Storey Building Sites Using Geophysical and Laboratory Engineering Testing Methods in Lagos, Nigeria
Oyedele Kayode Festus, Adeoti, Lukman, Oladele Sunday and Kamil Akintunde
83
Nutritional and Anti-Nutritional Composition of Bridelia Ferruginea Benth (Euphorbiaceae) Stem Bark Sample
Adesina Adeolu Jonathan, Akomolafe Seun Funmilola
92
A Study on the Relationship between Accounting Conservatism and Earnings Management in Teheran Stock Exchange Listed Companies
Abbas Ramezanzadeh Zeidi, Zabihollah Taheri and Ommolbanin Gholami Farahabadi
105
International Journal of Scientific Research in Knowledge, 2(2), pp. 57-66, 2014
Available online at http://www.ijsrpub.com/ijsrk
ISSN: 2322-4541; ©2014 IJSRPUB
http://dx.doi.org/10.12983/ijsrk-2014-p0057-0066
57
Full Length Research Paper
The Prediction of Disulphide Bonding in HIV and other lenti-viruses by Machine
Learning Techniques
Anubha Dubey
Research Scholar, Department of Bioinformatics, MANIT BHOPAL (M.P), INDIA
Email: [email protected]
Received 01 November 2013; Accepted 28 December 2013
Abstract. The introduction of disulphide bonds into proteins is an important mechanism by which they have evolved and are
evolving. Most protein disulphide bonds are motifs that stabilize the tertiary and quaternary protein structure. These bonds also
thought to assist protein folding by decreasing the entropy of the unfolded form. Amino acid cysteine plays a fundamental role
in formation of disulphide bonds. In the present study, proteomics of disulphide bonding in HIV is studied through a machine
learning model which has been developed to classify disulphide bonds from different species of lentiviruses like bovine
immunodeficiency virus (BIV), simian immunodeficiency virus (SIV), Feline immunodeficiency virus, murine infectious virus
(MIV) and equine infectious anaemia virus (EIV) and Human immunodeficiency virus (HIV). Phylogenetic relationship is also
studied by the prediction of disulphide bonding among these viruses. Hence by different algorithms of WEKA classifier J48
predicts better classification with an accuracy of 89.6104%.
Keywords: Disulphide bond, motifs, lentiviruses, Phylogenetic, WEKA.
I. INTRODUCTION
Disulfide bonds play an important role in the folding
and stability of some proteins, usually proteins
secreted to the extracellular medium (Savier and
Kaiser, 2002).Since most cellular compartments are
reducing environments; in general, disulfide bonds are
unstable in the cytosol, with some exceptions as noted
below, unless a sulfhydryl oxidase is present (Hatahet
et al., 2010).
Fig. 1: Cysteine is composed of two cysteines linked by a
disulfide bond (shown here in its neutral form)
Disulfide bonds in proteins are formed between the
thiol groups of cysteine residues. The other sulphur-
containing amino acid, methionine cannot form
disulfide bonds. A disulfide bond is typically denoted
by hyphenating the abbreviations for cysteine, e.g.,
when referring to Ribonuclease A the "Cys26-Cys84
disulfide bond", or the "26-84 disulfide bond", or most
simply as "C26-C84" (Ruoppolo et al., 2000). The
structure of a disulfide bond can be described by its
dihedral angle between the
atoms, which is usually
close to ±90°.
The disulfide bond stabilizes the folded form of a
protein in several ways:
1) It holds two portions of the protein together,
biasing the protein towards the folded topology. That
is, the disulfide bond destabilizes the unfolded form of
the protein by lowering its entropy.
2) The disulfide bond may form the nucleus of a
hydrophobic core of the folded protein, i.e., local
hydrophobic residues may condense around the
disulfide bond and onto each other through
hydrophobic interactions.
3) Related to #1 and #2, the disulfide bond link
two segments of the protein chain, the disulfide bond
increases the effective local concentration of protein
residues and lowers the effective local concentration
of water molecules. Since water molecules attack
amide-amide hydrogen bonds and break up secondary
structure, a disulfide bond stabilizes secondary
structure in its vicinity (Thorton, 1981; Wetzel, 1987).
Dubey
The Prediction of Disulphide Bonding in HIV and other lenti-viruses by Machine Learning Techniques
58
For the protein folding prediction, a correct
prediction of disulfide bridges can greatly reduce the
search space (Skolnick et al., 1997; Huang, 1999).
The prediction of disulfide bonding pattern helps, to a
certain degree, predict the 3D structure of a protein
and hence its function because disulfide bonds impose
geometrical constraints on the protein backbones.
Some recent research works had shown the close
relation between the disulfide bonding patterns and
the protein structures (Chuang, 2003; Vlijmen, 2004).
By stabilizing protein structure, disulphide bond can
protect proteins from damage and their half-life. Once
the disulphide bond is formed they remain unchanged
for the life of the protein.
In the realm of the disulfide bond prediction, four
problems are addressed. The first is the protein chain
classification: to classify if the protein contains
disulfide bridge(s) or not, the second is the residue
classification: to predict the bonding state of
cysteines, the third is the bridge classification and the
last is the prediction of the disulfide bonding pattern.
Over the past years, significant progress has been
made on the prediction of the disulfide bonding states
(Fariselli, 1999; Fiser and Simon, 2000; Martelli,
2002; Chen, 2004)) and the disulfide bonding pattern
(Vullo, 2004; Ceroni, 2006; Song, 2007; Rubinstein,R
2008). For disulfide bonding pattern prediction, with
the exception of the methods proposed by (Ferre,
Clote 2005, 2006; Chen et al., 2006) others are also
used with or without bonding state known.
A method for predicting disulphide bonds from
genomic data which organisms are rich in disulfide
bonds has been described in (Mallick, 2002;
O'Connor, 2004). In the present study, a similar
strategy was utilized in which proteomic sequences
are used first to generate phylogenetic relation
between HIV and other species of lentivirus and then
disulphide bond prediction is done among the species
to see the disulphide richness across the species. HIV
(Human Immunodeficiency Virus) is a member of
genus lentivirus, part of the family retroviridae.
Lentiviruses have many common morphologies and
biological properties. Many species are infected by
lentiviruses, which are characteristically responsible
for long-duration illnesses with a long incubation
period lentiviruses are transmitted as single-stranded,
positive-sense, enveloped RNA viruses. Here in this
paper we have introduced a disulphide bonding
relationship of HIV with Other six species of
Lentivirus like bovine immunodeficiency virus (BIV),
simian immunodeficiency virus (SIV), Feline
immunodeficiency virus, murine infectious virus
(MIV) and equine infectious anaemia virus (EIV) and
two types of HIV- HIV1 & HIV2.
Table1: The comparative features of HIV with other related viruses
S.N
o.
Featur
es
FIV BIV MLV EIAV SIV HIV
1. Occur Cats Cattles Cancer in
mouse
Horse African
green
monkey
Human
2. Genom
e size
80-100 nm
and
pleomorphic,d
iploid
genome
Mature
virus,110-
130 nm
with 8.4 kb
90 nm in
diameter
- - 120 nm
3. Enzym
es
RTase,integras
e, protease
RTase,integ
rase,
protease
RTase,integ
rase,
protease
RTase,integ
rase,
protease
RTase,integ
rase,
protease
RTase,integrase,protease,ri
bonuclease
4. Structu
ral
genes
Gag,pol,env Gag,pol,env Gag,pol,env Gag,pol,env Gag,pol,env Gag,pol,env
5. Open
reading
frames
absent Regions
between pol
& env
absent absent present Present
6. Access
ory
genes
Vif,vpr,rev Nif,tat,rev absent Tat, vif - Vif,vpr,nef,vpu,vpx
(HIV2),tat,rev,tev(fusion of
tat,rev,and env)
7. Envelo
pe and
core
Env codes for
surface
glycoprotein
and
transmembran
e glycoprotein
Envelope
present and
core
contains
gag,gag-pol
polyprotein
Gag,gag-
pol poly
protein
Gag poly
protein
Gag,pol,
polyprotein
Gag-pol polyprotein
8. Conser
ved
RNA
absent absent Present
called core-
encapsidati
on signal
absent Present
called core
encapsidati
on signal
Present in SR proteins,
RNA interface etc.
International Journal of Scientific Research in Knowledge, 2(2), pp. 57-66, 2014
59
One of the most important contributions of
biological sequences to evolutionary analysis is that
the sequences of different organisms are often related.
Hence role of disulphide bond is studied among
lentivirus species.
2. MATERIALS and METHODS
2.1. Data Preparation
The analysis has been done on the basis of protein
sequence data of BIV, FIV, EIAV, MLV, SIV, and
HIV which has been obtained from UNIPROT [30].
2.1.1 Phylogenetic methods
To study evolutionary relationship it is important to
do multiple sequence alignment (MSA). MSA is a
sequence alignment of three or more biological
sequences, generally protein, DNA, or RNA. In many
cases, the input set of query sequences are assumed to
have an evolutionary relationship by which they share
a lineage and are descended from a common ancestor.
Of the various software’s of MSA, CLUSTAL W2
(Chenna, 2003) is found to be suitable. In this
neighbour-joining method is used. Neighbour-joining
(NJ) is a bottom-up clustering method used for the
construction of phylogenetic trees. Usually used for
trees based on DNA or protein sequence data, the
algorithm requires knowledge of the distance between
each pair of taxa (e.g., species or sequences) in the
tree.
Phylogenetic methods play an important role in
evolutionary analysis and to obtain the evolutionary
relationship of HIV. The sequences taken as
S1,S2,S3,S4,S5,S6,S7,S8 represents
HIV1,HIV2,MLV,BIV, Lentivirus,Murine virus,
Feline immunodeficiency virus, Equine infectious
anaemia virus, Simian immunodeficiency virus.
Following figures are obtained by CLUSTAL-W2.
FFiigg.. 22:: ((aa)) NNeeiigghhbboouurr JJooiinniinngg UUnnrrooootteedd ttrreeee FFiigg.. 22:: ((bb)) NNeeiigghhbboouurr JJooiinniinngg RRooootteedd ttrreeee
Fig. 3: (a) Dendogram Unrooted tree Fig. 3: (b) Dendogram Rooted tree
Here the cladogram and NJ tree shows the HIV 1
& 2 is related with Simian Immunodeficiency Virus.
The alignment of a query sequence to a
homologous sequence infers a likely three-
dimensional mapping of the protein sequence in
question, yielding homology-based structural
predictions for many proteins. Considering all such
protein sequences from a given genome as a group,
Dubey
The Prediction of Disulphide Bonding in HIV and other lenti-viruses by Machine Learning Techniques
60
the tendency of each amino acid type to appear in
spatial proximity to every other type was then
analyzed, taking into account the overall abundances
of the 20 amino acid types. Enrichment in cysteine–
cysteine proximity above the expected value was
taken to indicate an enrichment of disulfide bonding.
Since cysteine–cysteine proximity can also indicate
metal-binding motifs, proteins were first filtered to
remove proteins with metal-binding sites that would
otherwise produce false-positive results.
2.1.2. Disulphide Bond Prediction
Of the various software’s available for disulphide
bond prediction, Disulphide Bonding Connectivity
Pattern (DBCP) server (Hsuan-Hung and Lin-Yu,
2010) is used as it predicts better disulphide bonding
positions with cysteines positions and also capable to
relate disulphide bonding with metal binding sites.
The working of server is as follows:
(1) Run Basic Local Alignment Search Tool
(BLAST) to get the template sequence of the input
sequence. The parameters of BLAST are set as
follows: the Expectation value (E) threshold for
saving hits is set to a very large value 10 000 and the
database is set to PDB that contains sequences derived
from the 3D structure records from the Protein Data
Bank. If the E-value of the template sequence is >10
or the template sequence shares identity <25% to the
input sequence, instead of going to Step 2, the method
previously developed by (20) to predict the disulfide
bonding pattern is used.
(2) Align the input sequence and the template
sequence.
(3) Feed the alignment file into MODELLER and
run the procedure to evaluate the model of the input
sequence using the template sequence.
(4) get the coordinate (X, Y, Z) of the Ca (a
Carbon) of each residue.
(5) Coding each cysteine pair as the NPD
(normalized pair distance), this will be the input to the
SVM (Hsuan-Hung and Lin-Yu, 2010).
(6) Feed the coding file into the Support Vector
Machine (SVM) to predict the bonding probability of
each cysteine pair with the trained model. The
multiple trajectory searches (Tseng and Chen, 2008)
are tightly integrated with the SVM training. For more
details, please refer to the Supplementary Data on the
DBCP web server.
(7) Coding the input file with the probabilities
from the SVM output and using the modified
weighted perfect matching algorithm to get the first
level disulfide bonding connectivity.
(8) Justify the first level disulfide bonding
connectivity with the thresholds to get the final result.
(9) Display the result on the web page or send the
result to the user. In Step 1, if the E-value of the
template sequence is >10 or the template sequence
shares identity <25% to the input sequence, a
previously proposed method (Lin, H.H., and Tseng,
L.Y. 2009) is used for prediction. In this method, the
position specific scoring matrix, the normalized bond
lengths, the predicted secondary structure of protein
and the physicochemical properties index of the amino
acid were used as features. The multiple trajectory
searches and the SVM training were tightly integrated
to train the predictor. More details can be obtained
from Lin and Tseng, 2009).
The DBCP server is free and open to all users.
2.1.2.1. Evaluation
After taking four websites of disulphide bond
connectivity pattern without prior knowledge of
bonding state of cysteine (Ferre et al., 2006; Song,
2007). We have tested our prediction by 10-fold cross
validation on the data set of FIV, BIV, MLV, EIAV,
SIV, HIV jointly named as VIRUS. And disulphide
bonds were observed with cysteine residues and some
of them are also shows metal binding sites. This was
again evaluated/ classified by J48 WEKA 3.7
algorithm. J48 is a decision tree classifier (Pfahringer
IHW, 1999).
The number of ways of forming p disulfide bonds
from n cysteine residues is given by the formula
2.1.2.2. Measurement of accuracy
A necessary step to the prediction of disulphide
connectivity is the prediction of the disulphide
bonding state of cysteines in proteins. In order to
evaluate the accuracy of the prediction two indexes
can be used: Qp and Qc.
For a protein PQp is defined as:
Qp= δ (Correct pattern, predicted pattern)
(1)
Where δ(x,y) is 1 if and only if the predicted
pattern coincides with the correct pattern.
Alternatively, Qc is defined as:
C
numberofcorrectlypredictedpairsQ
numberofpossiblepairs (2)
International Journal of Scientific Research in Knowledge, 2(2), pp. 57-66, 2014
61
The two indexes are equally suited and
complimentary for measuring the accuracy of the
prediction: Qp is a measure of the predictive
performance on each protein (either 1 or 0) and can be
averaged over a number of predicted proteins to give a
global measure of the accuracy of the method. Qc
quantifies the accuracy of the method based on the
number of pairs correctly predicted with respect to the
total number of possible pairs.
In order to score the method its performance was
also compared with that of a random predictor. The
probability of a predictor randomly performing (Rp)
on the prediction of the connectivity pattern can be
computed. In general, given 2B cysteines, the number
of possible connectivity pattern is:
Np= (2B-1)
( )(2 1) (2 1)i B iNp B (3)
The corresponding probability of Rp is:
1( )Qp Rp
Np (4)
For the random predictor (Rp), Qc is
1( )
(2 1)c pQ R
B
(5)
Evaluation of predictive accuracy:
The prediction accuracy was calculated by following
the standard conventions accuracy for prediction:
2c
o
NQ
N (6)
Where Nc is the total number of correctly predicted
cysteines and No is the total number of cysteines.
Specificity of the prediction:
X
X X
TNSpecificity
TN FP
(7)
Where x denotes the bonded cysteines or non-
bonded cysteines XFP is the number of false
negatives in the prediction and XTP is the number of
true positive predictions for bonding state x.
Sensitivity of the prediction was calculated as:
X
X X
TPSensitivity
TP FN
(8)
Where XFN is the number of false negatives for
bonding state x. The Matthews correlation
coefficient:
MCC is calculated as:
Mathews Correlation coefficient( ) ( )
( )( )( )( )
TP TN FP FNMCC
TP FN TP FP TN FP TN FN
Where XTN the true negatives of bonding state are
X. The value of MCC is an indication of how good is
the prediction. The closer the MCC is to 1, the closer
the prediction is to a perfect prediction.
2.1.3. Prediction of oxidation state of cysteines
Knowledge of the oxidation state of cysteines infer a
lot of information about the protein such as local
sequence environment the possible 3D structure of
protein and in some cases, the function and working
mechanisms of the protein. In this paper position of
oxidized cysteines was observed by DBCP software
(Hsuan-Hung and Lin-Yu, 2010).
2.1.4. Prediction of connectivity pattern of
cysteines
Connectivity pattern prediction is a challenging and
yet very biological meaningful task. It is challenging
because there are too many possibilities of disulphide
bonding for a given protein and many factor influence
the final connection pattern. The correct predictions of
disulphide bond provide in order to have a stabilized
three dimensional protein structure. Research is going
on for prediction of connectivity pattern of cysteines
(Hsuan-Hung and Lin-Yu, 2010)
2.1.5. Prediction of number of disulphide bridges
Analysis of prediction results shows that there is a
relationship between the sum S(p) of all the
probabilities of cysteines and the total number of
bonded cysteines (as predicted by DBCP serwer). The
Dubey
The Prediction of Disulphide Bonding in HIV and other lenti-viruses by Machine Learning Techniques
62
total number of bonded cysteines using linear
regression approach shows that the total number of
bonded cysteines is even and does not exceed the total
number of cysteines in sequence. When the number of
disulphide bridges increases in chains, the
performance decreases in general. The overall
specificity and sensitivity using four different input
schemes are around 51% to 55%. The variations of the
performance for chains with many disulphide bridges
(K>6) is large because there is few in the dataset.
Thus for proteins with a large number of disulphide
bridge (K>6), prediction must be used as caution. It is
very difficult to correctly predict the entire disulphide
connectivity pattern because the number of
connectivity pattern increases exponentially with K.
2.2. J48 algorithm
The disulphide bond prediction will be classified by
machine learning algorithm. J48 proves better for
small biological data. J48 is a decision tree
classifier.J48 is a Machine learning algorithm, a
branch of artificial intelligence, is a scientific
discipline concerned with the design and development
of algorithms that allow computers to evolve
behaviours based on empirical data, such as from
sensor data or databases. A learner can take advantage
of examples (data) to capture characteristics of interest
of their unknown underlying probability distribution.
Data can be seen as examples that illustrate relations
between observed variables. A major focus of
machine learning research is to automatically learn to
recognize complex patterns and make intelligent
decisions based on data; the difficulty lies in the fact
that the set of all possible behaviours given all
possible inputs is too large to be covered by the set of
observed examples (training data).
3. RESULTS AND DISCUSSIONS
Phylogenetic methods play an important role in
evolutionary analysis and to obtain the evolutionary
relationship of HIV with other species of lentivirus i.e.
MLV, BIV, Lentivirus, Murine virus, Feline
immunodeficiency virus, Equine infectious anaemia
virus, Simian immunodeficiency virus. CLUSTAL-
W2 result shows the cladogram and NJ tree of the
species which presents that HIV 1 & 2 is related with
Simian Immunodeficiency Virus. Disulphide bond is
predicted for these sequences to find out the similarity
among lentivirus species and then disulphide bond
based classification is studied by J48 a machine
learning technique.
3.1. Analysis of DBCP server
A web-based application system called the DBCP is
provided for the prediction of the disulfide bonding
connectivity pattern without the prior knowledge of
the bonding state of cysteines. To the best of our
knowledge, the best accuracy of disulfide connectivity
pattern prediction (Qp) and that of disulfide bridge
prediction (Qc) are found 81% and 82%, respectively,
on the data set of HIV and other related to HIV
molecular sequences with 10-fold cross validation.
Env gene plays a significant role in disulphide bond
prediction. Env, gag in FIV, env- pol in MLV, pol in
EIA, Env in SIV & HIV correctly predicts disulphide
bonds. Table 2 shows the species with position of
disulphide bond and this proves that disulphide bond
is conserved among species.
Table 2: Species with position of disulphide bonds
Species gene Position of disulphide bonds
FIV Env & gag 328-348
EIA POL 322-342
MLV ENV & POL 81-95,112-129,121-134,165-184: 536-538,561-576,
SIV ENV 99-207,106-198,180-193,230-242,300-333,382-457,389-
430,412-422
HIV ENV 118-200,125-191,213-242,223-234,291-328,374-435,381-
408
Stabilization of the native Env complex by disulfide
bond linkage is likely to impose constraints on Env
function because a certain degree of flexibility is
probably essential for Env to undergo the
conformational changes that eventually lead to fusion
of the viral and cellular membranes. The gp120 –
gp41 interface is considered to be structurally flexible,
so constraining its motion might have adverse effects.
3.2. J48 based classification
Again disulphide bond based classification of HIV
and other related viruses is done by machine learning
J48 algorithm which gives the accuracy of 89.6104%.
After 10 fold cross validation of training data (Virus)
the result obtained is shown as follows:
In the field of machine learning, a confusion
matrix is a specific table layout that allows
visualization of the performance of an algorithm,
International Journal of Scientific Research in Knowledge, 2(2), pp. 57-66, 2014
63
typically a supervised learning one (in unsupervised
learning it is usually called a matching matrix). Each
column of the matrix represents the instances in a
predicted class, while each row represents the
instances in an actual class.
Table 3: Statistics of J48 algorithm
Correctly Classified Instances 69 89.6104 %
Incorrectly Classified Instances 8 10.3896 %
Kappa statistic 0.8359
Mean absolute error 0.0528
Root mean squared error 0.1845
Relative absolute error 56.0893 %
Root relative squared error 56.0893 %
Total Number of Instances 77
Table 4: Detailed accuracy by class
TP
RATE
FPRATE PRECISION RECALL F-MEASURE ROC CLASS
0 0 0 0 0 0.474 EIA
0 0 0 0 0 0.48 BIV
0.833 0.042 0.625 0.833 0.714 0.893 FIV
1 0.032 0.875 1 0.933 0.99 MLV
0.714 0 1 0.714 0.833 0.814 SIV
0.976 0.083 0.93 0.976 0.952 0.918 HIV
0.896 0.053 0.885 0.896 0.884 0.899 Weighted
average
Table 5: Confusion Matrix between predicted and actual class Predicted class (column)
a b c d e f Classified as
0 0 1 0 0 0 a=EIA
0 0 1 0 0 0 b=BIV
0 0 5 1 0 0 c=FIV
0 0 0 14 0 0 d=MLV
0 0 0 1 10 3 e=SIV
0 0 1 0 0 40 f=HIV
The table 4 shows precision and recall which
actually are two widely used metrics for evaluating
the correctness of a pattern recognition algorithm.
They can be seen as extended versions of accuracy, a
simple metric that computes the fraction of instances
for which the correct result is returned. In a
classification task, the precision for a class is the
number of true positives (i.e. the number of items
correctly labeled as belonging to the positive class)
divided by the total number of elements labeled as
belonging to the positive class (i.e. the sum of true
positives and false positives, which are items
incorrectly labeled as belonging to the class). Recall in
this context is defined as the number of true positives
divided by the total number of elements that actually
belong to the positive class (i.e. the sum of true
positives and false negatives, which are items which
were not labeled as belonging to the positive class but
should have been).
Fig. 4: ROC for J48 Classifier
Dubey
The Prediction of Disulphide Bonding in HIV and other lenti-viruses by Machine Learning Techniques
64
3.3. Reciever Operating Curve (ROC)
It is a graphical technique for evaluating data mining
schemes, which are used in such a way the learner is
trying to select samples of test instances that have a
high proportion of positives a term used to
characterize the tradeoff between hit rate and false
rate.ROC curves depicts the performance of a
classifier without regard to class distribution or error
costs. They plot the number of positives included in
the samples on the vertical axis, expressed as a
percentage of the total number of positives, against the
total number of negatives on the horizontal axis. For
each fold of a 10 fold cross validation ,weight the
instances for a selection of different cost ratios train
the scheme on each weighted set ,count the true
positives and false positives in the test set, and plot the
resulting point on the ROC axes.
The correlation between MLV, SIV and HIV is
predicted as their data is sufficient for analysis of
correlation between these species, but the data of EIV,
FIV, BIV are less for any analysis. Let MLV is X3,
SIV is X2 and HIV is X1, then from correlation
coefficient r12= 0.98551, r13=0.30182, r23=0.47188.
This correctly shows that HIV is highly related to
SIV. Hence Pearson correlation coefficient has been
widely for the analysis of proteomic data. Its
popularity is likely due to its simplicity and
interpretability; therefore it essentially computes the
strength of the linear relationship between the two
quantities/ species.
4. CONCLUSION
Here a framework for disulphide bond prediction and
classification is presented with an accuracy of
89.6104%. Furthermore, DBCP is better for prediction
of disulphide bond with cysteines positions and also
this web server is able to find metal binding sites.
Other methods that can predict both the disulfide
bonds and the metal binding sites will be more
suitable for prediction. The high metal binding site
score (e.g. >0.5) indicates that there may be cysteines
involved in the metal binding sites. For protein
sequence analysis it was found that Env envelope
glycoprotein shows disulphide bond conservation
among all the species of retroviruses. The correlation
between HIV and SIV was also found to be 0.98551.
Hence disulphide bonds are evolutionary conserved
throughout the species.
The knowledge of disulfide richness in certain
organisms suggests practical applications, including
engineering enhanced protein stability and facilitating
protein-fold recognition. Disulfide-rich organisms
should allow the development of novel tools and
approaches for attacking such problems of current
interest. This work depends upon the availability of
sequenced proteomes, and the availability of
additional other proteomes has enabled the
identification of an enigmatic protein family as a
potential player in the biochemistry of cytoplasmic
disulfide bonds. We hope this study will promote
continued interest in sequencing more proteins from
diverse organisms so as to further enhance the scope
and resolution of comparative proteomics techniques.
As more proteomes become available, we anticipate
that the ease of discovery of specific proteomic
adaptations to the environment will improve and yield
further insights into molecular evolution and cell
biology.
REFERENCES
Sevier CS, Kaiser CA (2002). Formation and transfer
of disulphide bonds in living cells. Nature
Reviews Molecular and Cellular Biology,
3(11): 836–847.
Hatahet F, Nguyen VD, Salo KEH, Ruddock LW
(2010). Disruption of reducing pathways is not
essential for efficient disulfide bond formation
in the cytoplasm of E. coli. MCF, 9(67): 67.
Ruoppolo M, Vinci F, Klink TA, Raines RT, Marino
G (2000). Contribution of individual disulfide
bonds to the oxidative folding of rib
ribonuclease A. Biochemistry, 39(39): 12033–
42.
Thorton JM (1981). Disulphide bridges in globular
proteins. J. Mol.Biol.,151: 261-287.
Wetzel R (1987). Harnessing disulphide bonds using
protein engineering. Trends Biochem Sci., 12:
478-482.
Skolnick J, Kolinski A, Ortiz AR (1997).
MONSSTER: a method for folding globular
proteins with a small number of distance
restraints. J. Mol. Biol., 265: 217–241.
Huang ES, Samudrala R, Ponder JW (1999). Ab initio
fold prediction of small helical proteins using
distance geometry and knowledge-based
scoring functions. J. Mol. Biol., 290: 267–281.
Chuang CC, Chen CY, Yang JM, Lyu PC, Hwang JK
(2003). Relationship between protein structures
and disulfide-bonding patterns. Proteins, 55: 1–
5.
Van Vlijmen HWT, Gupta A, Narasimhan LS, Singh J
(2004). A novel database of disulfide patterns
and its application to the discovery of distantly
related homologs. J. Mol. Biol., 335: 1083–
1092.
Fariselli P, Riccobelli P, Casadio R (1999). Role of
evolutionary information in predicting the
disulfide-bonding state of cysteine in proteins.
Proteins, 36: 340–346.
International Journal of Scientific Research in Knowledge, 2(2), pp. 57-66, 2014
65
Fiser A, Simon I (2000). Predicting the oxidation state
of cysteines by multiple sequence alignment.
Bioinformatics, 16: 251–256.
Martelli PL, Fariselli P, Malaguti L, Casadio R
(2002). Prediction of the disulfide-bonding state
of cysteines in proteins with hidden neural
networks. Protein Eng., 15: 951–953.
Chen YC, Lin YS, Lin CJ, Hwang JK (2004).
Prediction of the bonding states of cysteines
using the support vector machines based on
multiple feature vectors and cysteine state
sequences. Proteins, 55: 1036–1042.
Fariselli P, Casadio R (2001). Prediction of disulfide
connectivity in proteins. Bioinformatics, 17:
957–964.
Vullo A, Frasconi P (2004). Disulfide connectivity
prediction using recursive neural networks and
evolutionary information. Bioinformatics, 20:
653–659.
Ferre`F, Clote P (2005). Disulfide connectivity
prediction using secondary structure
information and diresidue frequencies.
Bioinformatics, 21: 2336–2346.
Ferre F, Clote P (2006). DiANNA 1.1: An extension
of the DiANNA web server for ternary cysteine
classification.Nucleic Acids Res., 34: W182–
W185.
Chen BJ, Tsai CH, Chan CK, Kao CY (2006).
Disulfide connectivity prediction with 70%
accuracy using two-level models. Proteins, 64:
246–252.
Ceroni A, Passerini A, Vullo A, Frasconi P (2006).
DISULFIND: a Disulfide Bonding State and
Cysteine Connectivity Prediction Server.
Nucleic Acids Res., 34: W177–W181.
Cheng J, Saigo H, Baldi P (2006). Large-scale
prediction of disulphide bridges using kernel
methods, two-dimensional recursive neural
networks, and weighted graph matching.
Proteins, 62: 617–629.
Song J, Yuan Z, Tan H, Huber T, Burrage K (2007).
Predicting disulfide connectivity from protein
sequence using multiple sequence feature
vectors and secondary structure.
Bioinformatics, 23: 3147–3154.
Rubinstein R, Fiser A (2008). Predicting disulfide
bond connectivity in proteins by correlated
mutations analysis. Bioinformatics, 24: 498–
504.
Mallick P, Boutz DR, Eisenberg D, Yeates TO (2002).
Genomic evidence that the intracellular proteins
of archaeal microbes contain disulfide bonds.
Proc Natl Acad Sci USA, 99: 9679–9684.
O'Connor BD, Yeates TO (2004). GDAP: A web tool
for genome-wide protein disulfide bond
prediction. Nucleic Acids Res, 32:W360–
W364.
Poumbourios P, Maerz AL, Drummer HE (2003).
Functional evolution of the HIV-1 envelope
glycoprotein gp120-association site of gp41. J
Biol Chem., 278: 42149-42160.
Hsuan-Hung L, Lin-Yu T (2010). DBCP: A web
server for disulfide bonding, Connectivity
pattern prediction without the prior knowledge
of the bonding state of cysteines. Nucleic Acids
Research, Vol. 38, Web Server issue W503–
W507.
Pfahringer IHW (1999). WEKA: A Machine Learning
Workbench for Data, www.cs.waikato.ac.nz.
Lin HH, Tseng LY (2009). Predicting of disulphide
bonding pattern based on support vector
Machines with parameters tuned by multiple
trajectory search. WSEAS Trans. Compu., 9:
1429-1439.
Tseng LY, Chen C (2008). Multiple trajectories search
for large scale global optimization. Proceedings
of 2008 IEEE congress on Evolutionary
Computation, CEC’08, Hong-Kong, 3052-
3059.
Chenna R, Sugawara H, Koike T, Lopez R, Gibson
TJ, Higgins DG, Thompson JD (2003).
Multiple sequence alignment with the Clustal
series of programs Nucleic Acids Res., 31:
3497-3500.
Dubey
The Prediction of Disulphide Bonding in HIV and other lenti-viruses by Machine Learning Techniques
66
Anubha Dubey has submitted her PhD in Bioinformatics at Maulana Azad National Institute of
Technology, Bhopal. She received her first degree in Rani Durgawati University Jabalpur in 2005
awarded with Bachelors of Science in Biotechnology. She obtained degree in Master of Science in
Biotechnology from Barkatullah University Bhopal in 2007 with dissertation An Approach to
Investigate the Phenomenon of Genomic Instability in Cultured Human Foetal Lung Fibroblast cells by
modern Technologies. Her current research is focus on extracting information from HIV molecular
sequences by Machine learning techniques. To date, she has published several scientific articles related
to machine learning field.
International Journal of Scientific Research in Knowledge, 2(2), pp. 67-74, 2014
Available online at http://www.ijsrpub.com/ijsrk
ISSN: 2322-4541; ©2014 IJSRPUB
http://dx.doi.org/10.12983/ijsrk-2014-p0067-0074
67
Full Length Research Paper
Bioinformatics Prediction of Interaction Silver Nanoparticles on the Disulfide Bonds
of HIV-1 Gp120 Protein
Shahin Gavanji1*
, Hassan Mohabatkar2*
,Hojjat Baghshahi3, Ali Zarrabi
2
1Young Researchers and Elite Club, Isfahan (Khorasgan) Branch, Islamic Azad University, Isfahan, Iran
2Department of Biotechnology, Faculty of Advanced Sciences and Technologies, University of Isfahan, Isfahan, Iran
3Department of Animal Sciences, College of Agriculture, Isfahan University of Technology, Isfahan, Iran
*Corresponding Author: [email protected], [email protected]
Received 25 November 2013; Accepted 01 January 2014
Abstract. Silver nanoparticles have anti-HIV features in early stages of virus amplification. The two existing disulfide bonds
in carboxyl half of the HIV-1 GP120, which cooperate in conjugation of CD4 receptor, interact with silver nanoparticles.
Studies on protein disulfide bonds were done using Metal Detector Predicts V2.0 software. All Cys and His residues in amino
acid sequences were identified. Protein denaturation decreases disulfide bonds when silver ions couple with sulfhydryl groups.
CTRPNNNTRKRIRIQRGPGRAFVTIGKIGNMRQAHC amino acid sequence in GP 120 plays a key role. Breaking this bond
can alter spatial structure of the protein and prevents this part from connecting to CD4. Ultimately, nanosilver can prevent HIV
from connecting to CD4.
Key words: Metal binding sites, silver, GP 120, Bioinformatics
1. INTRODUCTION
Nanoparticles are defined as structures with a
dimension in the range of 1–100 nm. The widely
technological use of nanoparticles in commercial
industry will cause a great increase in price by 2015
(up to $3 trillion) (Ahamed et al., 2010). Silver is
often found in form of metal silver nanoparticles.
Since the particles are small in size, the exposure of
their surface in solution peaks, which leads to the
maximum possible effect per unit of silver. All the
interacting silver nanoparticles with the bacteria were
between 1 and 10 nm (Morones et al., 2005). This
dependency in size probably exists because the
combination of these particles can pass and react with
the cell membrane and their surface-to-volume ratio is
higher.
The smaller is the size of the silver nanoparticles,
the higher is the interaction between atoms. This
explains the reason for small (1−10nm) silver
nanoparticles’ interaction with the bacteria (Feng et
al., 2000). The existence of silver in this minute
amount doesn’t have any adverse effect on human
body (Berger et al., 1976). Nanoparticles have been
applied in medicine in order for experts to trace,
diagnose and cure different diseases. However, the
biological features of these particles are yet to be
studied (Bhattacharya and Mukherjee, 2008). In the
nineteenth century scientists discovered the use of
silver in medicine and local antibacterial agents
(Tokumaru et al., 1974). Studies run on anti-microbial
potential of silver nanoparticles have shown that these
nanoparticles are antibacterial agents against Gram-
negative and Gram positive bacteria (Shahverdi et al.,
2007). And are antiviral agents against HIV-1 (Sun et
al., 2005), hepatitis B virus (Lu et al., 2008),
respiratory syncytial virus (Sun et al., 2008), herpes
simplex virus type 1 (Baram-Pinto et al., 2009) and
monkey pox virus (Rogers et al., 2008). Many
biological processes in micro organisms can be
attacked by silver, namely the alteration of cell
membrane structure and functions (Pal et al., 2007).
This fact makes the application of silver very useful in
developing many biological and pharmaceutical
processes, products and devices some of which are
coating materials for medical devices (Raad and
Hanna, 2002), orthopedic or dental graft materials
(Hotta et al., 1998), wound repair topical aids
(Dowsett, 2004), water sanitization (Lin et al., 2002),
textile products (Takai et al., 2002) and washing
machines (Jung et al 2007). Everyday new forms of
silver nanoparticles are being produced. Parts of
clothing, food containers, wound dressings, ointments,
implant coating ant etc. are some of these products
Gavanji et al.
Bioinformatics Prediction of Interaction Silver Nanoparticles on the Disulfide Bonds of HIV-1 Gp120 Protein
68
(Arora et al., 2008; Kumar et al., 2008). Silver modes
of inhibitory action to microorganisms are
various(Clement and Jarrett, 1994), therefore,
compared to synthetic fungicides, the use of it in
controlling different forms of plant pathogens might
be safer (Choi, 2006).
This effect seems to be caused by those
mechanisms that are the cause of bactericidal effect of
silver ions. There are two groups of bactericidal
effects of silver: silver ions or silver nanoparticles.
Ions and particles have to be distinctively identified.
Silver ions are charged atoms (Ag+) while silver
nanoparticles are single crystals of nanosize
dimensions. Structural changes in the cell membrane
have been proven to be made by silver ions. The
enzyme-containing sulfate in the membrane of the
bacteria is in interaction with silver ions and. This
changes the membrane morphology by pacifying the
enzymes. Silver ions can easily penetrate the
membrane because of the mentioned inactivation
which makes the membrane vulnerable. The
interaction between silver ions and sulfate groups
located in the active site of the enzyme inside the cell
causes the destruction of the different parts of the cell
to continue. An inactivation of enzymes happens as a
result of this interaction with the active site. Another
interaction, which is proved to have severe effects, is
the interaction between silver ions and phosphorus
groups. The interaction between silver ions and DNA
backbone is an example of this, which is the reason
for bacteria’s inability in replicating itself or
transcribing mRNA for new proteins. All above-
mentioned changes cause a decrease in the speed of
bacteria growth and eventually lead to its death
(Alcamo, 1997). It was shown by another study that,
on the surface of cellular membrane, monovalent
silver ions (Ag+) replace hydrogen cations (H
+) of
sulfhydryl or thiol (S-H) groups.
This disables the production of proteins required to
sustain the cell, which finally kills the cell (Feng et
al., 2000). In addition, studies have suggested that
when silver ions enter the cell they intercalate into
bacterial DNA, and that prevents more pathogen
proliferation. The effectiveness of silver particles as
antimicrobial agents has been increased by
nanotechnology in recent years. The contact between
silver nanoparticles and microbes increases because of
the larger area-to-volume ratio of silver nanoparticles,
which results in the increase of their ability to
permeate cells, and that prevents more pathogen
proliferation. It is thought that silver’s mode of action
depends on Ag+ ions. These ions prevent bacterial
growth greatly by suppression of respiratory enzymes
and electron transport components and through
interference with DNA functions (Li et al., 2006).
Life is believed to be sustained in bacteria through
using an enzyme to metabolize oxygen. This enzyme
gets crippled by silver ions therefore oxygen
metabolizing gets stopped which suffocates the
bacteria and kills it. Virus growth happens when they
invade living cells and then produce replicas of
themselves. Since cell’s life depends on oxygen
metabolizing enzymes, silver ions kill viruses by
killing the cell by suffocation (Alvarez-Puebla et al.,
2004). The interactions between silver nanoparticles
and the bacteria are quite similar to the interactions
between silver ions and bacteria. This might mean that
silver nanoparticles and silver ions are similarly able
to interact with the same types of chemical groups,
and therefore can damage the bacteria in the same
form (Morones et al., 2005). More than any other
inorganic antibacterial agent, the antimicrobial
property of silver has been investigated and employed
extensively (Russell et al., 1994).
The slow and continual release of silver that
prolonged the antimicrobial effect is supposed to be
reason for this. Ionic or nanoparticle silver can
potentially be used for controlling spore-producing
fungal plant pathogens because of their antifungal
activity. In comparison with synthetic fungicides,
silver might have less toxicity for humans and
animals. Antibiotic and preservative qualities of silver
nanoparticles bring them close to human beings.
Bacteria play a key role in human digestion, and if
nanoparticles could eliminate the bacteria, it would
have a fatal effect. Some recent studies have shown
that silver nanoparticles have the ability to attach to
HIV-1virus and avoid it from making bonds with cells
(Elechiguerra et al., 2005). World Health
Organization has currently listed silver sulfadiazine as
an essential anti-infective topical medicine (Lara et
al., 2010). Virus adsorption tests were used to prove
that anti HIV activity of silver nanoparticles prevents
this virus from binding or fusion to the cells. AIDS
happens because of HIV-1infection (Lara et al., 2010;
Allan et al., 1985).
Since the binding between HIV and the target cell
surface, and therefore the cellular and viral membrane
fusion, is mediated by the envelope, its role is critical
in infection. Because preventing HIV from entering
the target cell abandons viral infectivity, replication,
and the cytotoxicity caused by virus-cell interaction,
fusion or entry inhibitors should be remarkably noted.
Also the presence of virucidal agents is necessary for
HIV/AIDS prevention because of their deactivating
quality against viral particle (virion), with which the
completion of viral replication cycle can be prevented
(Salzwedel et al., 1999). Silver nanoparticles act like
virucidal agent or viral entry inhibitor and therefore
deploy anti-HIV activity at early stages of viral
replication. Remarkable differences in antiviral
International Journal of Scientific Research in Knowledge, 2(2), pp. 67-74, 2014
69
activities of silver nanoparticles against different
drug-resistant strains were not found; therefore
changes made in antiretroviral HIV strains that confer
resistance do not have any influence on the efficacy of
silver nanoparticles (Al-Jabri and Alenzi, 2009).
Furthermore, an interaction between silver
nanoparticles and the two disulfide bonds located in
the carboxyl half of the HIV-1 gp120 glycoprotein
might happen. This mentioned area has been shown in
binding to the CD4 receptor (Lekutis et al., 1992).
The binding between sulfhydryl groups causes the
protein denaturation through reducing disulfide bonds.
Because silver nanoparticles have antiviral activity,
they are able to prevent HIV-1 infection from binding
to gp120. This inhabitation prevents CD4-dependant
viron binding, fusion and infectivity and inhibits HIV-
1 cell-free and cell-associated infection. This happens
because they act as virucidal agents (Borkow and
Lapidot, 2005). It is concluded that, through stopping
the activity of HIV particles at an early stage of viral
replication, silver nanoparticles act as effective
virucides by inactivating HIV particles in a short
period of time.
The aim of the study was to investigate the
bioinformatics prediction of interaction silver
nanoparticles on the disulfide bonds of HIV-1 Gp120
protein
2. MATERIALS AND METHODS
2.1. Preparing 3 dimensional structure GP120
protein
In the first step, amino acid sequences of GP120
protein with an accession number of p03378.1 were
taken from NCBI website (www.ncbi.nlm.nih.gov).
The GP120 protein consists of 478 amino acids
(Figure 1) and its molecular weight is 23.469Da .Then
the GP120 viral protein with the number of 1GC1
linked to CD4 protein was taken from Protein Data
Bank website. The CD4 and additional ligands were
separated from that GP120 viral protein through
Molegro software and this protein was prepared for
research, without having any other protein or ligand
linked to it (Figure 2) (www.rcsb.com). In the next
step, and silver with Ag formula (number 22394) were
provided from ChemSpider website
(www.chemispider.com).
Fig. 1: Amino acid sequence of GP120 protein
Fig. 2: Structure of GP120 protein
2.2. Studying disulphide bonds in proteins
2.2.1. Metal Detector Web Server
The web server Metal Detector classifies histidine
residues in proteins into one of two states (free or
metal bond) and cysteines into one of three states
(free, metal bond or disulfide bridged). It is freely
available at (http://metaldetector.dsi.unifi.it/v2.0).
This web server takes the protein sequence as input
and outputs predictions of transition-metal binding for
cysteine and histidine residues; for cysteines it also
predicts disulfide bonding bridges. Residues predicted
to coordinate the same ion will share the same
identifier. Every identifier is an integer ranging from 1
to 4. Its value has no special biochemical semantics
Gavanji et al.
Bioinformatics Prediction of Interaction Silver Nanoparticles on the Disulfide Bonds of HIV-1 Gp120 Protein
70
but lower values correspond to a higher level of
confidence for the predictor.
2.2.2. Studying disulphide bonds
Studies on disulfide bonds were performed using
Metal Detector Predicts v2.0 software. All Cys and
His residues in amino acid sequences were identified.
(http://metaldetector.dsi.unifi.it).
Fig. 4: disulfide bonds and metal binding site in GP120 protein
3. RESULTS AND DISCUSSIONS
3.1. Protein Structure Analysis
The GP120 protein consists of 478 amino acids, and
its molecular weight is 23.469 kD. This location is in
extracellular form, and the location of CD4-binding
loop is from amino acids 366 to 376 which contain 11
amino acids. According to the results taken from
Metal Detector Predicts v2.0 online server, the
disulfide bonds in GP120 protein amino acid sequence
are sited between cysteine amino acids in locations
186, 173, 164, 125, 99, 94, 87, 42, 22, 413, 386, 353,
346, 299, 264, 215, 207 and 196 as they are shown in
figure 4. The 346 location is the most important site of
this sequence, which is the site for metal group bonds.
Therefore, the sequence of
CTRPNNNTRKRIRIQRGPGRAFVTIGKIGNMRQA
HC in GP120 protein plays a key role (Figure 5). This
sequence is in the first and most important part of this
protein to bind GP120 and CD4 proteins. In Molegro
software the basis of the three dimensional structure
of this protein is positioned between 264 to 299 amino
acids where sulfide bonds are located. Breaking this
binding can change the spatial structure of this
protein, therefore prevents this part from binding to
CD4.
International Journal of Scientific Research in Knowledge, 2(2), pp. 67-74, 2014
71
Fig. 5: The structure of sequence amino acid in GP120 protein
Penetration and distribution inhibitors are
considered remarkable since inhibiting HIV from
penetrating the target cell will in turn lead to
prevention from viral infections, duplication, and
cytotoxicity caused by virus-cell interaction (Borkow
and Lapidot, 2005). The antivirus activity of
nanosilver particles let us inhibit HIV1 infections
regardless of virus types or resistant profiles against
GP120 binding. This inhibition is done so that viral-
particle-dependant CD4 cannot bind and there will not
be any infection and propagation, and it prevents
HIV-1 free cells and infected cells. Thus, nanosilver
particles are effective antivirus particles since they are
short term HIV deactivators. These particles act at the
beginning stages of viral propagation (penetration or
spread) and at pre-penetration stages (Lara et al.,
2010). The binding of silver ions to sulfhydryl groups
causes protein denaturation through reducing disulfide
bonds (McDonnell, 2007). Results of our research
show that the first and most important part of this
protein binds GP120 and CD4 proteins. In Molegro
software the basis of the three dimensional structure
of this protein is positioned between 264 to 299 amino
acids where sulfide bonds are located. Breaking this
binding can change the spatial structure of this
protein, therefore prevents this part from binding with
CD4.
4. CONCLUSION
At the first stages of HIV proliferation, silver nano
particles can act as antivirus agent for deactivation of
the virus in a short period of time. This process is
done through interaction of nano silver with 2
disulfide bonds located in carboxyl )HIV-1 gp120) so
that silver ions, through interaction with thiol group,
decrease disulfide bonds leading to protein
denaturation. Considering other similar scientific
investigations, one can conclude that the mentioned
process is fulfilled through substitution of single
valence Ag+ with H+ existed in thiol group.
Bioinformatic results using software and servers show
that silver ions make interaction with thiol group
through which the disulfide bonds decrease and the
virus will be destroyed.
REFERENCES
Achal V, Kumari D, Pan X (2011). Bioremediation of
Chromium Contaminated Soil by a Brown-rot
Fungus, Gloeophyllum sepiarium. Research
Journal of Microbiology, 6: 166-171.
APHA (1998). Standard Methods for Examination of
Water and Wastewater, 20th ed. American
Public Health Association, Washington, DC,
USA.
Gavanji et al.
Bioinformatics Prediction of Interaction Silver Nanoparticles on the Disulfide Bonds of HIV-1 Gp120 Protein
72
Ahamed M, AlSalhi MS, Siddiqui M (2010). Silver
nanoparticle applications and human health.
Clinica chimica acta, 411(23):1841-1848.
Allan J, Coligan J, Barin F, McLane M, Sodroski J,
Rosen C, Haseltine W, Lee T, Essex M (1985).
Major glycoprotein antigens that induce
antibodies in AIDS patients are encoded by
HTLV-III. Science, 228(4703):1091-1094.
Arora S, Jain J, Rajwade J, Paknikar K (2008).
Cellular responses induced by silver
nanoparticles: In vitro studies. Toxicology
letters, 179(2):93-100.
Alvarez-Puebla R, Dos Santos Jr D, Aroca R (2004).
Surface-enhanced Raman scattering for
ultrasensitive chemical analysis of 1 and 2-
naphthalenethiols. Analyst, 129(12):1251-1256.
Alcamo IE (1997). Fundamentals of Microbiology,
The Benjamin/Cummings Publishing Company.
Al-Jabri A, Alenzi F (2009). Vaccines, virucides and
drugs against HIV/AIDS: hopes and optimisms
for the future. The open AIDS journal, 3:1-3.
Berger T, Spadaro J, Chapin S, Becker R (1976).
Electrically generated silver ions: quantitative
effects on bacterial and mammalian cells.
Antimicrobial Agents and Chemotherapy,
9(2):357–358.
Bhattacharya R, Mukherjee P (2008). Biological
properties of “naked” metal nanoparticles.
Advanced drug delivery reviews, 60(11):1289-
1306.
Baram-Pinto D, Shukla S, Perkas N, Gedanken A,
Sarid R (2009). Inhibition of herpes simplex
virus type 1 infection by silver nanoparticles
capped with mercaptoethane sulfonate.
Bioconjugate Chemistry, 20(8):1497-1502.
Borkow G, Lapidot A (2005). Multi-targeting the
entrance door to block HIV-1. Current Drug
Targets-Infectious Disorders, 5(1):3-15.
Clement J, Jarrett P (1994). Antimicrobial silver.
Metal-Based Drugs, 1:467-482.
Choi SH (2006). A new composition of nanosized
silica-silver for control of various plant
diseases. The Plant Pathology Journal,
22(3):295-302.
Dowsett C (2004). The use of silver-based dressings
in wound care. Nursing Standard, 19(7):56–60.
Elechiguerra JL, Burt JL, Morones JR, Camacho-
Bragado A, Gao X, Lara HH, Yacaman MJ
(2005). Interaction of silver nanoparticles with
HIV-1. Journal of Nanobiotechnology, 3(6):1-
10.
Feng Q, Wu J, Chen G, Cui F, Kim T, Kim J (2000).
A mechanistic study of the antibacterial effect
of silver ions on Escherichia coli and
Staphylococcus aureus. Journal of biomedical
materials research, 52(4):662-668.
Hotta M, Nakajima H, Yamamoto K, Aono M (1998).
Antibacterial temporary filling materials: the
effect of adding various ratios of Ag-Zn-
Zeolite. Journal of oral rehabilitation,
25(7):485-489.
Jung WK, Kim SH, Koo HC, Shin S, Kim JM, Park
YK, Hwang SY, Yang H, Park YH (2007).
Antifungal activity of the silver ion against
contaminated fabric. Mycoses, 50(4):265-269.
Kumar A, Vemula PK, Ajayan PM, John G (2008).
Silver-nanoparticle-embedded antimicrobial
paints based on vegetable oil. Nature Materials,
7(3):236-241.
Lekutis C, Olshevsky U, Furman C, Thali M,
Sodroski J (1992). Contribution of disulfide
bonds in the carboxyl terminus of the human
immunodeficiency virus type I gp120
glycoprotein to CD4 binding. JAIDS Journal of
Acquired Immune Deficiency Syndromes,
5(1):78-81.
Li Y, Leung P, Yao L, Song Q, Newton E (2006).
Antimicrobial effect of surgical masks coated
with nanoparticles. Journal of Hospital
Infection, 62(1):58-63, 2006.
Lara HH, Ayala-Nuñez NV (2010). Ixtepan-Turrent
L, Rodriguez-Padilla C, Mode of antiviral
action of silver nanoparticles against HIV-1.
Journal of Nanobiotechnology, 8(1):1-8.
Lu L, Sun R, Chen R, Hui CK, Ho CM, Luk JM, Lau
G, Che CM (2008). Silver nanoparticles inhibit
hepatitis B virus replication. Antiviral Therapy,
13:253-262.
Lin YSE, Vidic RD, Stout JE, Yu VL (2002).
Negative effect of high pH on biocidal efficacy
of copper and silver ions in controlling
Legionella pneumophila. Applied and
environmental microbiology, 68(6):2711-2715.
Morones JR, Elechiguerra JL, Camacho A, Holt K,
Kouri JB, Ramírez JT, Yacaman MJ (2005).
The bactericidal effect of silver nanoparticles.
Nanotechnology, 16(10):2346.
McDonnell G (2007). Chemical disinfection.
Antisepsis, disinfection and sterilization, 111-
115.
Pal S, Tak YK, Song JM (2007). Does the
antibacterial activity of silver nanoparticles
depend on the shape of the nanoparticle? A
study of the gram-negative bacterium
Escherichia coli. Applied and environmental
microbiology, 73(6):1712-1720.
Rogers JV, Parkinson CV, Choi YW, Speshock JL,
Hussain SM (2008). A preliminary assessment
of silver nanoparticle inhibition of monkeypox
virus plaque formation. Nanoscale Research
Letters, 3(4):129-133.
International Journal of Scientific Research in Knowledge, 2(2), pp. 67-74, 2014
73
Raad II, Hanna HA (2002). Intravascular catheter-
related infections: new horizons and recent
advances. Archives of internal medicine,
162(8):871-878.
Russell A, Path F, Sl FP, Hugo W (1994).
Antimicrobial activity and action of silver.
Progress in medicinal chemistry, 31: 351-369.
Shahverdi AR, Fakhimi A, Shahverdi HR, Minaian S
(2007). Synthesis and effect of silver
nanoparticles on the antibacterial activity of
different antibiotics against Staphylococcus
aureus and Escherichia coli. Nanomedicine:
Nanotechnology Biology and Medicine, 3(2):
168-171.
Sun RWY, Chen R, Chung NPY, Ho CM, Lin CLS,
Che CM (2005). Silver nanoparticles fabricated
in Hepes buffer exhibit cytoprotective activities
toward HIV-1 infected cells. Chemical
Communications, (40):5059-5061.
Sun L, Singh AK, Vig K, Pillai SR, Singh SR (2008).
Silver nanoparticles inhibit replication of
respiratory syncytial virus. Journal of
Biomedical Nanotechnology, 4(2):149-158.
Takai K, Ohtsuka T, Senda Y, Nakao M, Yamamoto
K, Matsuoka J, Hirai Y (2002). Antibacterial
properties of antimicrobial-finished textile
products. Microbiology and immunology,
46(2):75-81.
Tokumaru T, Shimizu Y, Fox Jr C (1974). Antiviral
activities of silver sulfadiazine in ocular
infection. Research communications in
chemical pathology and pharmacology,
8(1):151-158.
Salzwedel K, West JT, Hunter E (1999). A conserved
tryptophan-rich motif in the membrane-
proximal region of the human
immunodeficiency virus type 1 gp41
ectodomain is important for Env-mediated
fusion and virus infectivity, Journal of virology,
73(3):2469-2480.
Wiederstein M, Sippl MJ (2007). ProSA-web:
interactive web service for the recognition of
errors in three-dimensional structures of
proteins. Nucleic acids research, 35(2):407-410.
Gavanji et al.
Bioinformatics Prediction of Interaction Silver Nanoparticles on the Disulfide Bonds of HIV-1 Gp120 Protein
74
Shahin Gavanji graduated in Biotechnology at MSc at the Department of Biotechnology, Faculty of
Advanced Sciences and Technologies, University of Isfahan, Isfahan, Iran. He has over 10 international
medals in invention. Shahin Gavanji's research has focused on Pharmacy and Pharmacology, Nano
Biotechnology, Bioinformatics, Biotechnology - Medical Biotechnology. He is editor in chief of
International Journal of Scientific Research in Inventions and New Ideas.
Dr. Hassan Mohabatkar is a faculty member at Department of Biotechnology, Faculty of Advanced
Sciences and Technologies, University of Isfahan, Isfahan, Iran. His research has focused on
Bioinformatics.
Hojjat Baghshahi graduated in Animal Science at MSc at the Department of Animal Sciences, College of
Agriculture, Isfahan University of Technology (IUT), Isfahan, IRAN.
Dr. Ali Zarrabi, Assistant Professor of Nano-biotechnology, Department of Biotechnology, Faculty of
Advanced Sciences and Technologies, University of Isfahan, Isfahan, Iran.
International Journal of Scientific Research in Knowledge, 2(2), pp. 75-82, 2014
Available online at http://www.ijsrpub.com/ijsrk
ISSN: 2322-4541; ©2014 IJSRPUB
http://dx.doi.org/10.12983/ijsrk-2014-p0075-0082
75
Full Length Research Paper
A Comparative Evaluation of Drug Release and Permeability of Ethylcellulose,
Cellulose Acetate and Eudragit RS100 Microspheres
Prakash Katakam1*, Saousen R. Diaf
1, Baishakhi Dey
2, Shanta K. Adiki
3, Babu R. Chandu
1, K.P.R. Chowdary
4
1Faculty of Pharmacy, University of Zawia, Libya
2School of Medical Science and Technology, IIT Kharaghpur, India
3Nirmala College of Pharmacy, Guntur, India
4College of Pharmaceutical Sciences, Andhra University, India
*Corresponding Author: [email protected]
Received 01 December 2013; Accepted 11 January 2014
Abstract. Present study aims at comparative evaluation of drug release and permeability of diclofenac sodium loaded
ethylcellulose (EC), cellulose acetate (CA) and eudragit (EU) microspheres. Microspheres of EC, CA and EU containing
diclofenac sodium were prepared by an emulsification-solvent evaporation (oil-in-oil, o/o) method and were investigated for a
comparative evaluation of various parameters. The microspheres were found discrete, free flowing, multinucleate, monolithic
and spherical. About 5560% of all microspheres prepared were in the size range of –20+30 (715 m) mesh size. The
encapsulation efficiency was in the range of 97.1106.4% with various polymers. The wall thickness of microspheres was in
the range of 13.69-74.97m which depended on polymer employed and was directly proportional to polymer concentration.
Diclofenac release from the microspheres was slow over longer periods of time and depended on the polymer used and
coat:core ratio. Release was diffusion controlled and followed first order kinetics. Good linear relationships were observed
between percent coat, wall thickness and release rate constant with all the three polymers. The slopes of percent coat vs release
rate (k1) plots were found to be 0.4117, 0.2351 and 0.9762; and those of wall thickness (h) vs drug release rate (k1) plots were
found 0.2549, 0.1863 and 0.7850 respectively for EC, CA and EU microspheres. The lower the slope the better is the
controlling effect. Cellulose acetate exhibited better release-controlling effect than that of ethylcellulose and eudragit. The
increasing order of diclofenac release rate and permeability observed with various microspheres was, cellulose acetate <
ethylcellulose < eudragit RS100. The possible permeability of drug from the prepared porous micrsopheres could be due to
osmotic pressure generated by diclofenac.
Keywords: Diclofenac sodium, Microspheres, Ethylcellulose, Cellulose acetate, Eudragit RS100 Release kinetics
1. INTRODUCTION
Microspheres are solid, approximately 1 to 1000 m
in size and are made of synthetic and natural
polymeric, waxy or other protective materials both
biodegradable and non-biodegradable (Vyas and
Khar, 2002). The internal structure of microspheres
varies as a function of polymer and the process
employed to prepare them (Brannon-Peppas, 1992).
Reservoir microcapsules have a core of drug coated
with a polymer. Whereas in monolithic microspheres,
the drug is distributed homogeneously throughout the
polymeric matrix. Microspheres have been widely
accepted as a means to achieve oral and parenteral
controlled release (Sau-hung et al., 1987).
Microspheres provide several advantages over other
sustained release systems, especially matrix type
tablets. They can be widely distributed throughout the
gastrointestinal tract, improve drug absorption and
minimize side effects due to localized buildup of
irritating drugs against the gastrointestinal mucosa (Li
et al., 1988).
The rate of drug release from microspheres dictates
their therapeutic action. Release is governed by the
molecular structure of the drug and polymer, the
resistance of the polymer to degradation and the
surface area and porosity of microspheres (Izumikawa
et al., 1991; Pitt and Schindler, 1983). Drug release
from polymeric systems with a variety of geometries
has been described (Cheung et al., 1988). Zero order
release kinetics may be more easily achieved with slab
or rod geometries than spheres. The rate of release
from spheres may result from polymer diffusion or
erosion (Cartensen, 1984; Crank, 1975).
Ethylcellulose, cellulose acetate and eudragit
RS100 are non-toxic, biocompatible polymers with
Katakam et al.
A Comparative Evaluation of Drug Release and Permeability of Ethylcellulose, Cellulose Acetate and Eudragit RS100 Microspheres
76
good film-forming properties and have been
extensively used in coating (Kent and Rowe, 1978)
and microencapsulation (Porter, 1989) to prepare
microspheres. Though they have been studied
(Prakash et al., 2007; Padala et al., 2009; Chowdary
and Ratna 1993; Kawashima, 1989; Hasan, 1992;
Lorenzo-Lamon et al., 1998) individually for
microspheres and controlled release, no reports on
comparative evaluation of their drug release and
permeability characteristics for diclofenac are
available. In the present study a comparative
evaluation of drug release and permeability of
ethylcellulose (EC), cellulose acetate (CA) and
eudragit (EU) microspheres has been made employing
diclofenac sodium as core. Oral controlled release
formulations are needed for diclofenac because of its
short biological half-life of 2.0 hr and gastrointestinal
irritation if present in larger concentrations (Gilman,
1991). In the present investigation a comparative
evaluation of drug release and permeability of
ethylcellulose (EC), cellulose acetate (CA) and
eudragit (EU) microspheres was made by employing
diclofenac sodium as core.
2. MATERIALS AND METHODS
2.1. Materials
Diclofenac sodium was a gift sample from M/s
Roland Pharmacetuicals, Berhampur, India.
Ethylcellulose (Loba Chemie, Mumbai, with an
ethoxy content of 47.5% by weight and a viscosity of
22 cps in a 5% concentration by weight, in a 80:20
toluene-ethanol solution at 25 oC), cellulose acetate
(Loba Chemie, Mumbai with a viscosity of 100140
cps in a 6% solution in 95% acetone-water mixture at
20 oC), eudragit RS100 (with a viscosity of 15 mPa s
in a 2% acetone-ethanol (1:1) solution at 25 oC), n-
hexane (Ranbaxy), acetone (Merck) and liquid
paraffin I.P. were procured from commercial sources.
All other reagents used were of analytical grade.
2.2. Preparation of microspheres
Microspheres of ethylcellulose, cellulose acetate and
eudragit containing diclofenac sodium were prepared
by an emulsification-solvent evaporation (oil-in-oil,
o/o) method. The polymer (EC, CA or EU) (2.0 gm)
was dissolved in acetone (100 mL) to form a
homogeneous polymer solution. The core material,
diclofen
added to the polymer solution (20 mL) and mixed
thoroughly. The resulting mixture was then added in a
thin stream to 200 mL of liquid paraffin contained in a
450 mL beaker, while stirring at 1000 rpm to emulsify
the added dispersion as fine droplets. A Remi medium
duty stirrer with speed meter (Model RQT 124) was
used for stirring. The solvent was then removed by
continuous stirring at room temperature (28 oC) for 3
hr to produce spherical microspheres. The
microspheres were collected by vacuum filtration and
washed repeatedly with n-hexane to remove adhering
liquid paraffin. The product was then air dried to
obtain discrete microspheres. Different proportions of
coat:core materials namely 1:9 (MC1), 2:8 (MC2), 3:7
(MC3), 4:6 (MC4) and 5:5 (MC5) were used in each
case to prepare microspheres with varying coat
thickness.
2.3. Evaluation of microspheres
Diclofenac sodium content in the microspheres was
estimated by using UV-spectrophotometric method
(The United States Pharmacopoeia, 1999) based on
measurement of absorbance at 276 nm in phosphate
buffer of pH 6.8. The method was validated for
linearity, accuracy and precision. The method obeyed
Beer-Lambert’s law in the concentration range 1-20
μg/mL. When a standard drug solution was assayed
repeatedly (n=6), the mean error (accuracy) and
relative standard deviation (precision) were found to
be 1.2% and 2% respectively. The encapsulation
efficiency was determined by estimating the drug
content in the microspheres and using the formula:
For size distribution analysis, different sizes in a
batch were separated by sieving using a range of
standard sieves. The amounts retained on different
sieves were weighed. Theoretical mean wall thickness
of the microspheres was determined by the method of
Luu et al (1973) using the equation:
12
1
d P1Pd 3
d P1 rh
where, h = wall thickness (m); r = mean radius of
microspheres (m); d1 = density of core material
(g/cm3); d2 = density of coat material (g/cm
3) and P =
proportion of medicament in the microspheres.
The microspheres prepared along with their
diclofenac content, encapsulation efficiency and wall
thickness measurements are given in Table 1. The
microspheres were observed under a scanning electron
microscope (SEM-LEICA, S430, UK). For SEM, the
microspheres were mounted directly on the sample
stub, using double-sided sticking tape, and coated
with gold film (thickness 200 nm) under reduced
pressure (0.001 torr).
International Journal of Scientific Research in Knowledge, 2(2), pp. 75-82, 2014
77
2.4. Drug release study
Release of diclofenac sodium from the microspheres
of size 20/30 was studied in phosphate buffer of pH
6.8 (900 mL) at 37 oC using an USP XXIV 6-stage
dissolution test apparatus (M/s. Campbell Electronics,
Mumbai) with a paddle stirrer at 100 rpm. A sample
of microspheres equivalent to 100 mg of diclofenac
sodium was used in each test. Samples were
withdrawn through a filter (0.4 m) at different
intervals of time and were assayed at 276 nm for
diclofenac sodium using an Elico UV-visible double
beam spectrophotometer. The drug release
experiments were conducted in triplicates.
From the drug release data, the permeability
coefficient (Pm) for various microspheres was
calculated using the equation as described by Koida et
al (1986).
s
app
A.C
.V.HKPm
where, Pm = permeability coefficient
(cm2/min), Kapp = apparent dissolution rate constant
calculated as mg/min from, the slope of the early
linear portion, V = volume of dissolution medium
(cm3), H = wall thickness of microspheres (cm), A =
surface area of the microspheres (cm2) and Cs =
solubility of the core in the dissolution medium (mg).
3. RESULTS AND DISCUSSION
Ethylcellulose, cellulose acetate and eudragit
microspheres of diclofenac sodium could be prepared
by the emulsification and solvent evaporation method
using acetone as solvent for the polymer as reported
by us (Prakash et al., 2007). The microspheres were
found to be discrete, spherical and free flowing.
Scanning electron microscogram (SEM) showed that
the microspheres prepared by all the three polymers
were nearly spherical with rough microporous surface
(Fig. 1). Regarding the internal structure the nature of
the method indicates that the microspheres produced
were of multinucleate monolithic type.
The sizes could be separated and a more uniform
size range of microspheres could readily be obtained
by sieving. The size distributions were normal in all
the batches with a large proportion, overall about
5560%, in the size range of –20+30 (715 m) mesh
size.
Fig. 1: Scanning electron micrograms of ethylcellulose (A), cellulose acetate (B) and eudragit RS100 (C) microspheres loaded
with diclofenac sodium.
Low coefficient of variation (< 4.7%) in percent
drug content indicated uniformity of drug content in
each batch of microspheres prepared with different
polymers (Table 1). The encapsulation efficiency was
found in the range 97.1103.8% with ethylcellulose,
99.4106.4% with cellulose acetate and
101.4104.2% with eudragit. As the microspheres
were spherical, the theoretical mean thickness of the
wall that surrounds the core particles in the
microspheres was calculated as per Luu et al (1973).
Microspheres prepared employing various ratios of
coat:core in each case (polymer) were found to have
different wall thickness.
Diclofenac release from the microspheres of size
20/30 was studied in phosphate buffer of pH 6.8 for a
period of 12 hr. Diclofenac release from all the
microspheres was slow and spread over extended
periods of time (Table 2). Plots of log percent drug
remaining vs time (Fig. 2) were found to be linear
(r>0.9516) with all the microspheres indicating that
the drug release from these microspheres was
according to first order kinetics with all the three
polymers. In a monolithic microsphere the path length
does not remain constant, since the drug in the center
has a longer path or travel than the drug near the
surface and therefore the rate of release decreases
exponentially with time. The release rates of various
microspheres are given in Table 2. At all coat:core
ratios the release rates were higher with eudragit and
the order of increasing release rates with various
Katakam et al.
A Comparative Evaluation of Drug Release and Permeability of Ethylcellulose, Cellulose Acetate and Eudragit RS100
Microspheres
78
polymers was CA < EC < EU. In each case the
release was depended on the percent of coat and wall
thickness of the microspheres.
Table 1: Coat:core ratio, drug content, encapsulation efficiency and wall thickness of the microspheres prepared.
Microspheres Coat:Core ratio Percent drug content Encapsulation
efficiency (%)
Wall thickness
(m) Theoretical Practical*
ECMC1 1:9 90 88.4 (1.6968) 98.2 18.90
ECMC2 2:8 80 77.7 (1.5444) 97.1 35.50
ECMC3 3:7 70 71.2 (3.6516) 101.7 50.16
ECMC4 4:6 60 61.5 (4.1951) 102.5 63.24
ECMC5 5:5 50 51.9 (2.3121) 103.8 74.97
CAMC1 1:9 90 93.5 (2.139) 103.8 13.69
CAMC2 2:8 80 83.1 (2.0457) 102.1 26.93
CAMC3 3:7 70 69.6 (3.8793) 99.4 39.76
CAMC4 4:6 60 62.3 (4.6548) 103.3 52.17
CAMC5 5:5 50 53.2 (4.1353) 106.4 64.20
EUMC1 1:9 90 93.2 (2.253) 103.5 14.18
EUMC2 2:8 80 81.9 (2.3199) 102.4 27.78
EUMC3 3:7 70 72.6 (2.8925) 103.7 40.83
EUMC4 4:6 60 62.5 (2.72) 104.2 53.35
EUMC5 5:5 50 51.7 (4.091) 101.4 65.39
* Figures in parentheses are coefficient of variation (cv) values.
Fig. 2: Log percent of drug remaining vs. time plots of ethylcellulose (A), cellulose acetate (B) and eudragit RS100 (C)
microspheres. MC1 (□), MC2 (), MC3(◊), MC 4 (Ο) and MC5(* ).
International Journal of Scientific Research in Knowledge, 2(2), pp. 75-82, 2014
79
Table 2: Release characteristics of microspheres prepared.
Microspheres Mean percent drug released at different time intervals (hr) (%±SD), n=3 T50
(hr)a k1 (hr-1)b
1.0 2.0 4.0 8.0 10.0 12.0
ECMC1 76.81±4.25 80.15±5.18 85.03±5.29 100.00±5.13 - - 7.09 0.3246
ECMC2 56.90±3.44 65.77±3.79 79.11±4.29 100.00±4.62 - - 7.37 0.3125
ECMC3 53.44±3.79 62.50±4.31 73.46±4.37 100.00±5.36 - - 8.21 0.2805
ECMC4 52.08±2.53 60.96±4.33 68.96±4.77 90.03±4.22 - - 8.74 0.2635
ECMC5 42.67±2.94 50.77±3.79 62.43±3.21 79.11±4.16 85.01±4.66 89.78±4.55 13.91 0.1656
CAMC1 63.56±4.19 79.23±3.61 88.48±4.45 100.00±4.97 - - 9.63 0.2392
CAMC2 54.36±3.88 64.46±4.34 70.36±5.49 86.87±4.37 100.00±5.14 - 11.98 0.1923
CAMC3 50.30±2.59 61.02±3.28 68.26±4.62 83.17±4.14 89.12±4.62 100.00±5.12 14.47 0.1591
CAMC4 48.35±3.18 56.49±3.46 65.40±5.18 80.94±3.85 86.39±4.26 91.78±4.52 14.58 0.158
CAMC5 39.15±3.31 45.22±3.11 56.36±4.73 74.45±4.28 82.71±4.76 86.17±5.18 16.52 0.1394
EUMC1 81.99±4.75 89.16±4.39 98.45±5.35 - - - 3.84 0.5995
EUMC2 76.93±6.39 83.70±5.44 94.61±5.97 - - - 4.86 0.4738
EUMC3 58.65±3.86 70.38±4.71 81.67±3.82 96.38±3.92 - - 6.54 0.3521
EUMC4 41.98±3.64 61.19±3.88 75.56±4.26 90.08±4.51 97.57±5.33 - 9.08 0.2535
EUMC5 38.93±2.71 56.34±3.29 63.39±3.74 80.58±4.89 96.26±4.27 - 11.02 0.2089 aTime for 50% release;
bFirst order release rate constant
Good linear relationships were observed between
percent coat (or) wall thickness and release rate
(Fig. 3). The relationships could be expressed by the
following linear equations.
y = 0.4117 x + 39.23 for ethyl cellulose
y = 0.2351 x + 24.26 for cellulose acetate
y = 0.9762 x + 64.72 for eudragit
where, x is percent coat and y is first order release rate
(k1 hr-1
) of the microspheres.
y = 0.2549 x + 39.31 for ethyl cellulose
y = 0.1863 x + 25.09 for cellulose acetate
y = 0.7850 x + 69.39 for eudragit
where, x is the wall thickness (m) and y is first order
release rate (k1 hr-1
) of the microspheres.
Fig. 3: Relationship between (A) percent coat and release rate and (B) wall thickness and release rate of ethyl cellulose (Ο),
cellulose acetate (□) and eudragit RS100 () microspheres.
The slope of the linear regression between percent
coat and release rate (k1) indicates the release
controlling effect of the polymer. The lower the slope
the better is the controlling effect. The slopes were
found to be 0.4117, 0.2351 and 0.9762 respectively
for the coat materials ethylcellulose, cellulose acetate
and eudragit. Similarly from the plots of wall
thickness (h) vs drug release rate (k1), the slopes
Katakam et al.
A Comparative Evaluation of Drug Release and Permeability of Ethylcellulose, Cellulose Acetate and Eudragit RS100
Microspheres
80
obtained for ethylcellulose, cellulose acetate and
eudragit microspheres were 0.2549, 0.1863 and
0.7850 respectively. Thus cellulose acetate was found
to have better release controlling effect than the other
two polymers (Pratap et al., 2012). The order of their
effectiveness in controlling drug release was found as
CA > EC > EU. Plots of amount released vs square
root of time (Fig. 4) were found to be linear with all
the three polymers indicating that the drug release
from these microspheres was diffusion controlled.
Fig. 4: Amount released vs. square root of time plots of ethylcellulose (A), cellulose acetate (B) and eudragit RS100 (C)
microspheres. MC1 (□), MC2 (), MC3(◊), MC 4 (Ο) and MC5(*).
Permeability of the microspheres was calculated
based on the release data as described by Koida et al
(1986). Permeability values of various microspheres
are summarized in Table 3. At all ratios of coat:core,
the microspheres of cellulose acetate were less
permeable than those of ethyl cellulose and eudragit
RS100. The order of increasing permeability of the
microspheres was CA < EC < EU. The permeability
of microspheres having porous surface occurs when
drug release is driven by osmotic pressure (Ozturk et
al., 1990). The possibe permeability of drug from the
prepared porous micrsopheres could be due to osmotic
pressure generated by diclofenac.
Table 3: Permeability coefficient (Pm) values of various microspheres prepared.
Microspheres Coat : core ratio Permeability coefficient, Pm (cm2/min) of microscpheres
EC CA EU
MC1 1:9 8.93 5.35 7.15
MC2 2:8 12.43 9.01 13.15
MC3 3:7 16.50 12.32 14.74
MC4 4:6 20.26 15.53 13.78
MC5 5:5 19.68 15.47 15.67
4. CONCLUSION
Ethylcellulose, cellulose acetate and eudragit RS100
microspheres containing diclofenac sodium could be
prepared by the emulsification-solvent evaporation
method using acetone as solvent for the polymer with
an encapsulation efficiency varying between
97.1106.4%. The microspheres were discrete, free
flowing, multinucleate, monolithic and spherical.
Diclofenac release from the microspheres was slow
over longer periods of time and depended on the
polymer used and coat:core ratio. Release was
diffusion controlled and followed first order kinetics.
Good linear relationships were observed between
percent coat, wall thickness and release rate constant
with all the three polymers. Cellulose acetate
exhibited better release-controlling effect than
ethylcellulose and eudragit. The order of increasing
diclofenac release rate and permeability observed with
various microspheres was cellulose acetate <
ethylcellulose < eudragit.
REFERENCES
Brannon-Peppas L (1992). Design and mathematical
analysis of controlled release from
International Journal of Scientific Research in Knowledge, 2(2), pp. 75-82, 2014
81
microsphere-containing polymeric implants. J.
Contr. Rel. 20: 201-207.
Cartensen JT (1984). Controlled Drug Delivery.
Muller BW Ed. Wissenschaftliche
Verlagsgesellschaft GmbH: Stuttgart. 132-145.
Cheung WK, Yakobi A, Silber BM (1988).
Pharmacokinetic approach to the rational design
of controlled or sustained release formulations.
J. Contr. Rel. 6: 263-270.
Chowdary KP, Ratna JV (1993). Comparative
evaluation of ethyl cellulose, methyl cellulose
and cellulose acetate microcapsules prepared by
a complex emulsion method. Indian Drugs. 30:
179-184.
Crank J (1975). The mechanics of diffusion, 2nd Ed.
Oxford Science Publications: Oxford. 1975.
Gilman AG, Theodore WR, Alans N, Taylor P (1991).
Goodman and Gilman’s the pharmacological
basis of therapeutics. 8th Edn., Mc Graw-Hill,
New York. 669.
Hasan M, Najib N, Suleiman M, El-Sayed Y, Abdel-
Hamid M (1992). Invitro and invivo evaluation
of sustained-release and enteric-coated
microcapsules of diclofenac sodium. Drug Dev.
Ind. Pharm. 18(18): 1981-1988.
Izumikawa S, Yoshioka S, Aso Y, Takeda Y (1991).
Preparation of poly (l-lactide) microspheres of
different crystalline morphology and effect of
crystalline morphology on drug release rate. J.
Contr. Rel. 15: 133-140.
Kawashima Y, Niwa T, Handa T, Takeuchi H,
Iwamoto T, Itoh Y (1989). Preparation of
prolonged release spherical micro-matrix of
ibuprofen with acrylic polymer by emulsion-
solvent diffusion method for improving
bioavailability. Chem. Pharm. Bull. 37: 425-
429.
Kent DJ, Rowe RC (1978). Solubility studies on
ethylcellulose used in film coating. J. Pharm.
Pharmcol. 30: 808-810.
Koida Y, Kobayashi M, Samejima M (1986). Studies
on microcapsules. IV. Influence of properties of
drugs on microencapsulation and dissolution
behavior. Chem. Pharm. Bull. 34: 3354-3361.
Li SP, Kowarski CR, Field KM, Grim MW (1988).
Recent advances in microencapsulation
technology and equipment. Drug Dev. Ind.
Pharm. 14: 354-376.
Lorenzo-Lamon ML, Remunan-Lopez C, Vila-Jato
JL, Alonso MJ (1998). Design of
microencapsulated chitosan microspheres for
colonic drug delivery. J. Contr. Rel. 52: 109-
118.
Luu SN, Carlier PF, Delort P, Gazzola J, Lafont D
(1973). Determination of coating thickness of
microcapsules and influence upon diffusion. J.
Pharm. Sci. 62: 452-455.
Ozturk AG, Ozturk SS, Palsson BO, Wheatley TA,
Dressman JB (1990). Mechanism of release
from pellets coated with an ethylcellulose-based
film. J. Control Release. 14: 203-213.
Padala NR, Prakash K, Bonepally CSR, Krishnaveni
B, Shantakumari K, Lakshmi NM (2009).
Stavudine Loaded Microcapsules using various
Cellulose Polymers: Preparation and In-Vitro
Evaluation. International Journal of
Pharmaceutical Science and Nanotechnology.
2(2): 551-556.
Pitt CG, Schindler A (1983). Kinetics of drug
cleavage and release from matrices containing
covalent polymer-drug conjugates. In: Bruck
SD Ed. Controlled drug delivery. CRC Press:
Boca Raton, FL, Vol. 1: 53-80.
Porter SC (1989). Controlled-release film coatings
based on ethyl cellulose. Drug. Dev. Ind.
Pharm. 15(10): 1495-1521.
Prakash K, Raju PN, Shanta KK, Lakshmi MN
(2007). Preparation and characterization of
lamivudine microcapsules using various
cellulose polymers. Trop. J. Pharm. Res. 6(4):
841-847.
Pratap KG, Chowdary KPR, Yasoda KK (2012).
Evaluation of starch acetate as
microencapsulating agent for controlled release
of carbamazepine in comparison to other known
polymers. International Journal of Pharma
Sciences. 2(4): 67-69.
Sau-hung, Leung S, Robinson JR (1987). In:
Controlled Drug Delivery, Fundamentals and
Applications, 2nd Ed. Marcel Dekker, Inc.,
New York. 448.
The united states pharmacopoeia (1999). 24rd Edn.,
The United States Pharmacopoeial Convention,
Inc., Rockville, MD, 547.
Vyas SP, Khar RK (2002). Eds. Targetted and
controlled drug delivery novel carrier systems,
1st Ed. CBS Publishes, New Delhi. 418.
Katakam et al.
A Comparative Evaluation of Drug Release and Permeability of Ethylcellulose, Cellulose Acetate and Eudragit RS100 Microspheres
82
Prakash Katakam, M.Pharm., Ph.D., has done his Ph.D. (Pharmaceutical Sciences) from Berhampur
University, Orissa, India. Research contribution includes design and biomedical evaluation in vitro and in
vivo, of various controlled release drug delivery systems like subgingival delivery films, microcapsules
and matrix tablets and conventional dosage forms. Other areas of research interest are biomaterials,
bioanalytical method development and natural medicine. He has guided two Ph.D. students and several
M.Pharm students. He has 80 peer reviewed research articles and four Indian patents to his credit. He is a
member of various professional bodies and editorial board member of several international journals.
Presently he is working on biomaterials and their novel modifications for pharmaceutical application.
Saousen R. Diaf, has completed her MSc Pharmaceutical Technology and Pharmacy Science in 2009 from
School of Pharmacy, University of Complutense, Madrid, Spain. She has completed her Master degree in
Pharmacognosy and Natural products in 2006 from College of Pharmacy, University of Alfateh, Tripoli,
Libya. She has been a recipient of Fulbright Exchange Libya Scholarship 2013, University of Nebraska
Medical Center, College of Public Health, Omaha, Nebraska, USA. Her areas of research interest are
Novel Drug Delivery systems and Industrial Pharmacy.
Ms Baishakhi Dey, M.Pharm., is a Faculty of Pharmacy, pursuing her PhD. Her research interests are
mostly in herbal anti-diabetic bio-actives, Nanotechnical approaches to drug delivery, Formulation of
nutraceuticals via innovative process technologies and pharmacoepidemiological surveys on disease
pattern. Currently she is having 12 peer reviewed journal articles, 5 conference papers, 2patents. Ms Dey is
the co-author of two books, and four book chapters.
Shanta Kumari Adiki, PhD, has done her PhD (Pharmaceutical Sciences) from Berhampur University,
India. She has completed her Master degree from Andhra University, India. She has 12 years of rich
experience Research and Teaching and published over 35 research articles in various international journals
and has two Indian patents. Her primary research area is Analytical chemistry and Bioanalytical Method
Development. Her other areas of research interest are formulation development, novel drug delivery
systems and natural products. She is presently guiding two PhD students and has guided several M.Pharm
projects.
Babu Rao Chandu, M.Pharm., Ph.D., has done his Ph.D. (Pharmaceutical Sciences) from Andhra
University, India. During his highly noticeable research experience over 20 years he has published over
140 papers in various international impact journals and one Indian patent. His research areas of interest
include, Phytochemistry, Natural medicine, Drug design and synthesis, Bioanalytical method development
and Formulation of dosage forms. He is a member of various professional bodies and editorial member of
several international journals. He has guided five Ph.D. students and several M.Pharm students.
Chowdary, KPR, M.Pharm., PhD., PGDAS, is a retired professor from University College of
Pharmaceutical Sciences, Andhra University. His primary area of research is Controlled Drug Delivery
Systems. He is well known for his research in the field of Microencapsulation and polymer science. He is
highly specialized in modification of drugs to improve their solubility by complexation. He has produced
over 60 PhDs and guided over 100 M.Pharm research projects.
International Journal of Scientific Research in Knowledge, 2(2), pp. 83-91, 2014
Available online at http://www.ijsrpub.com/ijsrk
ISSN: 2322-4541; ©2014 IJSRPUB
http://dx.doi.org/10.12983/ijsrk-2014-p0083-0091
83
Full Length Research Paper
Investigation of a Proposed Four Storey Building Sites Using Geophysical and
Laboratory Engineering Testing Methods in Lagos, Nigeria
Oyedele Kayode Festus*, Adeoti, Lukman, Oladele Sunday and Kamil Akintunde
Department of Geosciences, University of Lagos, Lagos, Nigeria
*Corresponding Author: Email: [email protected]
Received 06 December 2013; Accepted 11 January 2014
Abstract. The spate of engineering structures collapse in Lagos metropolis with its attendant loss of lives and properties has
assumed an alarming proportion in recent times. Efforts to mitigate such incidence has necessitated an integrated geophysical
and geotechnical investigation of a proposed four storey building sites with a view to determine the suitability of the site for
the proposed project. Resistivity investigation, un-drained multi-stage triaxial compression and oedometer consolidation tests
were carried out to determine the engineering properties of the subsurface. The results revealed peaty clay to silty sand
materials characterized by 35kN/m2 - 75kN/m
2 cohesion values, (5°-13°) internal friction, 29.3% - 64.5% natural water content
and 1.652 – 1.972 Mg/m3 bulk density. The allowable bearing capacity of 50 kN/m
2, volume compressibility from 0.115
m2/MN to 0.666 m
2/MN, initial void ratio and consolidation coefficient of 0.779 - 1.381 and 2.7 m
2/year - 8.3 m
2/year
respectively on the pressure range of 0 to 400 kN/m2
and estimated settlement values of 114 to 273 mm were obtained for the
site materials. These results are indicative of soft to stiff clays and presence of sands and silts in the essentially clayey deposit.
The study area is thus underlain by extensive zone of ductile and low strength founding materials having medium to high
compressibility and settlement value that exceeds the tolerable limit suitable for founding a four storey building and should
therefore be avoided. These characteristics preclude the use of conventional shallow foundations, piles or vibro-replacement up
to a depth of 30 m.
Keywords: Resistivity, vibro-replacement, multi-stage triaxial, geotechnical, uncomformably
1. INTRODUCTION
The incessant incidence of building failures is
becoming alarming in Nigeria and Lagos metropolis
in particular and has led to loss of life and properties
worth millions of dollars. These failures have been
attributed to factors such as inadequate information
about the subsurface geological material, poor
foundation design and poor building materials. Prior
to the commencement of design of a construction
project, investigations are traditionally carried out in
line with existing guides and codes regarding the
property and quality of the proposed site. Such
investigation is carried out in order to avert structural
failures, as these failures could lead to disasters which
pose serious threats to public safety. The ultimate goal
of site investigation is to have appreciable
understanding of the behaviour of the soils that will
bear load to be transmitted by the proposed structure.
More often than not, site investigation in Nigeria is
achieved by use of traditional geotechnical methods
such as boring and cone penetration testing while
undermining the growing importance of the
geophysical methods (Soupios et al., 2007) for the
geotechnical site chatacterization.
Geotechnical tests usually reveal discreet
information about the subsurface. However, to obtain
a clearer picture, geophysical methods are essential to
establishing lateral and vertical variations between the
points under investigation. The need to address the
lateral variations informed the integration of
geophysical and geotechnical methods in this study. In
the last decade, the involvement of geophysics and
geotechnical methods in civil engineering has become
a promising approach (Adepelumi et al., 2009; Al
Omosh et al., 2008; Schoor, 2002; Adepelumi and
Olorunfemi, 2000). However, it should be noted that
the use of geophysical methods in site investigation is
intended to supplement geotechnical methods and not
to serve as substitutes for the drilling, sampling, test
pitting and in-situ laboratory testing (Rowe, 2001).
For economic reasons, boreholes cannot be placed
close enough to one another to give an accurate
picture of subsoil conditions. The role of geophysics
is usually to describe the properties and geometry of
the subsurface (Sheriff, 2002) and to provide data
Festus et al.
Investigation of a Proposed Four Storey Building Sites Using Geophysical and Laboratory Engineering Testing
Methods in Lagos, Nigeria
84
between borings and at the same time reduce the
number of boreholes. Thus, combination of
geophysical data and geotechnical measurements may
greatly improve the quality of construction in civil
engineering as it will focus on the behaviour and
performance of soils and rocks in the design and
construction of civil engineering structures (Oyedele
et al., 2009). In this study, geophysical and field/
laboratory geotechnical methods were integrated with
a view to determining the suitability of subsoil at a
proposed four storey building development sites in
Ebute Metta (near Iponri), Lagos Nigeria, which will
serve as a guide for the design of the foundation for
the proposed structure.
2. GEOLOGIC SETTING
The Dahomey Basin is a combination of
inland/coastal/offshore sedimentary basin in the Gulf
of Guinea (Obaje, 2009). The lithology based
stratigraphic classification of Dahomey basin by Jones
and Hockey (Brownfield and Charpentier, 2006) is
suitable for this study in that lithology is a key
parameter in determining suitability of materials for
engineering purposes. The Ewekoro Formation, which
conformably overlies the Abeokuta formation is
Palaeocene in age and consists of limestone, shale and
clay members. The Ilaro formation overlies the
Ewekoro Formation and is of Eocene age. It is
composed of poorly sorted sandstone with clay
fractions and subordinate shale. The Coastal Plains
Sands unconformably overlies the Ilaro Formation and
is Pleistocene to Oligocene in age. The lithology
consists essentially of sands, silts and clay deposits
with traces of peat in parts. It directly underlies the
study area and is composed of deposits which can be
divided into the littoral and lagoonal sediments of the
coastal belt and the alluvial sediments of the major
rivers. They essentially of consist of unconsolidated
sands, clays and mud with a varying proportion of
vegetative matter.
3. METHODOLOGY
3.1. Geophysical Survey
A multi - electrode 2- D resistivity survey was carried
along four 200m traverses spaced approximately 30 m
from each other (Figure 1) with sixty-four electrode
SAS 4000 Terameter. A 5 m inter-electrode spacing
Wenner array was utilized owing to its high sensitivity
to lateral in homogeneities to provide a good idea of
variation of materials in a continuum around the site.
The acquired 2D data was inverted using the software
package RES2DINV (Loke, 1997).
3.2. Borehole Drilling
Eleven bore holes, distributed along the four traverses
were drilled on the site to a maximum depth of 30 m.
The drilling was carried out employing the shell and
auger drilling method with a fully equipped motorized
Pilcon Wayfarer drilling rig. Samples were collected
for inspection, description and laboratory analysis.
Sampling and in situ tests were carried out
progressively with the advancement of the drilling
activity through the sediments from which a number
of geotechnical samples intended for triaxial
compression and oedometer testings were collected.
3.3. Undrained Multi-Stage Triaxial Compression
Testing
Triaxial Compression Test is a test in which a
cylindrical specimen of soil or rock encased in an
impervious membrane is subjected to a confining
pressure and then loaded axially to failure in
compression. The triaxial apparatus has been
described in great detail by Bishop and Henkel
(1962). To prepare a triaxial specimen, field samples
were removed from its plastic sleeve and trimmed to a
length of about 200 mm.
The multistage triaxial test (Kovari and Tisa,
1975), as specified in BS1377: part 8:1990 and
described in Head (1992), was carried out to measure
the shear strength parameters of soils namely cohesion
and internal friction angle. In this multi-stage element,
a single specimen was compressed at three effective
stress stages, rather than using the more familiar three
individual specimens. The reason for using the multi-
stage approach is that fewer samples require less time
in the field, and that issues of non-uniformity between
samples were removed.
3.4. Oedometer Consolidation Test
The specimens were loaded and unloaded in several
steps. At each loading stage the change of the height
was recorded at suitable intervals while consolidation
takes place. At the end of the test, the final dial gauge
readings were taken. After removing the dial gauge
and the top plate, the measurements of the final height
of the specimen were determined by the calipers.
Immediately after that, the free water was removed
from the soil surface and specimen weighed. The
water content and void ratio were then determined.
The compressibility and cohesive values were
compared with the guidelines proposed by Bell
(2007).
International Journal of Scientific Research in Knowledge, 2(2), pp. 83-91, 2014
85
4. RESULTS AND DISCUSSIONS
4.1. 2D Resistivity and borehole logs
Results from the resistivity surveys show that much of
the subsurface beneath the site is underlain by a
simple nearly horizontal stratification which is
constituted essentially of clayey deposits. The
resistivity sections AA' to DD' (Figures 2, 4, 6 and 8)
generally shows alternation of Peaty Clay (<10 Ωm),
Silty Clay (approx 10 - 40 Ωm ) and Silty Sand (>40
Ωm) that completely submerged in water. These
materials are viewed as incompetent engineering
materials. Similar lithologies were delineated by
Oyedele et al. (2012), Adepelumi et al. (2009) and
Adepelumi and Olorunfemi (2000) using resistivity
measurements around the Lagos metropolis.
Considering the borehole lithology logs, two sets of
logs were observed. The first set having
predominantly clayey materials from ground surface
to the terminal depths of the boreholes at 30 m, while
the second set have more sandy contents within the
uppermost 10 m (Figures 3, 5, 7 and 9).
Fig. 1: Base Map of the Study Area.
Fig. 2: 2-D resistivity section along traverse AA'
Festus et al.
Investigation of a Proposed Four Storey Building Sites Using Geophysical and Laboratory Engineering Testing
Methods in Lagos, Nigeria
86
Fig. 3: Sections of Borehole Logs along traverse AA'
Fig. 4: 2-D resistivity section along traverse BB'
BH3 along traverse AA' (Figure 2) shows the
presence of silty sand between 4.5 m to 10 m which
was confirmed by the 2-D resistivity section. Though,
the 2 D shows that this layer is laterally continuous
but the resistivity of the silty sand greatly fluctuates
westward suggesting inhomogeneity (e.g at 55 – 90 m,
120 m) of the sandy material. This inhomogeneity
suggests changes in the quality and integrity of the
sandy material, implying unsuitability of the sandy
material for engineering foundation of a four storey
building.
Along traverse BB' (Figure 4), a BH6 at 40 m
mark encountered sandy materials at depth of about 9
m, indicating the presence of good founding medium
at that depth. However, the 2-D resistivity section
shows that the sandy medium which dipped steeply
easterly was replaced by peaty clay at 9 m so that it
was encountered at about 20 m. This is inimical to
engineering foundation.
Along traverse CC' (Figure 6), a borehole
drilled at lateral distance 90 m from the start of the
traverse (BH 8) appears to be underlain by sandy
materials from a depth of about 20 m, giving a wrong
notion of presence of good founding medium at that
depth. However, the 2-D resistivity section shows that
the sand medium is localized with much of the
subsurface beneath that traverse being constituted
essentially of clay materials.
Beneath traverse DD' (Figure 8), BH 10 and
BH 11 encountered peaty clay and silty sand
materials respectively at the shallow depth which are
in turn underlain by sandy/silty clay. The 2-D section
along this traverse shows that the sandy materials at
shallow depth beneath BH 11 is localized and lacks
lateral continuity and thus incongruous to bear
uniform load. The lateral continuity of silty sand at
about 25 m depth cannot be ascertained due to
discontinuity of data sets.
International Journal of Scientific Research in Knowledge, 2(2), pp. 83-91, 2014
87
Fig. 5: Sections of Borehole Logs along traverse BB'
Fig. 6: 2-D resistivity section along traverse CC'
Fig. 7: Sections of Borehole Logs along traverse BB'.
Festus et al.
Investigation of a Proposed Four Storey Building Sites Using Geophysical and Laboratory Engineering Testing
Methods in Lagos, Nigeria
88
Fig. 8: 2-D resistivity section along traverse DD'
4.2. Laboratory Results
Laboratory testing results obtained from multistage
triaxial compression and oedometer consolidation
tests are shown in tables 1 and 2 respectively.
4.3. Undrained Triaxial Compression Tests
Following the triaxial compression tests carried out on
retrieved samples from the boreholes, strength
parameters were obtained. A range of cohesion
values of 35 kN/m2 - 75 kN/m
2 was obtained for these
samples which are indicative of soft, firm to stiff
clays. The values of angles of internal friction (5°-
13°) is quite high for clayey deposits but can be
attributed to the presence of sands and silts in the
essentially clayey deposit. These strength parameters
typify low strength founding materials up to 30m
which are unsuitable for a four storey building. This is
at variance with Oyedele et al (2011) which
established the presence of competent materials at
16m depth in the southeastern part of the study area.
The natural water content of the samples ranges from
29.3% - 64.5%. These high values are due to
submergence of all materials in water because
groundwater level was encountered at 0.10 m from the
boreholes. This factor will contribute to the weakness
of the subsurface materials. Furthermore, considering
the essentially clayey nature of the subsurface
materials, settlement rate for structures placed on such
high water content material is expected to be high.
The bulk density result varies from 1.652 – 1.972
Mg/m3 which pinpoint a slightly compacted clayey
material.
International Journal of Scientific Research in Knowledge, 2(2), pp. 83-91, 2014
89
Table 1: Result of Multistage Triaxial Compression Test.
Table 2: Result of Oedometer Consolidation Test
Considering the cohesion and angle of internal
friction obtained from triaxial tests, the computed net
allowable bearing capacity of the soils within the
uppermost 1.2 m was found to be 50 kN/m2 without
using a safety factor of 3, this is grossly inadequate
for a four-floor building with an estimated load of 50
kN/m2.
4.4. Oedometer Consolidation Test
This coefficient of volume compressibility varies
between 0.115 m2/MN and 0.666 m
2/MN an indicative
of medium to high compressibility. Therefore such
materials will be unable to bear a four storey building.
The initial void ratio and consolidation coefficient
varies between ranges from 0.779 - 1.381 and 2.7
m2/year - 8.3 m
2/year on the pressure range of 0 to
400 kN/m2 respectively. Although settlement in
clayey materials may be slow because they drain
slowly, the settlement (subsidence) in the study area
will be eventually be large due to high initial void
ratio and consolidation coefficient. The estimated
settlement values obtained for the proposed four-floor
building using typical single and double wings range
from 80 to 192 mm and 114 to 273 mm respectively.
Festus et al.
Investigation of a Proposed Four Storey Building Sites Using Geophysical and Laboratory Engineering Testing
Methods in Lagos, Nigeria
90
These settlement values are far higher than the
tolerable limit of 50 mm for raft (shallow) foundation.
5. CONCLUSIONS
Geophysical method integrated with field and
laboratory geotechnical testing have shown that
electrical resistivity correlate well with engineering
strength parameters of the subsurface soils of a
proposed four storey building site. 2-D resistivity
sections proved useful in providing information
continuity between borings, depth and position of
changes in strata over large area and reduced the
number of boreholes necessary. The sections revealed
incompetent founding materials beneath the study
area. Laboratory tests conducted on the materials
obtained from boreholes at the site indicated materials
of grossly inadequate net allowable bearing capacity
and estimated settlement values that are higher than
the tolerable limit for the proposed four-floor
building. Therefore, the results of the various
investigations conducted in the study area prohibit the
use of conventional shallow foundation, piles or
vibro-replacement up to a depth of 30 m. It is
recommended that, for the proposed load, further
investigations be carried out by drilling more test
boreholes beyond 30 m with a view to establish the
strata with adequate bearing capacity.
REFERENCES
Adepelumi AA, Olorunfemi MO, Falebita DE,
Bayowa, OG (2009). Structural mapping of
coastal plain sands using engineering
geophysical technique: Lagos Nigeria Case
Study. Natural Science, 1: 2-9.
Adepelumi AA, Olorunfemi MO (2000). Engineering
geological and geophysical investigation
investigation of the reclaimed Lekki Peninsula,
Lagos, Southwest Nigeria. Bulletin of
Engineering, Geology and the Environment, 58:
125-132.
Bell FG (2007). Engineering Geology. Second
Edition. Elsevier Ltd. Oxford, U.K. 222, 223.
Bishop AW, Henkel DJ (1962). The measurements of
soil properties in the triaxia test. 2nd
edition
Edward Arnold (Publishers) LTD., London.
Brownfield ME, Charpentier RR (2006). Geology and
total petroleum systems of the West Central
Coastal Province (7203), W/Africa: U.S.
Geological Survey Bulletin 2207, 52 p.
BS1377 (1990) Method of test for soils for civil
engineering purposes. British Standards
Institution, London.
Head KH (1992). Manual of soil laboratory testing.
Vol. 3, effective stress tests, Wiley.
Kovari K, Tisa A, Einstein H, Franklin JA (1983).
Suggested methods for determining the
strength materials in triaxial compression. Int. J.
of Rock Mech. & Min. Sci. & Geomechs Abs.,
20: 283-290.
Loke MH (1997). RES2DINV ver. 3.3 for Windows
3.1, 95 and NT Advanced Geosciences Inc. 66
Obaje NG (2009). Geology and Mineral Resources of
Nigeria, Lecture Notes in Earth Sciences.
221p, Springer Dordrecht Heidelberg London
New York.
Oyedele K.F, Ayolabi EA, Adeoti L, Adegbola RB
(2009). Geophysical and Hydrogeological
Evaluation of Rising Groundwater Level in the
Coastal Areas of Lagos, Nigeria. Bull. Eng.
Geol. Environ. , 68: 137 - 143.
Oyedele KF, Oladele S, Adedoyin O (2011).
Application of Geophysical and Geotechnical
Methods to Site Characterization for
Construction Purposes at Ikoyi, Lagos, Nigeria.
Journal of Earth Sciences and Geotechnical
Engineering, 1(1): 87-100
Oyedele KF, Oladele S, Okoh C (2012). Geoassement
of Subsurface Conditions In Magodo Brook
Estate, Lagos Nigeria. International journal
of advanced scientific and technical research,
2(4): 731-741
Rowe RK (2001). Geotechnical and
Geoenvironmental Engineering Handbook,
Kluwer Academic Publishing, Norwell,
Mass., USA. 82.
Soupios P, Georgakopoulos P, Papadopoulos N,
Saltas V, Vallianatoss F, Sarris A, Makris J
(2007). use of engineering geophysics to
investigate a site for a building foundation. J.
Geophys. Eng, 4: 94-103.
Sheriff RE (2002). Encyclopedic Dictionary of
Applied Geophysics, Fourth Edition. The
Society of Exploration Geophysicists (S.E.G)
Tulsa OK USA. 323.
International Journal of Scientific Research in Knowledge, 2(2), pp. 83-91, 2014
91
DR Oyedele Kayode Festus, He is an Associate Professor. He is interested in Groundwater Exploration,
Geotechnical Investigation, Environmental Pollution, Petroleum Geophysics researches.
Dr. Lukumon Adeoti, He is a Senior Lecturer. He received his BS degree in Applied Physics-
Geophysics, 1997. He received his MSc degree in Exploration Geophysics, 2000. Also He received PhD
degree in Geophysics, 2007. He is interested in Exploration Geophysics / Borehole Geophysics.
Oladele Sunday, He is an Assistant Lecturer. He received MSc degree in Applied Geophysics. He is
interested in Petroleum geophysics, forensic geophysics, Earth imaging and modeling.
International Journal of Scientific Research in Knowledge, 2(2), pp. 92-104, 2014
Available online at http://www.ijsrpub.com/ijsrk
ISSN: 2322-4541; ©2014 IJSRPUB
http://dx.doi.org/10.12983/ijsrk-2014-p0092-0104
92
Full Length Research Paper
Nutritional and Anti-Nutritional Composition of Bridelia Ferruginea Benth
(Euphorbiaceae) Stem Bark Sample
Adesina Adeolu Jonathan1*
, Akomolafe Seun Funmilola2
1Department of Chemistry, Ekiti State University, PMB 5363, Ado Ekiti, Nigeria
2Department of Biochemistry, Ekiti State University, PMB 5363, Ado Ekiti, Nigeria
*Corresponding Author: Email: [email protected]
Received 06 December 2013; Accepted 11 January 2014
Abstract. Nutritional composition of Bridelia ferruginea Benth (Euphorbiaceae) stem bark sample was evaluated. The
percentage protein, fat, fibre and carbohydrate contents were: 15.7 ± 0.30, 5.45± 0.05, 4.35 ± 0.06 and 60.7 ± 0.71 respectively
while the total gross energy was 1501kJ/100g. The mineral composition ranged from 0.25 to 74.2 mg/100g, with phosphorus
being the most concentrated. The K/Na ratio (1.80) was higher than 1.0 recommended. The mineral safety index computation
showed only Zn to be in excess based on the recommended daily allowance (RDA). The amino acid contents ranged between
0.597 – 13.1 g/100g. All the essential amino acids were present in varying amount with Isoleucine being the most
concentrated. The % TEAA and % TNEAA were: 47.8 and 52.2 respectively. The P-PER (predicted protein efficiency ratio),
pI (Isoelectric point) and EAAI (essential amino acid index) values were: 1.69, 5.14, 1.18 respectively. Based on the whole
hen’s egg amino acid scoring pattern, methionine was limiting while with respect to FAO/WHO provisional amino acid
scoring pattern and essential amino acid scoring pattern based on the requirements of pre-school child Met + Cys was limiting.
The anti-nutritional factors analyzed; phytates, tannins, oxalates, phenolic content, saponins, phytin phosphorus and alkaloids
in the sample were lower than the range of values reported for most vegetables. This study revealed that the Bridelia
ferruginea stem bark consumed in Ekiti State and other states in the South-western part of Nigeria can contribute useful
amount of nutrients to human diet.
Keywords: Nutritional, anti-nutritional composition, Bridelia ferruginea stem bark
1. INTRODUCTION
Forage trees and shrubs play an essential and multiple
roles in the balance of the Sahelian and Sudanian
ecosystems exploited by man and his animals. This
role becomes more important as the dry season grows
longer, and decreases as the mean annual rainfall
increases. It therefore grows less important from north
to south according to the rainfall gradient, which is
about 1 mm per km, or 110 mm per each degree of
latitude. Throughout West Africa, especially in areas
prone to drought, previous studies demonstrate the
importance of edible wild plants as food sources
(Grivetti et al., 1987; Sena et al., 1998). Commonly,
drought is associated with inadequate food intake and
disease, where food scarcity and inadequate dietary
intakes clearly have led to increased incidence of
malnutrition and famine (Franke and Chasin, 1980).
Use of edible wild species in combination with
domesticated foods has remained a hallmark of many
African agro-pastoral societies (Grivetti, 1978, 1979;
Grivetti et al., 1987). Traditional medicine practiced
in rural Nigeria, reveals well-documented uses of
plant barks and bark extracts, fruits, leaves, nuts and
seeds, and tubers, but with few exceptions, dietary-
medical uses of edible wild plants as components of
West African traditional medicine have not been
widely documented (Ogugbuaja et al., 1997).
Bridelia ferruginea Benth (Euphorbiaceae) is a
medicinal plant that is widely used in African
folkloric medicine. In Nigeria, it is commonly called
Kirni, Kizni (Hausa); Marehi (Fulani); Iralodan
(Yoruba), Ola (Igbo), Kensange abia (Boki). Its
habitat is the savannah especially in the moister
regions extending from Guinea to Zaire and Angola.
The tree is 6-15m high, up to1.5m in girth and bole
crooked branching low down. The bark is dark grey,
rough and often markedly scaly (Kolawole and
Olayemi, 2003). The stem bark decoction is used in
African traditional medicine to treat diarrhea,
dysentery and gynaecological disorders (including
sterility). A decoction of the leaves is used to treat
diabetes. It has even been evaluated for antimalarial
(Kolawole and Adesoye, 2010), antimicrobial
Jonathan and Funmilola
Nutritional and Anti-Nutritional Composition of Bridelia Ferruginea Benth (Euphorbiaceae) Stem Back Sample
93
(Kareem et al., 2010), analgesic (Akuodor, et al.,
2011) and antidiabetic (Bakoma et al., 2011)
activities. B. ferruginea bark is used for treatment of
bacterial infections on wounds (Irobi et al., 1994). A
decoction of the leaves is used as a purgative and also
in the treatment of diabetes (Cimanga et al., 1999).
Roots and leaves extracts are used to cure piles,
diarrhea and dysentery (McNeely, 1990) and also
confirmed for anti-inflammation activities (Olajide et
al., 1999). Galocatechin has been isolated from the
bark (De Bruyne et al., 1997).
In western Nigeria, B. ferruginea is used as a
mouthwash and remedy for candidal oral thrush;
whereas in Northern Nigeria, the bark is used for
treatment of infections caused by poisoned arrow
wounds (Irobi et al., 1994). A decoction of the bark
extract has also been proven to have antibacterial
effect (Kolawole et al., 2006). It is also used as
purgative and a vermifuge. A macerated extract of the
bark is used in Northern Nigeria to harden beaten
laterite and mud floors (Kolawole and Olayemi,
2003). The chemical constituents of Bridelia
ferruginea have not been thoroughly examined. The
present study is therefore aimed at exploiting the
proximate, minerals, anti-nutritional and amino acid
profile of the back of Bridelia ferruginea as this
would promote its dietary-medical uses.
2. MATERIAL AND METHODS
2.1. Collection and preparation of samples
The sample was collected from local farms around
Iworoko- Ekiti, Irepodun-Ifelodun local Government
area of Ekiti State. It was properly sorted, washed,
dried, milled into fine powdered form and kept in an
air-tight plastic bottle prior to analysis.
2.2. Proximate analysis
Moisture, total ash, fiber and ether extract of the
samples were determined by the methods of the
AOAC 2005. Nitrogen was determined by a micro-
Kjeldahl method and the crude protein content was
calculated as N x 6.25 (Pearson, 1976). Carbohydrate
was determined by difference. All the proximate
results were reported in g/100 g dry weight. The
energy values obtained for carbohydrates (x 17 kJ),
crude protein (x 17 kJ) and crude fat (x 37 kJ) for each
of the samples. Determinations were in duplicate.
Table1: Proximate and some calculated parameters in the sample of Bridelia ferruginea stem bark
PEP = Proportion of total energy due to protein
PEF = Proportion of total energy due to fat PEC = Proportion of total energy due to protein
UEDP = Utilizable energy due to protein
2.3. Mineral analysis
The mineral elements were determined in the
solutions obtained above-Na and K by flame
photometry, Model 405 (Corning, Halstead Essex,
UK) using NaCl and KCl to prepare standards.
Minerals were analysed using the solutions obtained
by dry ashing the samples at 550 oC and dissolving it
in 10 % HCl (25 ml) and 5 % lanthanum chloride (2
ml), boiling, filtering and making up to standard
volume with deionized water. Phosphorus was
determined colorimetrically using a Spectronic 20
(Gallenkamp, London, UK) instrument, with KH2PO4
as a standard. All other elements (Ca, Mg, Zn, Fe, Mn,
Cu and Cr) were determined by atomic absorption
spectrophotometry, Model 403 (Perkin-Elmer,
International Journal of Scientific Research in Knowledge, 2(2), pp. 92-104, 2014
94
Norwalk, Connecticut, USA). All determinations were
made in duplicate. All chemicals used were of
analytical grade, and were obtained from British Drug
House (BDH, London, UK).
The detection limits for the metals in aqueous
solution had been determined just before the mineral
analyses using the methods of Varian Techtron, giving
the following values in µg/ml: Fe (0.01), Cu (0.002),
Na (0.002), K(0.005), Ca(0.04), Mg(0.002), Zn
(0.005), Mn (0.01) and Cr (0.02) (Varian Techtron,
1975). The optimal analytical range was 0.1 to 0.5
absorbance units with coefficients of variation from
0.9-2.2 %.
The coefficients of variation per cent were
calculated (Steel and Torrie, 1960). The percentage
contribution to energy due to protein (PEP), due to
total fat (PEF) and due to carbohydrate (PEC) as PEP
%, PEF % and PEC % respectively were calculated.
The percentage utilizable energy due to protein
(UEDP %) was also calculated. Ca/P, Na/K, Ca/Mg
and the millequivalent ratio of [K/(Ca +Mg)]; the
mineral safety index (MSI) of Na, Mg, P, Ca, Fe and
Zn were also calculated (Hathcock, 1985). To
calculate MSI, we have: RAI is recommended adult
intake; CV in the Table will represent calculated value
(CV) of calculated MSI from research results. The
differences between the standard MSI and the MSI of
the samples were also calculated.
2.4. Determination of Anti-nutritional factors
2.4.1. Determination of tannin
200mg of the sample was weighed into a 50ml sample
bottle. 10ml of 70% aqueous acetone was added and
properly covered. The bottles were put in an orbital
shaker and shaken for 2 hours at 300C. Each solution
was then centrifuged and the supernatant stored in ice.
0.2ml of each solution was pipetted into test tubes and
0.8ml of distilled water was added. Standard tannic
acid solutions were prepared from a 0.5mg/ml stock
and the solution made up to 1ml with distilled water.
0.5ml folin reagent was added to both sample and
standard followed by 2.5ml of 20% Na2CO3. The
solutions were then vortexed and allowed to incubate
for 40 minutes at room temperature after which
absorbance was red against a reagent blank
concentration of the sample from a standard tannic
acid curve (Makkar and Goodchild, 1996).
Table 2: Composition and some calculated mineral ratios in the Bridelia ferruginea bark sample
*milliequivalent ratio
2.4.2. Determination of oxalate
1g of the sample was weighed into 100ml conical
flask. 75ml of 1.5N H2SO4 was added and the solution
was carefully stirred intermittently with a magnetic
stirrer for about 1 hour and then filtered using
Whatman filter paper. 25ml of sample filtrate was
collected and titrated hot (80-900C) against 0.1N
KMnO4 solution to the point when a faint pink colour
appeared that persisted for at least 30 seconds (Day
and Underwood, 1986).
2.4.3. Determination of alkaloid
Alkaloid determination was carried out following the
procedure of Harborne(Harborne, 1973). 5.0g of the
Jonathan and Funmilola
Nutritional and Anti-Nutritional Composition of Bridelia Ferruginea Benth (Euphorbiaceae) Stem Back Sample
95
sample was weighed into a 250ml beaker and 200ml
of 10% acetic acid in ethanol was added and covered
and allowed to stand for 4h. This was filtered and the
extract was concentrated on a water bath to one
quarter the original volume. Concentrated ammonium
hydroxide was added drop wise to the extract until the
precipitation was complete. The whole solution was
allowed to settle and the precipitate was collected and
washed with dilute ammonium hydroxide and then
filtered. The residue is the alkaloid which was dried
and weighed.
2.4.4. Determination of saponin
The method used was that of Obadoni and Ochuko,
2001). 5g of the sample was put into a conical flask
and 100cm3 of 20% aqueous ethanol were added. The
sample was heated over a hot water bath for 4h with
continuous stirring at about 550C. The mixture was
filtered and the residue re-extracted with another
200ml 20% ethanol. The combined extracts were
reduced to 40ml over water bath at about 900C. The
concentrate was transferred into a 250ml separating
funnel and 20ml of diethyl ether was added and
shaken vigorously. The aqueous layer was recovered
while the ether layer was discarded. The purification
process was repeated. 60ml n-butanol was added. The
combined nbutanol extracts were washed twice with
10ml of 5% aqueous sodium chloride. The remaining
solution was heated in a water bath after evaporation;
the sample was dried in the oven to a constant weight.
The saponin content was calculated as percentage.
2.4.5. Determination of flavonoid
The method of Boham and Kocipai-Abyazan (Boham
and Kocipai-Abyazan, 1974) was followed in the
determination of flavonoid. 5g of the sample was
extracted repeatedly with 100ml of 80% aqueous
methanol at room temperature. The whole solution
was filtered through whatman filter paper (125ml).
The filtrate was later transferred into a crucible and
evaporated into dryness and weighed to a constant
weight.
2.4.6. Oxalate determination
The titration method as described by Day and
Underwood (1986) was followed. 1g of sample was
weighed into 100 ml conical flask. 75 ml 3M H2SO4
was added and stirred for 1 h with a magnetic stirrer.
This was filtered using a Whatman No 1 filter paper.
25 ml of the filtrate was then taken and titrated while
hot against 0.05 M KMnO4 solution until a faint pink
colour persisted for at least 30 s. The oxalate content
was then calculated by taking 1ml of 0.05 M KMnO4
as equivalent to 2.2 mg oxalate (Chinma and Igyor,
2007; Ihekoronye and Ngoddy, 1985).
2.4.7. Phytate content determination
This was determined by the method of Wheeler and
Ferrel (1971).100 ml of the sample was extracted with
3% trichloroacetic acid. The extract was treated with
FeCl3 solution and the iron content of the precipitate
was determined using Atomic Absorption
spectrophotometer (Pye Unicam 2900). A 4:6 Fe/P
atomic ratio was used to calculate the phytic acid
content (Okon and Akpanyung, 2005). Phytin
phosphorus (Pp) was determined and the phytic acid
content was calculated by multiplying the value of Pp
by 3.55 (Young and Greaves, 1940). Each milligram
of iron is equivalent to 1.19 mg of Pp.
Phytin phosphorus as percentage of phosphorus (Pp %
P) = Pp/P × 10
2.5. AMINO ACID ANALYSIS
The amino acid profile was determined using the
method described by Sparkman et al. (1958). Each
sample was dried to constant weight, defatted,
hydrolyzed, evaporated and loaded into the techno
sequential multi-sample amino acid analyzer (TSM).
Following the steps described below: 2 g of the dried
sample was weighed into extraction thimble and the
fat extracted with chloroform: methanol (2:1) mixture
using soxhlet extraction apparatus (AOAC, 2005).
Then, 1 g of the defatted sample was weighed into
glass ampoule. 7 ml of 6 N HCl were added and
oxygen expelled by passing nitrogen into the ampoule.
The glass ampoule was sealed and placed in an oven
preset at 1050C for 22 h. The ampoule was allowed to
cool before breaking open at the tip and the content
filtered. The filterate was then evaporated to dryness
and the residue dissolved with 5 ml acetate buffer (pH
2.0) and stored in plastic specimem bottles. 10 μl was
dispensed into the cartridge of the analyser which is
designed to seperate andanalyse free, acidic, neutral
and basic amino acids of the hydrolysate. The amount
of each amino acid present in the sample was
calculated in g/100 g protein from the chromatogram
produced.
International Journal of Scientific Research in Knowledge, 2(2), pp. 92-104, 2014
96
Table 3: Mineral safety index of Na, Mg, P, Ca, Fe, and Zn for the Bridelia ferruginea stem bark sample
TV = table value, CV = calculated value
D = difference (TV - CV)
Table 4: Anti-nutritional content of the Bridelia ferruginea stem bark samples
2.5.1. Determination of quality parameters
2.5.1.1. Determination of amino acid scores
Determination of the amino acid scores was fir
st, based on whole hen’s egg (Paul et al., 1976). In
this method, both essential and nonessential amino
acids were scored. Secondly, amino acid score was
calculated using the following formula (FAO/WHO,
1973):
Amino acid score = (amount of amino acid per test
protein (mg/g)) / (amount of amino acid per protein in
reference pattern (mg/g)).
In this method, Met + Cys and Phe + Tyr were
each taken as a unit. Also, only essential amino acids
determined were scored. Amino acid score was also
calculated based on the composition of the amino
acids obtained in the samples compared with the
suggested pattern of requirements for pre-school
children (2-5 years). Here, Met + Cys and Phe + Tyr
were each taken as a unit. Also, only essential amino
acids with (His) were scored.
2.5.1.2. Determination of the essential amino acid
index
The essential amino acid index (EAAI) was calculated
by using the ratio of test protein to the reference
protein for each of the eight essential amino acid plus
histidine in the equation that follows (Steinke et al.,
1980):
Methionine and cystine are counted as a single amino
acid value in the equation, as well as Phe + Tyr.
2.5.1.3. Determination of the predicted protein
efficiency ratio
The predicted protein efficiency ratio (P-PER) was
determined using one of the equations derived by
Alsmeyer et al. (1974), i.e.
P-PER = –0.468 + 0.454 (Leu) – 0.105 (Tyr).
2.5.1.4. Other determinations
Determination of the total essential amino acid
(TEAA) to the total amino acid (TAA), i.e.
(TEAA/TAA); total sulphur amino acid (TSAA);
percentage cystine in TSAA (% Cys/TSAA); total
aromatic amino acid (TArAA), etc; the Leu/Ile ratios
were calculated while the isoelectric point (pI) was
calculated using the equation of the form
(Olaofe and Akintayo, 2000):
Jonathan and Funmilola
Nutritional and Anti-Nutritional Composition of Bridelia Ferruginea Benth (Euphorbiaceae) Stem Back Sample
97
pIm IPiXi
Where pIm is the isoelectric point of the mixture of
amino acids, IPi is the isoelectric point of the ith amino
acid in the mixture and Xi is the mass or mole fraction
of the ith amino acid in the mixture (Finar, 1975).
Table 5: Amino acids profile of the Bridelia ferruginea
bark samples (g/100g cp)
3. RESULTS AND DISCUSSION
The results of proximate composition of the Bridelia
ferruginea back sample are shown in Table 1. The
crude protein content (15.7 ± 0.30 %) was comparably
higher than the values reported for some vegetables
consumed in Nigeria: 5.91 % (Cnidoscolus
chayamansa), 3.31 % (Solanium nodiflorum), 3.03 %
(Senecio biafrae) (Adeleke and Abiodun, 2010);
4.60% (A. hydridus), 4.30 % (T. occidentallis)
(Fafunso and Basir, 1977) but favourably compared
with the values reported for C. maxima (18.6 %), A.
viridis (19.2 %) and B. alba (18.0%) (Adesina, 2013)
and S. indicum (18.59%), B.aegypiaca (15.86 %)
(Kubmarawa et al., 2008).
The ash content (6.54 %) of Bridelia ferruginea
back sample is an indication of the levels of minerals
or inorganic component of the sample. These minerals
act as inorganic co-factors in metabolic processes
which mean in the absence of these inorganic co-
factors there could be impaired metabolism
(Iheanacho and Udebuani, 2009). Table 1 still
contains other parameters calculated from the
proximate values. It shows the various energy values
as contributed by protein, fat and carbohydrate. The
daily energy requirement for an adult is between
2500-3000 kCal (10455-12548 kJ) depending on his
physiological state while that of infants is 740 kCal
(3094.68 kJ) (Bingham, 1978). This implies that while
an adult man would require between 590-709 g
(taking the calculated energy of 1501 kJ/100 g) of his
energy requirement, infants would require 174.9 g
(taking the calculated energy of 1501 kJ/100 g). On
the whole this meant that samples with higher energy
value would require lower quantity of sample to
satisfy the energy needs of man and infants. The
utilizable energy due to protein (UEDP %) for the
sample (assuming 60 % utilization) 10.7 %. The
UEDP % compared favourably with the recommended
safe level of 8 % for an adult man who requires about
55 g protein per day with 60 % utilization. The PEF %
values were generally low in the sample (13.4 %) and
far below the recommended level of 30 % (NACNE,
1983) and 35 % (COMA, 1984) for total fat intake;
this is useful for people wishing to adopt the
guidelines for a healthy diet.
Table 6: Concentrations of essential, non-essential, acidic,
neutral, sulphur, aromatic, basic, etc. (g/100g crude protein)
of Bridelia ferruginea stem bark samples (dry matter of
sample)
International Journal of Scientific Research in Knowledge, 2(2), pp. 92-104, 2014
98
Table 2 gives the list of the nutritionally important
minerals as well as the computed mineral ratios in
Bridelia ferruginea back sample. Minerals are
important in human nutrition. It is well known that
enzymatic activities as well as electrolytic balance of
the blood fluid are related to the adequacy of Na, K,
Mg and Zn. Potassium is very important in
maintaining the blood fluid volume and osmotic
equilibrium. Metal deficiency syndrome like rickets
and calcification of bone is caused deficiency.
Appreciable levels of all the essential minerals were
present in the the sample. The sample was apparently
high in phosphorus (74.2 ± 0.105 mg/100g), a value
which was comparably higher than what was reported
for F. asperifolia and F. sycomorus (Nkafamiya et al.,
2010). The levels of K, Na, Ca and Mg were
comparably higher than Mn and Cu levels. The Ca/P
(0.326) was comparably lower than 0.5 which is the
minimum ratio required for favourable calcium
absorption in the intestine for bone formation
(Nieman et al., 1992) although the level of Ca/P has
been reported to have some effects on calcium in the
blood of many animals (Adeyeye et al., 2012). The
value of ratio (0.556) was lower than 0.6, the value
that favours non-enhancement of high blood pressure
disease in man. Although for normal retention of
protein during growth and for balancing fluid a K/Na
ratio of 1.0 is recommended (Helsper et al., 1993), the
high value of K/Na ratio (1.80) obtained in the present
report suggests that bringing the ratio down would
require the consumption of food sources rich in Na.
the Ca/Mg value obtained for the present sample
(1.14) was fairly higher than the 1.0 recommended. It
means both that both Ca and Mg would need
adjustment for normal for normal health.
The milliequivalent ratio of [K/(Ca+Mg)] (1.31)
was comparably lower than 2.2 recommended, which
means the sample would not promote
hypomagesaemia in man (NRC, 1989; Adeyeye and
Adesina, 2012). Iron and Zinc are among the required
elements for humans and their daily requirements for
adults are 10 and 15 respectively. Levels obtained in
the present report (4.55 ± 0.02 mg/100g) (Fe) and 25.7
mg/100g (Zn)) compared favourably with the values
reported for F. asperifolia and F. sycomorus
(Nkafamiya et al., 2010). However zinc requirements
can easily be met by consuming this sample (25.7
mg/100g). Generally from the recommendation set out
by NRC/NAS, the daily requirement of Zn, Mn and
Cu can easily be met while the diets may need be
supplemented with foods high in K, Na, Ca and P.
The mineral safety index (MSI) values of the
sample are shown in Table 3. The standard MSI for
the elements are Na (4.8), Mg (15), P (10), Ca (10), Fe
(6.7) and Zn (33). For Ca, P, Mg, Fe, Cu and Na, the
MSI values ranged from 0.16 – 2.75, with all the
differences between the standard and calculated MSI
values being positive.
Let us take Na for an example, the calculate MSI
value is 0.16 and the difference is +4.64, this meant
that no amount of the sample might be overloading
the body with sodium that can lead to secondary
hypertension. For Ca, Mg and P all the calculated MSI
were lower than standard MSI and hence within the
USRDA (Hathcock, 1985). For Zn, the odd sample
out, was 56.5 and the difference was -23.5. The
implication of the above is that abnormally high level
Zn was present in the sample. The sample could cause
the reduction of Zn absorption in the small intestine in
children. The Zn MSI greater than 33 are above the
recommended adult intake. The minimum toxic dose
is 500 mg, or 33 times the RDA (Hathcock, 1985).
High doses of Zn can be harmful. Zinc supplements
can decrease the amount of high density lipoprotein
(HDL) circulating in the blood, increasing risk of
heart disease. Excess Zn interacts with other minerals,
such as Cu and Fe, decreasing their absorption. In
animals, Zn supplements decrease the absorption of
Fe so much that anaemia is produced (Adeyeye et al.,
2012).When patients are given 150 mg of Zn per day,
Cu deficiency results. Intakes of Zn only 3.5 mg/day
above the RDA decrease Cu absorption (Nieman et
al., 1992). In animals, Cu deficiency causes scarring
of the heart muscle tissue and low levels of Ca in the
bone (Adeyeye et al., 2012).
The antinutrient content of the sample are listed in
Table 4. These are compounds that limit the wide use
of many plants due to their ubiquitous occurrence of
them as natural compounds capable of eliciting
deleterious effect in man and animals (Kubmarawa et
al., 2008).
The antinutrient factors; oxalate, tannin, saponin,
phytate, alkaloids, phenolic content, phytin
phosphorus and flavonoids were present in varying
amounts in the sample.
These anti nutritional factors tend to bind to
mineral elements there by forming indigestible
complex (Nkafamiya and Manji, 2006). Oxalate for
instance tends to render calcium unavailable by
binding to the calcium ion to form complexes
(calcium oxalate crystals). These oxalate crystal
formed prevent the absorption and utilization of
calcium. The calcium crystals may also precipitate
around the renal tubules thereby causing renal stones
(Ladeji et al., 2004; Nkafamiya and Manji, 2006). In
general the levels of antinutrients in Bridelia
ferruginea bark sample are low to significantly
interfere with nutrients utilization. They are below the
established toxic level (Nkafamiya and Manji, 2006).
The amino acid profile of the Bridelia ferruginea
back sample is shown in Table 5. The levels of the
Jonathan and Funmilola
Nutritional and Anti-Nutritional Composition of Bridelia Ferruginea Benth (Euphorbiaceae) Stem Back Sample
99
amino acid ranged between 0.597 ± 0.001 – 13.1±
0.450 g/100g.
All the essential amino acids were present in the
sample, with Isoleucine having the highest
concentration (6.20 ± 0.014 g/100g).
Table 6 shows the concentrations of total AA
(TAA), total essential AA (TEAA), total acidic AA
(TAAA), total neutral AA (TNAA), total sulphur AA
(TSAA), total aromatic AA (TArAA) and their
percentage values. The Leu/Ile ratios, their differences
and percentage differences are contained in Table 6.
Non- essential amino acids have the highest %
concentration (52.2) while TEAA total essential
amino acids had a % concentration of 47.8. The
content of TEAA of 43.1 g/100g crude protein was
close to the value for egg reference protein (56.6
g/100g cp) (Paul et al., 1976); comparably close to the
values reported for soya bean (44.4 g/100g cp)
(Altschul, 1958).
The TAA in the current report was 90.1 g/100g cp,
this value was comparably close to the values of
reported for the dehulled African yam bean (AYB)
(70.3 – 91.8 g/100g cp) (Adeyeye, 1997) but
comparably higher than the values reported for raw
and processed groundnut seeds (Adeyeye, 2010). The
content of TSAA (1.42 g/100g) was lower than the 5.8
g/100g cp recommended for infants
(FAO/WHO/UNU, 1985) The ArAA range suggested
for ideal infants protein (6.8 – 11.8 g/100g cp )
(FAO/WHO/UNU, 1985), the present report has its
value better than the minimum, i.e 8.96 g/100g cp.
The ArAA are the precursors of epinephrine and
thyroxin (Robinson, 1987). The % ratios of TEAA/
TAA in the sample was 47.8 % which was well above
39 % considered to be adequate for ideal protein food
for infants ; 26 % for children and 11 % for adults
(FAO/WHO/UNU, 1985). The TEAA/ TAA was
strongly comparable to that of egg (50 %)
(FAO/WHO, 1990), and 43.6 % reported for pigeon
pea flour (Oshodi et al., 1993).
Most animal protein are low in Cys and hence in
Cys in TSAA (Adeyeye and Adamu, 2005). In
contrast many vegetable proteins contain substantially
more Cys than Met. The reverse is the case in the
present in which the % Cys in TSAA was 42.0 %.
Information on the agronomic advantages of
increasing the concentration of sulphur-containing
amino acid in staple foods shows that Cys had
positive effects on mineral absorption, particularly
zinc (Mendoza, 2002).
The P-PER value (1.69) was higher than 1.21
(cowpea), close to (Salunkhe and Kadam, 1989); 1.62
(millet ogi) and 0.27 (sorghum ogi) (Oyarekua and
Eleyinmi, 2004). A common feature of sorghum and
maize is that the proteins of these grains contain a
relatively high proportion of leucine. It was therefore
suggested that an amino acid imbalance from excess
leucine might be a factor in the development of
pellagra (FAO, 1995). Clinical, biochemical and
pathological observations in experiments conducted in
humans and laboratory animals showed that high
leucine in the diet impairs the metabolism of
tryptophan and niacin and is responsible for niacin
deficiency in sorghum eaters (Ghafoorunissa and
Narasinga Rao, 1973). High leucine is also a factor
contributing to the pellagragenic properties of maize
(Belavady and Gopalan, 1969). These studies
suggested that the leucine/isoleucine balance is more
important than dietary excess of leucine alone in
regulating the metabolism of tryptophan and niacin
and hence the disease process. The present Leu/Ile
ratios were low in value. The pI value was 5.14. The
pI of any organic matter is important when the protein
isolate is to be prepared. The EAAI can be useful as a
rapid tool to evaluate food formulations for protein
quality. The EAAI for soy flour is 1.26 (Nielsen,
2002) which is better than the current result of 1.18.
Table 7 showed that Met had the lowest score with
a value of 0.26. to correct for the AA needs from the
sample, it becomes 100/26 or 3.85 times as much raw
sample protein to be taken (eaten) when they are the
sole source of protein in the diet (Bingham,1977). In
Table 8 two different scoring patterns were presented:
Scorea
= Essential amino acid scores based on
FAO/WHO (1973) scoring pattern, Scoreb
= Essential
amino acid scores based on requirements of pre-
school child (2-5 years)(FAO/WHO/UNU, 1985).
In both patterns Met + Cys had the lowest score
(limiting amino acid), with values 0.41 and 0.57. and
would need a correction of 100/41 or 2.44 according
to the essential amino acid scoring pattern 100/57 or
1.75 times as much raw sample protein to be taken
(eaten) when they are the sole protein in the diet
(Bingham, 1977).
International Journal of Scientific Research in Knowledge, 2(2), pp. 92-104, 2014
100
Table 7: Amino acid score of the Bridelia ferruginea stem
bark samples based on whole hen's egg amino acid
4. CONCLUSION
This study has revealed that the Bridelia ferruginea
back consumed in Ekiti State and other states in the
South-western part of Nigeria can contribute useful
amount of nutrients including amino acids to human
diet. Interestingly, the anti-nutritional content of the
sample was low, much lower than is obtainable in
most Nigerian vegetables. This implies that, its overall
nutritional value will not be affected. Indeed, this part
of the plant consumed largely by the rural populace in
Ekiti State is not inferior to the conventional popular
Nigerian vegetables. There is need, however, to
determine the vitamins and fatty acids present in the
sample. Understandably, nutrient loss is of great
concern during blanching and cooking of vegetables,
therefore there is need to study the effects of cooking
and processing procedures on nutrient availability of
the Bridelia ferruginea back sample. This will help to
adequately establish their importance in human
nutrition and provide basis for maximum utilization of
the plant’s part.
Table 8: Essential amino acid scores of the Bridelia ferruginea stem bark samples
Scorea = Essential amino acid scores based on FAO/WHO (1973) scoring pattern, Scoreb = Essential amino acid scores based on requirements of pre-school
child (2-5 years)(FAO/WHO/UNU, 1985)
REFERENCES
Adeleke RO, Abiodun OA (2010) Chemical
Composition of Three Traditional Vegetables in
Nigeria. Pakistan Journal of Nutrition, 9 (9):
858-860.
Adesina AJ (2013). Proximate, Minerals and Anti-
nutritional Compositions of Three Vegetables
Commonly Consumed in Ekiti State,Nigeria.
International Journal of Pharmaceutical and
Chemical Sciences, 2(3): 1631-1638.
Adeyeye EI, Adamu AS (2005). Chemical
composition and food properties of
Gymnarchus niloticus (Trunk fish). Biosci.
Biotechn. Res. Asia, 3(2): 266-272.
Adeyeye EI (1997). Amino acid composition of six
varieties of dehulled African yam bean
(Sphenostylis stenocarpa) flour. Int. J. Food
Sci. Nutr., 48: 345-351.
Adeyeye EI (2010). Effect of cooking and roasting on
the amino acid composition of raw groundnut
(Arachis Hypogaea) seeds. Acta Sci. Pol.,
Technol. Aliment., 9(2): 201-216.
Adeyeye EI, Adesina AJ (2012). Nutritional
composition of African breadfruit (Treculia
africana) seed testa. Journal of Agric. Res. &
Dev., 11(1): 159- 178.
Jonathan and Funmilola
Nutritional and Anti-Nutritional Composition of Bridelia Ferruginea Benth (Euphorbiaceae) Stem Back Sample
101
Adeyeye EI, Orisakeye OT, Oyarekua MA (2012).
Composition, mineral safety index, calcium,
zinc and phytate interrelationships in four fast-
foods consumed in Nigeria. Bulletin of
Chemical Society of Ethiopia, 26(1): 43-54.
Akuodor GC, Mbah C C, Anyalewechi NA, Idris-
Usman M, Iwuanyanwu TC, Osunkwo, UA
(2011). Pharmacological profile of aqueous
extract of Bridelia ferruginea stem bark in the
relief of pain and fever. Journal of Medicinal
Plants Research, 5(22): 5366-5369.
Alsmeyer RH, Cunningham AE, Happich ML
(1974). Equations to predict PER from amino
acid analysis. Food Technol., 28, 34-38.
Altschul AM (1958). Processed plant protein
foodstuff. Academic Press, New York.
AOAC (2005). International Official Methods of
Analysis (18th edition). Association of
Analytical Chemists, Washington DC.
Bakoma B, Eklu-Gadegkeku K, Agbonon A,
Aklikoku K. Bassene E, Gbeassor M. (2011).
Preventive effect of Bridelia ferruginea against
highfructose diet induced tolerance, oxidative
stress and hyperlipidamia in male wistar rats.
Journal of pharmacology and toxicology, 6(3):
249 - 257.
Belavady B, Gopalan C (1969). The role of leucine in
the pathogenesis of canine blacktongue and
pellagra. Lancet 2, 956-957.
Bingham S (1977). Dictionary of nutrition. Barrie and
Jenkins London.
Bingham S (1978). Nutrition: A consumer’s guide to
good eating. Transworld Publishers.London.
Boham BA, Kocipai-Abyazan R (1974). Flavonoids
and condensed tannins from leaves of Hawairan
vaccinium valiculatum andV. calycinium.
Pacific Sci., 48: 458-463.
Chinma CE, Igyor MA (2007). Micronutrients and
anti-nutritional contents of selected tropical
vegetables grown in Southeast, Nigeria. Niger.
Food J., 25(1): 111- 116.
Cimanga K, DeBruyne T, Apers S, Pieter L, Totte J,
Kambu K, Tona L, Gill LS Akinwumi C
(1999). Nigeria medicinal plants practice and
belief of Ondo people.J. Ethnopharmacol, 18:
257-266.
Committee on Medical Aspects (COMA) (1984).
Food Policy Diet and cardiovasculardisease.
HMSO. London.
Day (Jnr) RA, Underwood AL (1986). Quantitative
analysis 5th ed., Prentice
HallPublication,London..
De Bruyne T, Cimanga K, Pieters L, Claeys M,
Dominisse R, Vlietinck A (1997).
Galocatechin; Epigallocatechin. A New
Biflavonoid Isolated from Bridelia ferruginea.
Nat. Prod. Let., 11: 47- 52.
Fafunso M, Bassir O (1977). Variations in the Loss of
Vitamins in Leafy vegetables with various
methods of food preparation, Food Chem, 21:
51-55.
Finar IL (1975). Organic chemistry. ELBS and
Longman London.
FAO/WHO (1990). Protein quality evaluation Report
of Joint FAO/WHO Expert Consultation.FAO
Food and Nutrition Paper 51. FAO Rome.
FAO/WHO/UNU (1985). Energy and protein
requirement. WHO Technical Report Series
724, WHO Geneva.
FAO/WHO (1973). Energy and protein requirements.
Technical Report Series 522. WHO, Geneva
Switzerland.
FAO (1995). Sorghum and millets in human nutrition.
FAO Food Nutrition Series 27. Food and
agriculture Organization of the United Nations.
Rome Italy.
Franke RW, Chasin BH (1980). Seeds of Famine:
Ecological Destruction and the Development
Dilemma in the West African Sahel. Montclair,
NJ: Allanheld and Osmun.
Ghafoorunissa S, Narasinga-Rao BS (1973). Effect of
leucine on enzymes of the tryptophan niacin
metabolic pathway in rat liver and kidney.
Biochem. J., 134: 425-430.
Grivetti LE, Frentzel CJ, Ginsberg KE, Howell KL,
Ogle BM (1987). Bush foods and edible weeds
of agriculture: perspectives on dietary use of
wild plants in Africa, their role in maintaining
human nutritional status and implications for
agricultural development.Health and disease in
Tropical Africa. Geographical and Medical
Viewpoints, ed. R Akhtar, pp. 51–81. London:
Harwood.
Grivetti LE (1979): Kalahari agro-pastoral hunter-
gatherers. The Tswana example. Ecol. Food
Nutr., 7: 235– 256.
Grivetti LE (1978). Nutritional success in a semi-arid
land. Examination of Tswana agro-pastoralists
of the Eastern Kalahari, Botswana. Am. J. Clin.
Nutr., 31: 1204–1220.
Harborne JB (1973). Phytochemical methods.
Capman and Hall, Ltd., London, 49-188.
Hathcock JN (1985). Quantitative evaluation of
vitamin safety. Pharmacy Times. 104-113.
Helsper JPFG, Hoogendijk M, Van Norel A and
Burger-Meyer K (1993). Antinutritional factors
in faba beans (Vicia faba L.) as affected by
breeding toward the absence of condensed
tannin. J Agric Food Chem.41:1058-1061.
Iheanacho ME, Udebuani AC (2009).Nutritional
Composition of Some Leafy Vegetables
International Journal of Scientific Research in Knowledge, 2(2), pp. 92-104, 2014
102
Consumed in Imo State, Nigeria. J. Appl. Sci.
Environ. Manage., 13(3): 35–38.
Ihekoronye AI, Ngoddy PO (1985). Integrated Food
Science and technology or tropics. Macmillian
publisher. London, pp.257-264.
Irobi ON, Moo-Young M, Anderson WA, Daramola
SO (1994). Antimicrobial activity of bark
extract of Bridelia ferruginea (Euphorbiaceae).
Journal of Ethnopharmacology, 43 (3): 185-
190.
Kareem KT, Kareem SO, Adeyemo OJ, Egberongbe
RK (2010). In vitro antimicrobial properties of
Bridelia ferruginea on some clinicalisolates.
Agriculture and Biology Journal of North
America, 1(3): 416-420.
Kolawole OM, Olayemi AB (2003). Studies on the
efficacy of Bridelia ferruginea Benth bark in
water purification. Nigerian Journal of Pure &
Applied Science, 18: 1387-1394.
Kolawole OM, Oguntoye SO, Agbede OO, Olayemi
AB (2006). Studies on the efficacy of Bridelia
ferruginea Benth bark extract on reducing the
coliform load and BOD5 of domestic
wastewater. Ethnobotanical Leaflet, 10: 228–
238.
Kolawole OM, Adesoye AA (2010). Evaluation of the
antimalarial activity of bridelia ferruginea
benth bark. SENRA Academic Publishers,
Burnaby, British Columbia, 4(1): 1039-1044.
Kubmarawa D, Andeyang IF, Magomya H (2008).
Amino Acid Profile of Two Non- conventional
Leafy Vegetable, Sesamum and Balanites
aegyptiaca. Afr. J. Biotechnol., 7(19): 3502-
3504.
Ladeji O, Ahin CU, Umaru HA (2004). Level of
antinutritional factors in Vegetables commonly
eaten in Nigeria. Afr.J. Nat. Sci.7:71-73.
Makkar AOS, Goodchild AV (1996). Qualification of
tannins. A laboratory manual, ICARDA,
Aleppo, Syria.
Mendoza C (2002). Effect of genetically modified
low phytic acid plants on mineral absorption.
Int. J. Food Sci. Nutr., 37: 759-767.
McNeely JA (1990). Conserving the World’s
Biological diversity. Transaction Books. New
Brunsick. 208. National Advisory Committee
on Nutrition Education (NACNE) (1983).
Proposal for nutritional guidelines for healthy
education in Britain. Health Education
Council.London.
National Research Council (NRC) (1989). Food and
Nutrition Board Recommended Dietary
Allowances (10th edition). National Academy
Press. Washington DC.
Nelson SS (1994). Introduction to the Chemical
Analysis of Foods. Jones and Bartletes
Publishers, Londoon pp. 93-201.
Nieman DC, Butterworth DE, Nieman CN (1992).
Nutrition. Wm. C. Brown Publishers.Dubuque.
Nielsen SS (2002). Introduction to the chemical
analysis of foods. CBS Publ. Distrib.
NewDelhi.
Nkafamiya II, Osemeahon SA, Modibbo UU, Aminu
A (2010). Nutritional status of non conventional
leafy vegetables,Ficus asperifolia and Ficus
sycomorus. African Journal of Food Science,
4(3): 104-108.
Nkafamiya II, Manji AJ (2006). A Study of
Cyanogenetic Glucoside Contents of some
Edible Nuts and Seeds. J. Chem. Soc. Niger.
31(1 and 2): 12-14.
Ogugbuaja VO, Akinniyi JA, Ogarawu VC,
Abdulrahman F (1997). Elemental Contents of
Medicinal Plants. Faculty of Science
Monograph Series, No. 1, pp. 1–42. Faculty of
Science, University of Maiduguri, Nigeria, pp.
1–42.
Obadoni BO, Ochuko PO (2001). Phytochemical
studies and comparative efficacy of the crude
extracts of some Homostate plants in Edo and
Delta States of Nigeria. Global J Pure Appl
Sci.:8b:203-208.
Okon EU, Akpanyung EO (2005). Nutrients and
Antinutrients in selected Brands of Malt- drinks
Produced in Nigeria. Pak. J.Nutr., 4(5):352-355.
Olaofe O, Akintayo ET (2000). Prediction of
isoelectric points of legume and oilseed
proteins from their amino acid compounds. J.
Techno-Sci., 4: 49-53.
Olajide OA, Makinde JM, Awe SO (1999). Effect of
aqueous extract of Bridelia ferruginea stem
bark corrageenan-induced oedema and grand
coma tissue formation rats and mice. J.
Ethnopharmacol., 66 (1): 113-117.
Osagie AU, Offiong AA (1998). Nutritional Quality
of Plant Foods. Ambik Press, Benin City, Edo
State Nigeria pp. 131-221.
Oshodi AA, Olaofe O, Hall GM (1993). Amino acid,
fatty acid and mineral composition of pigeon
pea (Cajanus cajan). Int. J. Food Sci. Nutr., 43:
187-191.
Oyarekua MA, Eleyinmi AF (2004). Comparative
evaluation of the nutritional quality of
corn,sorghum and millet ogi prepared by
modified traditional technique. Food
Agric.Environ. 2 (2), 94-99.
Paul AD, Southgate AT, Russel J (1976). First
supplement to McCance and Widdowson’s. The
composition of foods. HMSO,London.
Jonathan and Funmilola
Nutritional and Anti-Nutritional Composition of Bridelia Ferruginea Benth (Euphorbiaceae) Stem Back Sample
103
Pearson D (1976). Chemical Analysis of Foods (7th
edition). Churchill. London.
Varian Techtron (1975). Basic AtomicAbsorption
Spectroscopy-A Modern Introduction.
Dominica Press. Victoria.
Robinson DE (1987). Food biochemistry and
nutritional value. Longman Sci. Techn. London.
Salunkhe DK, Kadam SS (1989). Handbook of world
food legumes, nutritional chemistry, processing
technology and utilisation. Boca Raton, CRC
Press Florida.
Sena LP, VanderJagt DJ, Rivera C, Tsin AT,
Muhamadu I, Mahamadu O, Millson M,
Pastuszyn A, Glew RH (1998). Analysis of
nutritional components of eight famine foods of
the Republic of Niger. Plant Foods Hum. Nutr.
52: 17– 30.
Sparkman DH, Stein EH, Moore S (1958). Automatic
Recoding Apparatus for Use in
Chromotography of Amino acids. Anal.Chem.,
30: 119.
Sravan PM, Venkateshwarlu GK, Vijaya BP, Suvarna
D, Dhanalakshmi CH (2011). Effects of anti
inflammatory activity of Amaranthus viridis
Linn. Annals Biological Research, 2(4): 435-
438.
Steel RGD, Torrie JH (1960). Principles and
Procedures of Statistics. McGraw-Hill. London.
Steinke FH, Prescher EE, Hopkins DT. (1980).
Nutritional evaluation (PER) of isolated
soybean protein and combinations of food
proteins. J. Food Sci., 45: 323-327.
Wheeler H, Ferrel J (1971). In:Okon, E.U and
Akpanyung EO (2005).Nutrients and
Antinutrients in selected Brands of Malt Drinks
Produced in Nigeria. Paki. J. Nutr., 4(5): 352-
355.
Young SM, Greaves JS (1940). Influence of variety &
treatment onphytin content of wheat, Food Res.,
5: 103-105.
International Journal of Scientific Research in Knowledge, 2(2), pp. 92-104, 2014
104
Adesina, Adeolu, J. is a Ph.D candidate in Food/ Analytical Chemistry of the Department of
Chemistry, Ekiti State University, Ado Ekiti, Nigeria. He received his first degree from
University of Ilorin, Kwara State, Nigeria 2001 awarded with Bachelor of Science in Pure
Chemistry. He obtained degree in Master of Science in Analytical Chemistry from University of
Ibadan, Nigeria in 2006. His current research focuses on Food chemistry and quality. To date, he
has published several scientific journal articles related to Food chemistry and quality evaluation
field.
Akomolafe Seun Funmilola is a Ph.D candidate in Food biochemistry and toxicology of the
Department of Biochemistry, Federal University of Technology, Akure, Nigeria. She received her
first degree from Ekiti State University, Ado Ekiti Nigeria. 2005 awarded with Bachelor of Science
in Biochemistry. She obtained degree in Master of Science in Biochemistry from University of
Ibadan, Nigeria in 2009. Her current research focuses on Food Biochemistry and quality. To date,
she has published several scientific articles related to food Biochemistry and toxicology field.
International Journal of Scientific Research in Knowledge, 2(2), pp. 105-115, 2014
Available online at http://www.ijsrpub.com/ijsrk
ISSN: 2322-4541; ©2014 IJSRPUB
http://dx.doi.org/10.12983/ijsrk-2014-p0105-0115
105
Full Length Research Paper
A Study on the Relationship between Accounting Conservatism and Earnings
Management in Teheran Stock Exchange Listed Companies
Abbas Ramezanzadeh Zeidi1*
, Zabihollah Taheri2, Ommolbanin Gholami Farahabadi
3
1Department of Accounting, Neka Branch, Islamic Azad University, Neka, Iran
2Department of Accounting, Payamenour University, Sari, Iran
3Department of Accounting, Payamenour University, Neka, Iran
*Corresponding Author: E-mail: [email protected]
Received 06 December 2013; Accepted 26 January 2014
Abstract. The present study focuses on the link between accounting conservatism and earnings management in Teheran Stock
exchange listed companies. To this aim, the researches selected a statistical sample consisting of 154 companies and gathered
statistical data for time period from 1385 to 1390. Using multiple variable combinational regressions, the researchers extracted
the proper research model and examined the research hypothesis. The models developed for conservatism and earnings
management were respectively book value to market value ratio of the stockholders' equity, and Jones's adjusted model.
Primarily, research findings indicated that the models are insignificant and a significant link between conservatism and
earnings management does not exist. However, when the researches fitted the examination based on logarithm of conservatism,
they found out that there is significant and negative link between conservatism and earnings management.
Keywords: conservatism, earnings management, discretionary accruals
1. INTRODUCTION
Regarded as the basic providers of companies'
resources, investors always require accurate and
comprehensive databases from the companies.
Accounting information appears in financial
statements and investors regularly refer to such
information without adjustingitto the changes in
accounting methods or the way those information
were calculated (Hendriksen et al., 1982). Income
statements are a key tool among the financial
statements for evaluating the performance as well as
the profitability of the business enterprise. The
information needs to be shared in a way that itenables
the investors to evaluate preceding performance and to
effectively assess and forecast the profitability of the
business enterprise. As a result, the profit reported in
the statements helps the investors evaluate the
performance and profitability of the firm and fulfill
their expectations about theiridealreturn profit.
Therefore, boththe reported profitand the qualitative
characteristics of the profit mean a lot to the investors
(Francis et al., 2004). Earnings management is a
means managers take advantage of to manipulate
reported profit. It is, in effect, a targeted interference
bypersonal motives of managers in the process of
financial reporting to the individuals outside the
business enterprise and is achieved by manipulating
the information of the current period. In other words,
managers let their personal judgments meddle in the
process of financial reporting and manipulate the
mechanism of transactions to make changes in the
financial reports.
Conservatism has been a controversial premise
from the outset and plays an important part in the
practice of Accounting. A conservative approach
defines a level of caution in forecasting the profit;
however, itdoes the same with the possible lossesonly
if it is risk-free for the upcoming cash flow. Preparing
conservative financial statements heightens reliability
of accounting data and indicates the ability of
accounting profit to illustrate financial profit (positive
dividend yield) and financial loss (negative dividend
yield). Conservative approach stresses on
distinguishing between positive and negative dividend
yield (financial profit and loss) (Basu, 1997). The
notions of profit management and conservatism in
accounting have other functions in financial reporting,
and each of them solely is capable of influencing
immensely the quality of financial reporting and
consequently the efficiency of capital market and also
the behavior of investors, creditors and in general the
Zeidi et al.
A Study on the Relationship between Accounting Conservatism and Earnings Management in Teheran Stock Exchange
Listed Companies
106
users of financial statements. Therefore, the
researchers believe that studying the correlation
between these two parameters can be a step forward
and contribute significantly to the literature on this
subject. The researchers seek to answer the following
questions: does conservatism in accounting procedure
have any influence on profit management? Is there
any correlation of any kind between conservatism and
profit management?
2. THEORETICAL FRAMEWORK AND
RELATED LITERATURE
2.1. Defining conservatism
Basu (1997) defines conservatism as the necessity to
gainhigh degree of certainty to differentiate between
desirable news e.g. profit from undesirable ones e.g.
loss. Such a definition views conservatism from a
profit/loss standpoint. However, other definitions
(Feltham and Ohlson, 1995) examine conservatism in
the balance sheets. Based on this, where there is an
actual uncertainty in selecting from among a number
of reporting methods, that method is preferable which
has the least desirable effect on the rights of the
shareholders. The third definition (Givoly and Hayn,
2000) combines the two aforementioned methods of
balance sheets and profit / loss. In this third approach,
conservatismis defined as an accounting notion and
results in a reduction in reported cumulated
dividendcaused by belated acknowledgement of profit
and prompt acknowledgement of expenses, low
evaluation of assets and high evaluation of debts.
Ryan (2006) draws another category to define
conservatism i.e. conditional and unconditional
conservatism. Conditional conservatism is obligated
by accounting standards. This translates into prompt
recognition of the losses in case of any undesirable
news and not acknowledging the profit in case of any
desirable news. For instance, applying the law of
minimum cost or that of net sales value in inventory
evaluation is an example of conditional conservatism,
also called profit / loss or retrospective conservatism.
On the other hand, unconditional conservatism is not
obligated by the widely accepted accounting
standards. This type of conservatism does not go
beyond indicating net book value of the assets through
traditional accounting procedures. It is also known as
balance sheet or futuristic conservatism.
2.2. Literature Review
Many studies have ever been conducted on topics
related to profit management and conservatism. Basu
(1997) studied the link between earnings and dividend
using regression to estimate the conservatism index.
He realized that in companies whose dividend yield is
negative, the dividend yield has higher correlation
with earnings compared with companies whose
dividend yield is positive. He also found out that in
periods of judicial and court trials, conservatism
increases. Watts and Zimmerman (1978) hold that
companies with higher political costs tend to apply
more conservative accounting procedures. Just to
prove this fact, Ahmed et al. (2002) showed that big
companies apply conservative accounting procedures
more than other companies. Their study also revealed
that in case of any discrepancies between the interests
of the loaners and those of the shareholders
concerning distribution of the income, the managers
of the borrowing companies are more likely to apply
conservative accounting procedures. He also
discovered that there is negative relation between
conservatism and profit management. In another study
(2007), Ahmed concludes that conservative
accounting discourages managers from investing on
projects with negative return.
In addition, Nikolav (2008) examined the relation
between conservative accounting and the limitations
of debt covenants. He found out that the more limited
the debt covenant is, the more the conservatism
grows. This fact had already been reached at in Ball et
al. (2007). Watts (2003), believes that if a company's
contract with different groups e.g. investors and
creditors should be based on accounting figures, then
the companies' managers, due to discrepancies
between their own interests and those of the groups,
will try to unfairly manipulate the figures in their own
favor. For example, they may increase the profit or
asset and on the other hand decrease the debts. In such
cases, conservatism, as an effective regular
mechanism, neutralizes the manager's unfair
manipulations by postponing the acknowledgement of
profit and helping a prompt recognition of the debt
and loss. In their research Zhou and Lebov (2006)
found out that companies that offer conservative
financial statement are able to handle profit
management more efficiently. However, as Zhou
discovered, such companies normally do not get
involved in earning management. Richardson (2005)
claims that accrual items have proven to be more
reliable and can predict loss and profits of the coming
year as well.
Bur Gestaller et al. (2006) concluded in their
research that private companies make bigger profits
compared with state-run companies and in countries
with more efficient judicial system, these companies
have a smaller share in profit management. Lafond
and Wattz (2007) showed that informational
asymmetry between aware and unaware investors
gives rise to conservatism in financial statements.
Conservatism lowers managers' motiveand ability to
International Journal of Scientific Research in Knowledge, 2(2), pp. 105-115, 2014
107
manipulate accounting figures and consequently,
informational asymmetry and great losses it is
responsible for are reduced and the value of the
company increases. Moreover, in another study (2008)
Lafond and Watz pointed out that conservative
financial reporting is part of a regulation system that
makes managers less able to manipulate profit and to
raise cash flow in the company. Ball and Shivakumar
(2006) believe that once the managers realize that they
can no longer postpone loss recognition to the coming
years, they appreciate conservative accounting since it
helps solve potential issues and restricts company's
investments on projects with negative Net Present
Value (NPV).
3. THESIS HYPOTHESIS
Based on the primary studies that were already carried
out in the field, the hypothesis of the research is as
follows: There is a meaningful link between
accounting procedures and profit management
4. MTHODOLOGY
4.1. Statistical population and sample volume
Statistical population of the present research includes
Tehran Stock Exchange listed companies. The
statistical sample has been narrowed down using a
systematic omission method regarding the following
requirements:
(1) The business enterprise must not be an
investment company, a leasing company or a bank,
due to their field of activity
(2) The end of fiscal year of the business enterprise
must coincide the end of Esfand
(3) The business enterprise must not experience a
shift in fiscal year during the study period
(4) Financial data of the business enterprise must
be available during the study period
Considering these requirements, the researchers
selected 154 companies for a study period from 1385
until 1390.
4.2. Research Methodology
The present study is categorized as a descriptive-
explorative research. It studies the status quo and
describes it regularly trying to examine its different
features in relation to the variables. Such a research is
significant both for applied and theoretical areas. The
findings may well be put into use in decision and
policy making, and the explorations can contribute
substantially to theories since they have been reached
at through deductive methods.
4.3. Research parameters
4.3.1. Independent variable
In the present study profit management has been
considered as the independent variable. Based on the
study, the proper variable to indicate profit
management is the accrual items. These items may be
subcategorized into non-discretionary and
discretionary accrual items. The former is determined
by activity levels and is out of the control of the
managers; the latter is within the control of the
managers and may simply be manipulated. The
researchers hold that the residual of accrual items
model is a criterion of discretionary accrual items and
may be considered as profit management, meaning
that after estimating the model and ensuring that its
statistical qualities are effective, the residual amount
of the model is considered as profit management
variable. Dechow et al. (1995) and Guay et al. (1996)
argued that Jonse's adjusted model is the most
practical one among the existing models to estimate
discretionary accrual items. Because of this, the
researchers have applied this model in the present
study. The model is formulated as follows:
Where, TACit = total accruals for company i in year t;
TAit-1 = Lagged total asset for company I; ΔREVit =
change in operating revenues for company i in year t;
ΔRECit = change in net receivables for for company i
in year t; PPE it = gross property, plant and equipment
for company i in year t; α0j - α3j = regression
parameters; E = error term
4.3.2. Dependent variable
Conservatism is the dependent variable in this study.
According to other studies ever carried out on the
topic (Ahmed et al., 2002; Zang, 2007; Lebov et al.,
2008; Jean and Rezaee, 2004) the ratio of book value
to the market value of stockholders' interest has been
chosen to represent conservatism. Therefore, if the
ratio of book value to the market value of
stockholders' interest is less than one, then it can
indicate accounting conservatism.
Zeidi et al.
A Study on the Relationship between Accounting Conservatism and Earnings Management in Teheran Stock Exchange
Listed Companies
108
4.3.3. Control variables
In this research, variables such as company size,
financial leverage and shareholders' dividend yield
have been used as control variables in an attempt to
control the effects of other factors. Zimmerman
(1983) states that, because of more political
sensitivities, bigger companies tend to apply
conservatism more than other companies. Previous
researches prove that we may use common logarithm
of total asset at the end of each fiscal period (Derashid
AND Zang, 2003) and also logarithm of the total sales
income (Zimmerman, 1983) as a criterion to measure
a company size. Since the total sales income has direct
effect on the profit, it can influence the results of the
study in a way that are not desirable, therefore, the
researchers have decided that common logarithm of
total asset at the end of each fiscal period is an
acceptable criterion to indicate the company size.
Besides, the study conducted by Chen et al. (2007)
proved that companies with lower profitability tend to
manage profits more cautiously and effectively.
Kowthari et al. also believe that discretionary accrual
items are connected with a company's performance
which was calculated and evaluated through Return on
Equity (ROE).
On the other hand, accounting methods relate to
financial leverage, since one of the essential criteria of
the creditors in Iran (banks mainly) is the company's
debt ratio. Therefore the higher the company's debt
ratio is, the less it tends to apply conservative
methods. As a result, managers are expected to apply
less conservative methods in their financial statements
in a bid to minimize the risk that their offer to receive
loans from the banks may not be granted, and to stop
to be imposed a burden of higher interest
rates.Therefore, the researchers have decided that debt
ratio represents financial leverage. The ratio is
estimated by dividing the total debt by total asset.
4.4. Research model
The following regression model has been applied in
the research to study the links between the use of
conservatism in accounting procedures and earnings
management:
EM = β0 + β1 CON + β2 Size + β3 ROE + β4 Lev +e
Where, EM: Earnings management / CON:
Conservatism / SIZE: Company's Size / REO: Return
On Equity; LEV: Financial Leverage / β: fixed
parameter / e: error term
5. STATISTICAL METHODS AND
TECHNIQUES
Since the objective of the present study is to examine
the links between conservatism and earnings
management, multiple variable regression models has
been used based on mixedmethod data analysis to test
the research hypothesis. In this analysis, the proper
models were fitted based on the results from Chaw
test and Hausman test. To run significant test for the
fitted regression model, Fischer statistic was used in
95% assurance level. Respectively, the researchers
used T-student statistic to studyvariable coefficient of
regression model; Durbin-Watson test to study auto-
correlation among observations; and finally adjusted
determining variable statistic to examine how
explainable the model is.
6. DATA ANALYSIS AND EXAMINATION OF
THE RESEARCH HYPOTHESES
In the present research, the ratio of book value to
market value of the stockholders' equity has been
considered as a criterion of conservatism in
accounting procedures. The reason why such a
criterion was chosen was that notable researches such
as Ahmad et al. (2002), Zhang (2007) and Lobo et al.
(2008) also took advantage of this criterion in their
studies. They found out that there is a meaningful
negative link between conservatism and discretionary
accruals. Jain & Rezaee showed in their research that
when the ratio of book value to market value of the
stockholders' equity is less than one, it can indicate
accounting conservatism. Besides, as once mentioned
earlier, residual sum of adjusted Jones's discretionary
accruals model has been used to estimate earnings
management. To examine the research core model, the
researchers embedded the variable of book value ratio
to market value of the stockholders' equity as an
indicator of accounting conservatism asindependent
variable. On the other hand, they embedded residual
sum of adjusted Jones's discretionary accruals model
that is the same as discretionary accruals as
independent variable to indicate earnings
management. Moreover, control variable such as
company size, company leverage and return on equity
have been exerted to control the undesirable effects.
6.1. Descriptive statistics of research core model
variables
Descriptive statistics of research variables include
central tendency, variability and distribution
indicators. In this research, the relevant data related to
mean and median have been presented in category of
central tendency, standard deviationin category of
International Journal of Scientific Research in Knowledge, 2(2), pp. 105-115, 2014
109
variability, and finally elongation and skewness in
category of distribution. Moreover, Jarko-bra statistic
and relevant significant level have been presented in
this chart to test normality of distribution of research
variables. Descriptive statistics of research core model
variables have been presented in Table 1.
Table 1: Descriptive Statistics of the Main Model Variables
variable Mean Median Standard
deviation
Skewness Elongation Jarco-bra
statistic Probability
Con 0.880815 0.550846 4.854001 22.24455 591.4461 13218915 0.000000
Em -7.49E-19 0.003025 0.233599 -6.818365 154.5776 857948.6 0.000000
Size 13.38333 13.17000 1.483872 0.808653 4.018428 137.1351 0.000000
ROE 0.405161 0.261357 5.445384 4.407300 131.4621 624521.4 0.000000
Lev 0.765846 0.649520 1.332373 15.5463 288.6741 624521.4 0.000000
The observations indicate that in average, book
value of the sample companies is about 88% of their
market value. As mentioned earlier, Jain & Rezaee
believe that when the ratio of book value to market
value of the stockholders' equity is less than one, it
can indicate accounting conservatism. Since this ratio
is below one in the sample companies, therefore on
may conclude that conservative accounting exists in
these companies. The ratio for half of the companies
is above 0.55 and for the second half it is below 0.55.
According to the observations, earnings management
in the sample companies is on average -7.49E-19. The
negative mark implies that the companies either have
adopted an earnings reduction policy or have not
taken any measure at all to manage the earnings.
However, this does not suggest any lack of earnings
management in those companies. This might be due to
the fact that the average accrual items are negative.
These items have already been referred earlier in this
paper. In other words, companies on average possess
negative accrual items.
The mean size of the sample companies according
to asset logarithm is 13.38. The median of this
variable decreased by 0.21 unit and amounted to
13/17. Mean ROE of the sample companies are 40.5
percent, which means that net profit is on average 40.5
percent of shareholders' equity. In half of the
companies the equity is above 26 percent and in the
other half it is below 26 percent. The observations
suggest that the sample companies' debts make up an
average of 76.5 percent of their assets. In half of the
companies, this ratio is above 65 percent and in the
other half it is less than 65 percent. All the research
variables have positive skewness except for earnings
management indicator. Positive skewness implies that
distant samples from central tendency indicator are
located on the right domain of the measurement scale.
When earnings management indicator has negative
skewness then distant samples from central tendency
indicator are located on the left domain of the
measurement scale. Besides, all the research variables
have positive elongation which means that variable
distribution curve is longer than normal distribution
curve. As Jarco-Bra statistic and its corresponding
significant level suggests, not all the research
variables have normal distribution.
6.2. Examining correlation among the research
variables
In this section, researchers examine the correlation
among the core model variables of the research
applying Pearson correlation coefficient. Table 2
presents correlation coefficient matrix.
Table 2: Correlation coefficients between the variables of the main model
con Size ROE Lev
Con : Pearson Correlation
Sig.
1.000000
Size: Pearson Correlation
Sig.
0.003540
0.9175
1.000000
ROE: Pearson Correlation
Sig.
-0.010421
0.7604
-0.043804
0.1996
1.000000
Lev: Pearson Correlation
Sig.
-0.025484
0.4557
0.068145
0.0459
-0.017358
0.6114
1.000000
As the table shows, the only significant correlation
regarding level of significance of correlation
coefficient among the variables (below 0.05) is the
correlation between size and leverage, being about
0.068. This means that the correlation is affirmative;
however, the correlation intensity between them is
evaluated as weak. Therefore, applying the variables
Zeidi et al.
A Study on the Relationship between Accounting Conservatism and Earnings Management in Teheran Stock Exchange
Listed Companies
110
to the model at the same time will not cause any
interference concerning collinearity.
6.3. Examination of research core models
6.3.1. Examining the model on links between
conservatism and earning management
The researchers primarily examined required tests in
each case to select a proper pattern. According to F
statistic ofChaw test and the sum of related probability
(above 0.05) the model lacks required effects. Since
this test does not recommend applying mixed data
along with the effects, therefore there is no need to
Hausman test and the model is immediately fitted. The
results are presented in Table 3. Table 3: The Results of Choosing a Model for Model Test of Relation between conservatism and earning management
Test type Sample
statistic
Statistic
quantity
df sig
Chow test F 0.008238 153,723 1.0000
Table 4: Estimating the model of relation between conservatism index and earning management
Dependent variable: discretionary accruals
Explanatory variable Coefficients Standard error T statistic sig
The width of source: (α0) 0.000168 0.002285 0.073643 0.9413
BV to MV ratio -0.000345 .0000800 -0.431327 0.6663
Fischer statistic: 0.083745 Adjusted coefficient of explanation statistic : 0.000043
Probability of Fisher’s Statistic: 0.772354 Durbin-Watson’s Statistic : 2.188045
The results from model estimation indicate that t-
statistic and its relevant probability show lack of (α0)
significance and book value to market value ratio in
the model. In other words, there is no evidence that
there is significant link between conservatism and
earnings management. The adjusted R2 model statistic
indicates that conservatism cannot explain earning
management efficiently. Since all the findings
suggested the inefficiency of the model to explain the
link between conservatism and earnings management,
once again another model was fitted based on
logarithm of book value to market value ratio to
examine the link more closely.
6.4. Testing the model of links between
conservatism and earning management
As it was earlier mentioned, to examine the links
between the variables more closely, another model
was fitted based on logarithm of book value to market
value ratio. In other words, logarithm of book value to
market value ratio has been considered as a criterion
of conservatism. To select a proper pattern, the
researchers examined required tests in each case.
According to F statistic of Chaw test and the sum of
related probability (above 0.05) the model lacks
required effects. Since this test does not recommend
applying mixed data along with the effects, therefore
there is no need to Hausman test and the model is
immediately fitted. The results are presented in Table
5. Table 6 introduces the results of model estimation
using combinational method without exercising the
effects.
Table 5: The Results of Choosing a Model for Model Test of Correlation between conservatism and earning management
Test type Sample statistic Statistic
quantity
df sig
Chow test F 0.992227 153,698 0.5139
Table 6: Estimating the model of correlation between conservatism index and earnings management
Dependent variable: discretionary accruals
Explanatory variable Coefficients Standard error T statistic sig
The width of source (α0) -0.003538 0.002619 -1.350927 0.1771
BV to MV ratio -0.005332 0.002352 -2.267174 0.0236
Fischer statistic: 4.641173 Adjusted coefficient of explanation statistic : 0.014601
Probability of Fisher’s Statistic: 0.031494 Durbin-Watson’s Statistic : 2.201023
International Journal of Scientific Research in Knowledge, 2(2), pp. 105-115, 2014
111
Regression equation extracted from model estimation
is as follows:
EM = -0.003538 - 0.005332*LNCONS
Having studied the results from model estimation,
the researchers found out that t statistic and its related
probability (less than 0.05) indicates significance of
conservatism (logarithm of Book value to market
value ratio). Negative coefficient of the variable
proves that it has negative and significant effect on the
model. This means that there exists a negative and
significant link between conservatism and earnings
management. In other words, the higher the
conservatism index becomes, the less the earnings
management is exercised, and vice versa. (α0),
however, does not influence the model significantly.
Adjusted R2 statistic of the model indicates that only
1.4 percent of the earnings management may be
explained via conservatism. Of course, coefficient of
explanation of the model is too low. One may
conclude that the model is poorly explainable. Durbin-
Watson statistic of the model suggests that the
remaining model is still independent.
6.5. Testing the model of links between
conservatism and earning management along with
control variables
To control the undesirable effects, researchers
involved other variables such as company size,
company leverage and performance as control
variables. First, Chaw and Hausman test were used to
select a proper model. The results of these tests have
been presented in Table 7. According to F statistic of
Chaw test and the sum of related probability (below
0.05) the model carries the required effects. Again,
with regard to X2 statistic of Hausman test the related
sum of the probability (higher than 0.05) the model
contains random effect. Table 8 introduces the results
of model estimation using random method
Table 7: The Results of Choosing a Model for Model Test of Correlation between conservatism index and Earnings
Management at the Presence of the Control Variables
Test Type Sample Statistic Statistic
Quantity
df sig
Chow Test F 1.423763 153,658 0.0018
Hausman Test X2 7.972217 4 0.0926
Table 8: Estimating the model of correlation between conservatism index and earning management at the presence of the
control variables
Dependent variable: discretionary accruals
Explanatory variable Coefficients Standard error T statistic sig
The width of source (α0) 0.0222000 0.028989 0.765811 0.4441
BV to MV ratio -0.028492 0.013198 -2.158818 0.0312
size -0.003145 0.001755 -1.792030 0.0736
ROE 4.72E-05 0.000508 0.092848 0.9261
Lev 0.012039 0.002036 5.913195 0.0000
Fischer statistic: 1.418300 Adjusted coefficient of explanation statistic: 0.074572
Fischer statistical probability: 0.001843 Durbin-Watson statistic: 1.714347
Regression equation extracted from model estimation
is as follows:
EM= 0.0222 – 0.028492 * Ln Cons – 0.003145 * Size
+ 4.72E-05* ROE + 0.012039* Lev+ [CX=R]
Having studied the results from model estimation,
the researchers found out that F statistic and its related
probability (less than 0.05) indicates significance of
the whole model. Negative coefficient of the variable
proves that it has negative and significant effect on the
model. This means that there exists a negative and
significant link between conservatism and earnings
management. In other words, the higher the
conservatism index becomes, the less the earnings
management is exercised, and vice versa. (α0),
however, does not influence the model significantly. T
statistic and the related probability (less than 0.05)
indicates significance of leverage variable. Positive
coefficient of the variable proves that it has Positive
and significant effect on the model. In other words,
the higher the leverage (debt to asset ratio) becomes,
the more the earnings management is exercised, and
vice versa. T statistic and the related probability (more
than 0.05) indicates lack of significance of size and
ROE variables.However, size variable up to 10
percent error has negative significance on the model.
In this case, as the size increases (assets logarithm)
earnings management increases, and vice versa.
6.6. Testing the model of links between
conservatism and earning management along with
significant control variables
In this model, researchers involved only company size
variable, since it had significant effect on the model.
First, Chaw and Hausman test were used to select a
proper model. The results of these tests have been
presented in Table 9. According to F statistic of Chaw
Zeidi et al.
A Study on the Relationship between Accounting Conservatism and Earnings Management in Teheran Stock Exchange
Listed Companies
112
test and the sum of related probability (below 0.05)
the model carries the required effects. Again, with
regard to X2 statistic of Hausman test the related sum
of the probability (lower than 0.05) the model lacks
proper consistent effects. Table 10 introduces the
results of model estimation using consistent method.
Table 9: The Results of Choosing a Model for Model Test of Correlation between Earnings Management and Earnings
Response Coefficient at the Presence of the Control Variables
Test Type Sample Statistic Statistic
Quantity
df sig
Chow Test F 1.457958 153,683 0.0009
Hausman Test X2 8.323279 2 0.0156
Table 10: Estimating the model of correlation between conservatism index and earnings management along with significant
control variables
Dependent variable: discretionary accruals
Explanatory variable Coefficients Standard error T statistic sig
The width of source (α0) -0.020441 0.002852 -7.167389 0.0000
BV to MV ratio -0.029073 0.003773 -7.706612 0.0000
Lev 0.010837 0.000666 16.26696 0.0000
Fischer statistic: 8.231768 Adjusted coefficient of explanation statistic: 0.572214
Fischer statistical probability: 0.000000 Durbin-Watson statistic: 2.238676
Regression equation extracted from model estimation
is as follows:
EM= -0.020441 – 0.029073*Ln Cons + 0.010837*
Lev = [CX=F]
Having studied the results from model estimation,
the researchers found out that F statistic and its related
probability (less than 0.05) indicates significance of
the whole model. T statistic and the related probability
(less than 0.05) indicates significance of conservatism
(logarithm of book value to market value ratio)
Negative coefficient of the variable proves that it has
negative and significant effect on the model. This
means that there exists a negative and significant link
between conservatism and earnings management. In
other words, the higher the conservatism index
becomes, the less the earnings management is
exercised, and vice versa. T statistic and the related
probability (less than 0.05) indicates significance of
the width of source, however, it influences the model
negatively and significantly. T statistic and the related
probability (less than 0.05) indicates significance of
leverage variable. Positive coefficient of the variable
proves that it has Positive and significant effect on the
model. In other words, the higher the leverage (debt to
asset ratio) becomes, the more the earnings
management is exercised, and vice versa. Adjusted R2
statistic of the model indicates that about 57.2 percent
of the earnings management may be explained via
conservative variables in relation to leverage control
variable. It is noteworthy that coefficient of
explanation of the model has increased dramatically
compared with prior models, to the extent that the
model is explainable more than 50 percent.
7. CONCLUSION
With regard to the tests that were run, the findings
indicate that the whole model lacks significance. Book
value to market value ratio lacks significance, as well.
This means that no significant link was explored
between conservatism and earnings management.
Since all the cases that were examined showed
inefficiency of the model, therefore another model
was fitted based on logarithm of book value to market
value ratio. This time the findings proved that the
whole model as well as logarithm of book value to
market value ratio is significant. Negative coefficient
of this variable indicates negative and significant
effect of the variable on the model. Afterwards, to
control undesirable effects, other variables such as
company size, company leverage and ROE were also
involved in the study. Having fitted the model, the
researchers found out that the whole model as well as
logarithm of book value to market value ratio is
significant. The findings also indicate that leverage
variable is significant, since positive coefficient of the
variable signifies its positive and significant effect.
Next, the final model was fitted applying the leverage,
a significant control variable. Eventually, the final
findings of the study proved that the whole model as
well as logarithm of book value to market value ratio
is significant. Adjusted coefficient of determination of
the model indicates that about 57.2 percent of earnings
management by conservatism is explainable using
leveraging control variable.
7.1. Research implications
(1) As it was once mentioned in the descriptive
statistics chapter, the observations showed that on
International Journal of Scientific Research in Knowledge, 2(2), pp. 105-115, 2014
113
average, book value of the sample companies are less
than their market value. That the ratio of book value to
market value of the stockholders' equity is below one
indicates accounting conservatism is very likely to be
there. Since this ratio is below one in sample
companies, therefore one may rightly conclude that
accounting conservatism is being exercised in these
companies. The researchers recommend the possible
users to refer to the findings and take this matter into
their consideration when making financial decisions.
(2) Since there is a negative and significant link
between conservatism and earnings management, it is
expected that as conservatism increases, earnings
management decrease. In some instances in which
earnings management takes a negative aspect, the
negative and significant link between conservatism
and earnings management can stop manipulating the
earnings. Therefore, analysts and possible users of the
research findings are recommended to take this matter
into consideration when making financial decisions.
7.2. Suggestions for further research
(1) The researchers suggest a similar study with an
emphasis on the notion of earnings quality.
(2) The researchers suggest that links between
conservatism and income smoothing be studied.
(3) The researchers suggest that links between
conservatism and diverse features of time series of
profit such as profit stability, profit predictability and
the like are studied.
REFERENCES
Ahmed AS, Billings BK, Morton RM, Stanford-Harris
M (2002). The Role of Accounting
Conservatism in Mitigating Bondholder-
Shareholder Conflicts over Dividend Policy and
in Reducing Debt Costs. The Accounting
Review, 77(4): PP.867–890.
Ahmed AS, Duellman S (2007). Evidence on the role
of accounting conservatism in corporate
governance. Journal of Accounting and
Economics, 43: PP, 411–437.
Basu S (1997). The Conservatism Principle and the
Asymmetric Timeliness of Earnings. Journal of
Accounting and Economics, 24: 3-37.
Ball R, Shivakumar L (2006). The role of accruals in
asymmetrically timely gain and loss
recognition. Journal of Accounting Research,
44: 24-402 .
Burgstahler D, Hail L, Leuz C (2006).The Importance
of Reporting Incentives: Earnings Management
in European Private and Public Firms. The
Accounting Review, 81(5): 983-1016.
Chen Q, Hemmer T, Zhang Y )2007(. On the relation
between conservatism in accounting standards
and incentives for earnings management.
Journal of Accounting Research, 45(3): 541.
Dechow P, Sloan R, Sweeney A )1995(. Detecting
earnings management. The Accounting Review,
70 4) ): 193-225.
Ding Y, Stolowy H (2007). Timeliness and
conservatism changes over time in the
properties of net income in France. on line:
http://www.ssrn.com.
Dimitrios V, Kousenidis, Anestis C, Ladas, ChristosI
Negakis.( 2009),” Value relevance of
conservatism and non-conservatism accounting
information”, the international journal of
accounting; 44:219-238.
Feltham G, Ohlson JA (1995) Valuation and clean
surplus accounting for operating and financial
activities, Contemporary Accounting Research,
11(2): 689–731.
Francis J, LaFond R, Olsson P, Schipper K (2004).
Cost of Equity and Earnings Attribute. The
Accounting Review, 79 (4): 967-1010.
Guay WR, Kothari SP, Watts RL (1996). A Market-
based Evaluation of Discretionary Accruals
Models. Journal of Accounting Research, 34:
83-105.
Givoly D, Hayn C (2000). The changing time-series
properties of earnings, cash flows and accruals:
Has financial reporting become more
conservative?. Journal of Accounting and
Economics, 29(3): 287-320.
Givoly D, Hayn C, Natarajan A (2006). Measuring
reporting conservatism. The Accounting
Review, 82(1): 65-106.
Guay WR, Kothari SP, Watts RL (1996). A Market-
based Evaluation of Discretionary Accruals
Models. Journal of Accounting Research, 34:
83-105.
Hendriksen Eldon S, Van Breda, Michael F (1982).
Accounting Theory. McGraw-Hill.
Jain P, Rezaee Z (2004). The Sarbanes-Oxley Act of
2002 and Accounting Conservatism. Working
Paper Series, available at: www.ssrn.com.
Lafond R, Watts RL (2007). The Information Role of
Conservative Financial Statements.
http://ssrn.com/abstract=921619 or
http://dx.doi.org/10.2139.
Nikolaev V )2008(. Debt covenants and accounting
conservatism: complements or substitutes?
www.ssrn.com.
Penman SH, Zhang XJ (2002). Accounting
Conservatism the quality of earnings, and Stock
Returns. Accounting Review, 77: 237-264.
QIchen Hemmer T, Zhang Y (2007). On the relation
between conservatism and incentive for earning
Zeidi et al.
A Study on the Relationship between Accounting Conservatism and Earnings Management in Teheran Stock Exchange
Listed Companies
114
management. Journal of Accounting Research,
45(3).
Richardson S, Sloan R, Soliman M, Tuna I (2005).
Accrual Reliability, Earnings Persistence and
Stock Prices. Journal of Accounting and
Economics, 39: 437-485.
Ryan S (2006). Identifying conditional conservatism.
European Accounting Review, 15(4): 511-525.
Watts R, Zimmerman J (1978). Towards a Positive
Theory of the Determinants of Accounting
Standards. The Accounting Review, 53(1): 112-
134.
Watts, R, (2003), Conservatism in accounting, Part II:
evidence and research opportunities.
Accounting Horizons 17: 287-301.
Zimmerman J, watts R (1986). Positive accounting
theory. prentice –hall, inc.
Zhang J (2008). The Contracting Benefits of
Accounting Conservatism to Lenders and
Borrowers. Journal of Accounting and
Economics, 45: 27–54.
Zhou J (2008). Financial Reporting After the
Sarbanes-Oxley Act: Conservative or Less
Earnings Management?, Research in
Accounting Regulation, 20: 187-192.
International Journal of Scientific Research in Knowledge, 2(2), pp. 105-115, 2014
115
Abbas Ramezanzadeh Zeidi received his MA in Accounting from Tehran Branch, Islamic Azad
University. Currently, he is PhD Candidate at AMU India. He has more than 20 papers in the referees
journals and conferences. He is faculty member of Neka Branch, Islamic Azad University.
Zabihollah Taheri received his MA in Accounting from Tehran Branch, Islamic Azad University. He is
faculty member of Payamenour University, Sari, Iran.
Ommolbanin Gholami Farahabadi received her MA in Accounting from Payamenour University,
Behshahr, Iran. She is faculty member of Payamenour University, Neka, Iran.