Vol.3, No.2, 77-81 (2011) Health
doi:10.4236/health.2011.32014
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/HEALTH/
Adenylate kinase locus 1 genetic polymorphism and
type 2 diabetes
Fulvia Gloria-Bottini1*, Elena Antonacci2, Anna Neri1, Andrea Magrini1, Egidio Bottini1
1Division of Biopathology of Human Population and Environmental Pathology, Department of Biopathology and Imaging Diagnos-
tics, Unive r si ty of Rome Tor Vergata, Rome, Italy; *Corresponding Author:
2Centre of Diabetology, Local Sanitary Unit, Penne, Italy.
Received 9 August 2010; revised 14 February 2011; accepted 18 August 2011
ABSTRACT
AK1 catalyzes the reversible reaction ATP+AMP
 2ADP thus contributing to the regulation of
relative concentration of these important nu-
cleotides. Intracellular ATP is a storage of en-
ergy for cellular processes, moreover extracel-
lular ATP together with ADP, AMP and adeno-
sine are critical signalling molecule for sending
messages to nearby cells acting on P1 and P2
receptors. AK1 shows a genetic polymorphism
and recently our group has shown that the cor-
relation between blood glucose and glycated
haemoglobin in T2D is dependent on AK1 phe-
notype. In the present paper we have carried
further studies on the relationship between AK1
phenotypes and T2D. Possible interactions with
ABO blood groups and ACP1 polymorphism
have also been investigated. We have
re-examined the data on 280 subjects with type
2 diabetes from the White population of Penne
(Central Italy). 384 consecutive healthy new-
borns from the same population have been also
studied. A three way contingency table analysis
was carried out according to Sokal and Rohlf
and other statistical analyses by SPSS pro-
grams. T2D patients with AK12-1 phenotype
have higher values of blood glucose level and
glycated haemoglobin and an increased ten-
dency to dyslipidemia and retinopathy. In addi-
tion there is an interaction of AK1 with ABO
blood group s and with ACP1 polymorphism. The
different activity between AK1 phenotypes could
influence the relative concentration of ATP, ADP,
AMP and adenosine with important effects on
metabolic activity thus explaining the associa-
tion of AK1 with clinical manifestation of T2D.
Keywords: AK; T2D; Glycemia; ATP;
Glycated Haemoglobin
1. INTRODUCTION
Adenylate kinase (AK) is an ubiquitous enzyme that
catalyzes the nucleotide phosphoryl interconversion ATP
+ AMP  2ADP. The products of this reaction are in-
volved in the regulation of many cellular function and
relationship. Intracellular ATP is a storage of energy for
cellular processes. The energy from extracellular ATP is
used for sending messages to nearby cells.
At first, extracellular ATP was seen simply as a neuro-
transmitter, subsequently was studied the mechanism of
interactions between extracellular ATP signalling with
other signalling systems outside the cells named ec-
toATPases. This large family of AK enzyme removes
from ATP, one by one, its phosphates producing adeno-
sine diphosphate (ADP), adenosine monophosphate
(AMP) and adenosine that have different effect on sev-
eral cells by binding themselfes to P2 family (for ADP,
AMP) and to P1 (for adenosine) receptors 1].
The family of AK enzymes includes seven genes,
AK1-AK7, with each other different functions,molecular
weight and kinetic property. The network of these en-
zymes are distributed throughout intracellular compart-
ments, interstitial space and body fluids to regulate en-
ergetic and metabolic signaling circuits and to fasten an
efficient economy of cell energy, signal communication
and stress response.Mutations in AK1, AK2 or AK7 genes
have been found associated with hemolytic anemia, re-
ticular dysgenesis and ciliary dyskinesia 2].
Adenylate kinase locus 1 (AK1) belongs to AK family
and plays an important role in the synthesis of nucleo-
tides requested for many metabolic functions. It is pre-
sent in the cytosol of skeletal muscle, brain, and eryth-
rocyte. The enzyme is polymorphic and shows three
phenotypes with different activity, in the order AK11 >
AK12-1 > AK12 corresponding to the presence of two
codominant alleles, AK1*1 and AK1*2 at an autosomal

F. Gloria-Bottini et al. / Health 3 (2011) 77-81
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/HEALTH/
78
locus on chromosome 9. Rare alleles AK1*3, AK1*4 and
AK1*5 have been also referred.
AK1 was identified because of its association with a
rare genetic disorder causing nonspherocytic hemolytic
anemia where a mutation in the AK1 gene was found to
reduce the catalytic activity of the enzyme and the re-
placement of Arg-128 with Trp 3].
Our group has recently shown that the correlation
between blood glucose and glycated haemoglobin in
T2D is dependent on AK1 phenotype. In addition, blood
glucose is significantly higher in AK12-1 than in AK11
subjects 4]. This prompted us to study in more details
the relationship between AK1 and T2D. We have also
considered possible interaction s o f AK 1 with ABO blood
groups that is linked to AK1 5] and with ACP1 that is
associated with clinical manifestation of T2D 6].
2. MATERIAL AND METHODS
We have re-examined the data on 278 subjects with
type 2 diabetes from the population of Penne (Italy) and
have studied 384 consecutive healthy newborns from the
same population. T2D patients have been considered in a
previous study 6] and were a random sample of a popu-
lation of about 2000 subjects under care at Centre of
Diabetology of the local Hospital. Penne is a rural town
located in Eastern side of Central Italy. This homoge-
nous population are the descendant of an old Italic po-
pulation called “Vestini”. The total number of subjects
shown in the tables are not always the same due to some
random missing value for t he variables considered.
The patients have been controlled in the Centre of
Diabetology according to a regular schedule. In occasion
of the control blood glucose and glycated haemoglobin
levels have been measured (more than 9 hours since the
last meal). Blood sample for determination of genetic
markers were also obtained. Written informed consent
was obtained from patients and from mothers of new-
borns to participate to this study that was approved by
the Institutional Review Board.
2.1. Laboratory Analysis
Serum glucose concentration was measured by the
automated Roche/Hitachi cobas C501 system based on
enzymatic reaction with exochinase. Glycated haemo-
globin was determined using the Menarini Diagnostics
HA-8160 automated equipment based on inverse ex-
change cationic chromatography.
AK1 phenotype was determined by starch gel electro-
phoresis of haemolysate 7]. Samples were examined at
pH7. The insert were made from Whatman, n˚3 filter
paper. After electrophoresis the gels were sliced and then
covered with a 0.75% agar solution at 45˚C made in 0.1
M tris buffer pH 8 and containing glucose 10 mM, mag-
nesium chloride 20 mM, adenosine diphosphate (ADP) 1
mM, nicotinammide adenine dinucleotide phosphate (NA-
DP) 0.4 mM, phenazine methosulphate (PMS) 0.012%,
tetrazolium salt (MTT) 0.012%, glucose-6-phosphate de-
hydrogenase (G6PD) 0.04 units/ml and hexokinase 0.08
units/ml. The agar was allowed to set and then the gel
incubated at 37˚C for two hours.
At the sites of AK activity ADP is converted into AMP
and ATP. The ATP reacts with glucose in the presence of
hexokinase to produce ADP and glucose-6-phosphate
(G6P), this is oxidized to 6-phosphogluconate by G6PD
with concomitant reduction of NADP. The reduced
NADP in the presence of PMS causes the reduction of
MTT to give a blue-coloured insoluble formazan, which
is thus deposited at the sites of AK activity. In Caucasian
populations three distinct types of electrophoretic p attern
are recognized referred as AK11, AK12-1 and AK12 cor-
responding to the presence of two codominant alleles:
AK1*1 and AK1*2 at an autosomal locus.
ACP1 phenotype has been determined by starch gel
electrophoresis on red blood cell haemolysates according
to Spencer et al. 8]. The acid phosphatase pattern is
revealed by a solution of phenolphthalein diphosphate:
The addendum of ammonium solution reveals the area
where phenophtalein has been liberated in the areas of
gel where ACP1 activity is present. In European popula-
tions, the presence of three common alleles *A, *B and
*C determines the occurrence of six phenotypes with
enzymatic activity increasing in the order: A < AB < B<
AC < BC < C. Each of the homozygous A, B, and C
phenotypes are composed of two fractions F and S cor-
responding to a fast and slow component of electropho-
retic pattern. Heterozygous phenotypes have a pattern
corresponding to a mixture of homozygous types.
2.2. Statistical Analysis
Statistical analyses have been carried out by SPSS pro-
gram 9]. Three way contingency table analysis has been
performed according to Sokal and Rohlf 10].
3. RESULTS
Clinical and demographic data in subjects with T2D
and in healthy newborns are shown in Table 1.
In T2D the frequency of AK12-1 is slightly greater in
comparison to healthy newborn s but the difference is not
statistically significant (see Table 2). The data suggest
that Ak1 may not be an important factor primarily in-
volved in the susceptibility to type 2 diabetes.
The distributions of blood glucose and glycated Hb
are shown in Tab le 3. Both parameters show a greater
value in AK12-1 than in AK11 but only for bloo d glucose

F. Gloria-Bottini et al. / Health 3 (2011) 77-81
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/HEALTH/
7979
Table 1. Clinical and demographic data of the samples studied.
MeanS.E. %ProportionTotal n˚
T2D
Age (yrs) 66 0.59 280
Age at onset of disease (yrs) 54 0.65 275
Duration of disease (yrs) 11.90.48 269
BMI 29.50.29 277
Male proportion 47% 279
Dyslipidemia 26.8% 276
Healthy Newborns
Male proportion 47.6% 424
Birthweight (g) 3348.322.0 417
Gestational age (wks) 39.70.06 411
Table 2. Distribution of AK1 phenotypes in T2D and in healthy
newborns.
AK11 AK12-1 Total n˚
T2D subjects
93.5% 6.5% 278
Healthy newborns
94.3% 5.7% 384
Chi square test of independence
2 df p
0.054 1 0.816
Table 3. Blood glucose and glycated haemoglobin levels ac-
cording to AK1 phenotype in T2D.
Blood glucose Glycated Hb
Mean S.E. Mean S.E.
AK11 133.9 2.3 7.51 0.11
AK12-1 156.9 11.8 8.18 0.39
T-test for differences between means
p = 0.015 p = 0.121
the difference is statistically significant.
The proportion of subjects with dyslipidemia and
retinopathy is greater in AK12-1 than in AK11 but the
difference does not reach the level of statistical signifi-
cance (see Table 4).
In T2D the proportion of AK2-1 phenotype is very
high in B blood group and very low in A group . The dif-
ference between A and B blood groups is highly signifi-
cant (O.R. = 10.173 C.I. 2.151-39.163). No significant
association has been observed between ABO and AK1
phenotypes in healthy newborns (see Table 5).
Table 6 shows in T2D subjects the distribution of
blood glucose and glycated Hb in relation to the joint
ABO-AK1 phenotypes. Both parameters in A and O sub-
jects show higher values in those carrying the AK12-1
phenotype than in those carrying the AK11 phenotype
while in B subjects there is a slight tenden cy to a reduc-
tion of both parameters in AK12-1phenotype as com-
pared to Ak1 phenotype.
Figure 1 displays the relationship between serum
glucose level an d ACP1 enzymatic activity in AK11 and
AK12-1 subjects with T2D. In AK11 phenotype the con-
centration of serum glucose is positively correlated with
ACP1 activity while in AK12-1 phenotype the associa-
tion follows an opposite pattern.
4. DISCUSSION
The present data show that T2 D patients with AK12-1
phenotype have higher values of blood glucose level and
glycated haemoglobin and have an increased tendency to
dyslipidemia and retinopathy. In addition there is an in-
teraction of AK1 with ABO blood groups and ACP1
polymorphism.
AK1 belongs to AK family and, at the sites of AK1 ac-
tivity, ADP is converted into AMP and ATP. The reaction
is reversible thus contributing to the regulation of rela-
tive concentration of these important nucleotides 1].
Extracellular ATP plays a physiological role in the
maintenance of glucose homeostasis by regulating insu-
lin secretion 11]. ATP is able to stimulate the release of
pancreatic insulin modulating glucose transport via
GLUT1 acting on P2 purinergic ionotropic (P2X) and P2
metabotropic (P2Y) receptors through a mechanism that
involves beta-cell metabolism and a rise of intracellular
calcium. P2Y receptors are deficient in fibroblasts from
T2D patients 12].
Table 4. Prevalence of dyslipidemia and retinopathy in T2D
subjects according to AK1 phenotype.
Dyslipidemia Retinopathy
% of subjects with
dyslipidemia Total n˚ % of subjects with retinopathy Total n˚
AK11 26.0% 258 15.5% 258
AK12-1 38.9% 18 27.8% 18
Chi square test of independence
2 df p 2 df p
1.431 1 0.232 1.938 1 0.152

F. Gloria-Bottini et al. / Health 3 (2011) 77-81
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/HEALTH/
80
Table 5. Distribution of the joint ABO-AK1 phenotypes i n T2D
and in healthy newborns.
ABO phenotypes
T2D subjects
A B AB O
Proportion of AK12-1
phenotype 2.7% 21.9% 0.0% 6.2%
Total n˚ 112 32 4 130
Chi square test of independence
2 df p
A vs B vs AB vs O 15.497 3 0.001
A vs B vs O 15.012 2 0.0005
A vs B 11.378 3 0.0007
O.R. = 10.173(B vs A/AK12-1 vs AK11)
95% C.I. 2.151-39.163
Healthy newborns
Proportion of AK12-1
phenotype 6.8% 9.7% 0.0% 4.5%
Total n˚ 162 31 12 178
Chi square test of independence
2 df p
A vs B vs AB vs O 2.458 3 0.483
A vs B vs O 1.653 2 0.437
A vs B 0.036 1 0.849
O.R.=1.471(B vs A/AK12-1 vs AK1)
95% C.I. 0.027-6.230
Three way contingency table analysis by a log linear model
x = ABO (A vs B);
y = AK1 (AK11 vs AK12-1);
z = sample (newborns vs T2D)
G df p
xyz interaction 4.177 1 0.042
ODDS ratio analysis (B vs A)/(T2D vs newborns)
AK11 phenotype O.R. = 1.236 95%C.I. 0.656-2.330
AK12-1 phenotype O.R. 8.55 95% C.I. 0.99-92.86
An increase in the AMP/ATP ratio activates adenine
monophosphate activated protein kinase (AMPK) that
regulates glycolysis, glucose uptake, lipid oxidation, fatty
acid synthesis, cholesterol synthesis and gluconeogene-
sis. 13,14]. The AMPK channelopathies are present in
hyperinsulinemia, neonatal diabetes mellitus and are a
risk factor for the aetiology of T2D 15,16].
Ta b l e 6 . Blood glucose and glycated Hb in T2D according to
the joint ABO-AK1 phenotype.
ABO phenotype
Blood glucose
A B O
AK11
Mean 129.5 141.4 135.3
S.E. 3.6 8.2 3.3
AK2-1
Mean 164.7 139.1 169.5
S.E. 34.3 15.2 20.0
Significance of difference
between means (t-Student)(p) N.S. N.S. N.S.
Glycated Hb
AK11
Mean 7.31 8.15 7.55
S.E. 0.19 0.36 0.15
AK12-1
Mean 8.57 7.35 8.75
S.E. 1.15 0.59 0.54
Significance of difference
between means (t-Student) (p)N.S. N.S. 0.042
The differences in enzymatic activity between AK1
phenotypes could influence the relative concentration of
ATP, ADP, AMP and adenosine influencing metabolic
parameters and contributing to explain the association
with clinical manifestation in T2D. Further in vestigation
in this area could be rewarding. Our research is in line
with the recent interest on genetic variants that through
the regulation of insulin secretion by -cells could be
involved in the pathogenesis and clinical manifestations
of type 2 diabetes. 17].
Genetic variability of ABO blood groups substances
that are important components of cell membrane struc-
ture may influence AK1 ecto-enzyme activity thus ex-
plaining the associations reported in Tables 5 and 6.
In general high activity of ACP1 is associated with
high blood glucose level. Low AK1 activity modifying
the ratio among ATP, ADP and AMP may influence the
effect of ACP1 activity on blood glucose resulting in
lower glucose level in carriers of high ACP1 activity
genotypes.
5. CONCLUSION
Low adenylate kinase activity associated to AK12-1
phenotype influencing the relative concentration of ATP,

F. Gloria-Bottini et al. / Health 3 (2011) 77-81
Copyright © 2011 SciRes. Openly accessible at http://www.scirp.org/journal/HEALTH/
8181
Figure 1. Blood glucose in T2D patients in relation to
AK1 and ACP1 phenotypes. “AK1-ACP1 blood glu-
cose interaction” ACP1 three classes ((A + BA) vs B
vs (CA+CB)) “p = 0.079 ACP1 two classes ((A + BA)
vs (C+CB))” p = 0.030.
ADP, AMP and adenosine could have negative effects on
the clinical evolution of T2D.
REFERENCES
[1] Khakh, B.S. and Burnstock, G. (2009) The double life of
ATP. Scientific American, 301, 84-90.
doi:10.1038/scientificamerican1209-84
[2] Dzeja, P. and Terzic, A. (2009) Adenylate kinase and
AMP signaling networks: Metabolic monitoring, signal
communication and body energy sensing. International
Journal of Molecular Sciences, 10, 1729-1772.
doi:10.3390/ijms10041729
[3] Matsuura, S., Igarashi, M., Tanizawa, Y., Yamada, M.,
Kishi, F., Kajii, T., Fujii, H., Miwa, S., Sakurai, M. and
Nakazawa, A. (1989) Human adenylate kinase deficiency
associated with hemolytic anemia. A single base substitu-
tion affecting solubility and catalytic activity of the cy-
tosolic adenylate kinase. Journal of Biological Chemistry,
264, 10148-10155.
[4] Gloria-Bottini, F., Antonacci, E., Cozzoli, E., De Acetis,
C. and Bottini, E. (2010) The effect of genetic variability
on the correlation between blood glucose and glycated
hemoglobin levels. Metabolism, 60, 250-255.
doi:10.1016/j.metabol.2010.01.003
[5] Weitkamp, L.R., Sing, C.F., Shreffler, D.C. and Guttormsen,
S.A. (1969) The genetic linkage relations of adenylate
kinase : Further data on the ABO-AK link age group. Ameri-
can Journal of Human Genetics, 21, 600- 605.
[6] Bottini, N., Gloria-Bottini, F., Borgiani, P., Antonacci, E.,
Lucarelli, P. and Bottini, E. (2004) Type 2 diabetes and
the genetics of signal transduction: a study of interaction
between adenosine deaminase and acid phosphatase lo-
cus 1 polymorphisms. Metabolism, 53, 995-1001.
doi:10.1016/j.metabol.2004.03.006
[7] Fildes, R.A. and Harris, H. (1966) Genetically deter-
mined variation of adenylate kinase in man. Nature, 209,
261-263. doi:10.1038/209261a0
[8] Spencer, N., Hopkinson, D.A. and Harris, H. (1964)
Quantitative differences and gene dosage in the human
red cell acid phosphatase polymorphism. Nature, 201, 299-
300. doi:10.1038/201299a0
[9] SPSS/PC+ Version 5.0 (1992) Chicago: SPSS Inc.
[10] Sokal, R.R. and Rohlf, J.F. (1981) Biometry, WH Free-
man, New York.
[11] Petit, P., Lajoix, A.D. and Gross, R. (2009) P2 purinergic
signalling in the pancreatic beta-cell: Control of insulin
secretion and pharmacology. European Journal of
Pharmaceutical Scien ces, 37, 67-75.
doi:10.1016/j.ejps.2009.01.007
[12] Solini, A., Chiozzi, P., Morelli, A., Passaro, A., Fellin, R.
and Di Virgilio, F. (2003) Defective P2Y purinergic re-
ceptor function: A possible novel mechanism for im-
paired glucose transport. Journal of Cellular Physiology,
197, 435-444. doi:10.1002/jcp.10379
[13] Misra, P. and Chakrabarti, R. (2007) The role of AMP
kinase in diabetes. Indian Journal of Medical Research,
125, 389-398.
[14] Miranda, N., Tovar, A.R., Palacios, B. and Torres, N. (2007)
AMPK as a cellular energy sensor and its function in the or-
ganism. Revista de Investigacion Clinica, 59, 458- 469.
[15] Remedi, M.S. and Koster, J.C. (2010) K(ATP) channelo-
pathies in the pancreas. Pfl ugers Archiv, 460, 307-320.
[16] Wasada, T. (2002) Adenosine triphosphate-sensitive po-
tassium (K(ATP)) channel activity is coupled with insulin
resistance in obesity and type 2 diabetes mellitus. Inter-
nal M edicine, 41, 84-90.
[17] Chistiakov, D.A., Potapov, V.A., Khodirev, D.C., Sham-
khalova, M.S., Shestakova, M.V. and Nosikov, V.V.
(2009) Genetic variations in the pancreatic ATP-sensitive
potassium channel, beta-cell dysfunction, and suscepti-
bility to type 2 diabetes. Acta Diabetol, 46, 43-49.