Decreased ADP-ribosyl cyclase activity in peripheral blood mononuclear cells from diabetic patients with nephropathy.

Abstract

ADP-ribosyl-cyclase activity (ADPRCA) of CD38 and other ectoenzymes mainly generate cyclic adenosine 5'diphosphate-(ADP-) ribose (cADPR) as a second messenger in various mammalian cells, including pancreatic beta cells and peripheral blood mononuclear cells (PBMCs). Since PBMCs contribute to the pathogenesis of diabetic nephropathy, ADPRCA of PBMCs could serve as a clinical prognostic marker for diabetic nephropathy. This study aimed to investigate the connection between ADPRCA in PBMCs and diabetic complications.
PBMCs from 60 diabetic patients (10 for type 1 and 50 for type 2) and 15 nondiabetic controls were fluorometrically measured for ADPRCA based on the conversion of nicotinamide guanine dinucleotide (NGD(+)) into cyclic GDP-ribose.
ADPRCA negatively correlated with the level of HbA1c (P = .040, R(2) = .073), although ADPRCA showed no significant correlation with gender, age, BMI, blood pressure, level of fasting plasma glucose and lipid levels, as well as type, duration, or medication of diabetes. Interestingly, patients with nephropathy, but not other complications, presented significantly lower ADPRCA than those without nephropathy (P = .0198) and diabetes (P = .0332). ANCOVA analysis adjusted for HbA1c showed no significant correlation between ADPRCA and nephropathy. However, logistic regression analyses revealed that determinants for nephropathy were systolic blood pressure and ADPRCA, not HbA1c.
Decreased ADPRCA significantly correlated with diabetic nephropathy. ADPRCA in PBMCs would be an important marker associated with diabetic nephropathy.

Full text

Page 1

Hindawi Publishing Corporation
Experimental Diabetes Research
Volume 2008, Article ID 897508, 8 pages
doi:10.1155/2008/897508
ResearchArticle
Decreased ADP-Ribosyl Cyclase Activity in Peripheral Blood
Mononuclear Cells from Diabetic Patients with Nephropathy
Michio Ohtsuji,1Kunimasa Yagi,1Miyuki Shintaku-Kubota,1Yukiko Kojima-Koba,1Naoko Ito,1
Masako Sugihara,1Naoto Yamaaki,1Daisuke Chujo,1Atsushi Nohara,2Yoshiyu Takeda,1
Junji Kobayashi,1Masakazu Yamagishi,2and Haruhiro Higashida3
1Division of Endocrinology and Diabetology, Kanazawa University Graduate School of Medicine, 13-1 Takaramachi,
Kanazawa 920 8640, Japan
2Division of Cardiovascular Medicine, Department of Internal Medicine, Kanazawa University Graduate School of Medicine,
13-1 Takaramachi, Kanazawa 920 8640, Japan
3Department of Biophysical Genetics, Kanazawa University Graduate School of Medicine, 13-1 Takaramachi,
Kanazawa 920 8640, Japan
Correspondence should be addressed to Kunimasa Yagi, diabe@med.kanazawa-u.ac.jp
Received 30 September 2008; Accepted 12 December 2008
Recommended by Andreas Pf¨ utzner
Aims/hypothesis. ADP-ribosyl-cyclase activity (ADPRCA) of CD38 and other ectoenzymes mainly generate cyclic adenosine
5’diphosphate-(ADP-) ribose (cADPR) as a second messenger in various mammalian cells, including pancreatic beta cells and
peripheral blood mononuclear cells (PBMCs). Since PBMCs contribute to the pathogenesis of diabetic nephropathy, ADPRCA
of PBMCs could serve as a clinical prognostic marker for diabetic nephropathy. This study aimed to investigate the connection
between ADPRCA in PBMCs and diabetic complications. Methods. PBMCs from 60 diabetic patients (10 for type 1 and 50
for type 2) and 15 nondiabetic controls were fluorometrically measured for ADPRCA based on the conversion of nicotinamide
guanine dinucleotide (NGD+) into cyclic GDP-ribose. Results. ADPRCA negatively correlated with the level of HbA1c (P = .040,
R2= .073), although ADPRCA showed no significant correlation with gender, age, BMI, blood pressure, level of fasting plasma
glucose and lipid levels, as well as type, duration, or medication of diabetes. Interestingly, patients with nephropathy, but
not other complications, presented significantly lower ADPRCA than those without nephropathy (P = .0198) and diabetes
(P = .0332). ANCOVA analysis adjusted for HbA1c showed no significant correlation between ADPRCA and nephropathy.
However, logistic regression analyses revealed that determinants for nephropathy were systolic blood pressure and ADPRCA, not
HbA1c. Conclusion/interpretation. Decreased ADPRCA significantly correlated with diabetic nephropathy. ADPRCA in PBMCs
would be an important marker associated with diabetic nephropathy.
Copyright © 2008 Michio Ohtsuji et al.ThisisanopenaccessarticledistributedundertheCreativeCommonsAttributionLicense,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
1.INTRODUCTION
Diabetes mellitus is characterized by chronic hyperglycemia
and the development of diabetes-specific microvascular
complications. As a consequence of these complications,
diabetes is a leading cause of end stage renal disease
(ESRD)[1].Actually,diabetic nephropathyisthesinglemost
common cause of ESRD in Japan [2]. Recent epidemiologic
data indicated that the number of Japanese type 2 diabetic
patients on renal replacement therapy has tripled within
less than 15 years, and Japanese type 2 diabetic patients
might be particularly predisposed to nephropathy [3, 4].
Therefore, earlier detection of high-risk Japanese subjects
for the diabetic nephropathy is quite important for early
intervention to prevent ESRD.
Chronic low-grade inflammation and activation of the
innate immune system are closely involved in the patho-
genesis of diabetes and its microvascular complications [5,
6]. Inflammatory events are central to the pathogenesis of
diabetic nephropathy [7–9], and inflammatory cytokines
are involved in its development and progression [10].
Macrophages derived from circulating monocytes have been
recently focused on as playing central roles in the pro-
gression of diabetic nephropathy [11, 12]. Since PBMCs

Page 2

2Experimental Diabetes Research
mainly consist of lymphocytes and monocytes, both of
which contribute to the progression of diabetic nephropathy,
intracellular regulatory signals involved in the activation of
PBMCs could serve as a possible therapeutic target and/or
clinical prognostic marker for diabetic nephropathy.
Cyclic adenosine 5’diphosphate-ribose (cADPR) plays
a second messenger role in a variety of mammalian cells,
including pancreatic beta cells, kidney mesangial cells, and
PBMCs [13]. The rapid release of Ca2+from the smooth
endoplasmic reticulum (SER), the most characterized Ca2+
organelle, is evoked by stimulation of two kinds of receptors
on the SER for ryanodine and inositol (1,4,5)-triphosphate.
cADPR is catalyzed from beta-NAD+by ADP-ribosyl-
cyclase activity (ADPRCA) of CD38, CD157, and other
ectocellular membrane-bound enzymes [14–18]. cADPR
has been recently demonstrated to stimulate a variety of
mammaliancellfunctions,includingproliferation,secretion,
contraction, and vasodilation [19–25]. For example, several
studies have examined ADPRCA in kidney mesangial cells
[26, 27], and cADPR has been reported to induce smooth
muscle contraction in small renal arteries [28].
Since cADPR modulates adaptive immune recognition
process of PBMCs [29], we hypothesized that impaired
regulatory signals of ADPRCA in PBMCs would result in
the progression of diabetic vascular complications. However,
no clinical studies have been performed to examine the
relationship between CD38 and diabetic complications.
Based on this background, this study investigates the
connection between ADPRCA in PBMCs and diabetic
vascular complications, especially nephropathy.
2. MATERIALS AND METHODS
2.1.Studypopulation
Patients were eligible for the study if they exhibited diabetes
mellitus. Diabetes mellitus was diagnosed and classified
according to World Health Organization (WHO) criteria
[30,31].Subjectswithotherendocrinediseasesorsignificant
renal and/or hepatic diseases were excluded. The study,
approved by the ethics committee of Kanazawa University
Graduate School of Medical Science, was conducted in
accordance with the Declaration of Helsinki (1964), and all
patients gave written informed consent before participating
in the study.
Definitionofdiabeticcomplications
ssessment and diagnosis of diabetic complications were
performed as below [32]. Nephropathy was diagnosed as
the existence of albuminuria, proteinuria, and/or creati-
nine clearance <60mL/min. Albuminuria was defined as
urinary albumin excretion between 20 and 200mg/24hr
or urinary albumin to urinary creatinine ratio between
30 and 300mg/gCr. Proteinuria was defined as Albustix
[Ames] positive. Creatinine clearance values were calculated
by the Cockroft-Gault formula. Retinopathy was diagnosed
on the basis of direct ophthalmoscopy (through a dilated
pupil) by an experienced ophthalmologist and/or by flu-
orescein angiography. Peripheral neuropathy was assessed
by questioning patients about symptoms of neuropathy,
including paresthesia, dulled sensation, and pain in legs
and feet and was based on clinical examination (i.e.,
measuring abnormal knee/ankle reflexes and a Semmes-
Weinsteinmonofilament)andconfirmedbymeasurementof
conduction velocities of ulnar (motor and sensory), tibial,
peroneal, and sural nerves. Macroangiopathy was considered
in subjects who met the following three conditions: (1)
a history of a cardiovascular event and/or the presence
of angina and/or permanent ischemic electrocardiogram
abnormalitiesatrestorischemicabnormalitiesinastresstest
(usually combined with a cardiac noninvasive imaging tech-
nique); (2) claudication and/or abolished peripheral pulses
and/or foot lesions due to vascular disease demonstrated
by Doppler echography and/or angiography; or (3) carotid
vascular disease, as assessed by Doppler echography.
Medicationfordiabetes
Medication for type 2 diabetics was classified into three
categories:dietarytherapywithoutmedication,insulininjec-
tion, and oral hypoglycemic agents including sulphonylureas
and alpha-glucosidase inhibitors. No type 2 diabetic subjects
being treated with insulin sensitizers (including thiazolidine-
diones and biguanides) were enrolled in this study.
2.2.Laboratorymeasurements
Body mass index (BMI) was calculated as weight (in
kilograms) divided by height (in meters) squared. Venous
blood samples were obtained after a 12-hour overnight
fast. Blood glucose was measured with the glucose oxidase
method and HbA1c by high-pressure liquid chromatogra-
phy. Serum total cholesterol (TC) and triglycerides (TG)
were determined by enzymatic methods, and high-density
lipoprotein cholesterol (HDL-C) levels were measured by a
polyanion-polymer/detergent method. Low-density lipopro-
tein cholesterol(LDL-C)wascalculatedusingthe Friedewald
formula.
2.3. MeasurementofADPRCA
IsolationofPBMCmembranousfraction
After obtaining written informed consent, peripheral blood
samples were collected. PBMCs were isolated using Ficoll-
Paque PLUS (Sigma) [33–35] and centrifuged. Isolated
PBMCsweresuspendedin10mMTris/HClsolution,pH7.4,
with 5mM MgCl2 (2 mL for each sample) at 4◦C for 30
minutes. The suspension was homogenized in a Teflon-glass
homogenizer; the resultant homogenate was centrifuged at
4◦C for 5 minutes at 1000×g to remove unbroken cells
and nuclei. Crude membrane fractions were prepared by
centrifugation (twice) of homogenates at 105000×g for 15
minutes. The supernatant was removed, and precipitates
weresuspendedin10mMTris/HClsolution,pH6.7[36,37].

Page 3

Michio Ohtsuji et al.3
For each experiment, membranes were freshly prepared and
used immediately for enzymatic reactions.
FluorometricmeasurementofADP-ribosylcyclaseactivity
ADPRCAwasdeterminedfluorometricallyusingatechnique
based on measurement of the conversion of β-NGD+into
the fluorescent product cGDP-ribose [36–39]. In brief,
2.5mL of reaction mixtures containing 60μM β-NGD+,
50mM Tris/HCl, pH 6.6, 100mM KCl, 10μM CaCl2, and
membranes (62–297μg of protein) were maintained at 37◦C
under constant stirring. The samples were then excited
at 300nm, and fluorescence emission was continuously
monitored at 410nm in a Shimadzu RF-5300PC spectroflu-
orophotometer (Kyoto, Japan). Activity was calculated from
thedatapointsrecordedduring5minutespermgmembrane
protein as reported previously [37].
2.4. Statisticalanalysis
All data are shown as mean ± SD. Continuous variables
were compared by one-way analyses of variance (ANOVA)
or covariance (ANCOVA) after being adjusted for age, BMI,
and sex. Normality of the distribution of Ln(ADPRCA) was
confirmed by Shapiro-Wilk’s W-test. Differences between
the two groups were compared by chi2analysis (categorical
variables) or nonparametric Mann-Whitney U-test (contin-
uous variables). Logistic regression analyses were performed
to clarify the clinical parameters contributing to categorical
variables. Stepwise multivariate regression analyses were
performed to clarify the clinical parameters contributing to
the level of continuous variable. All statistical analyses were
conducted with JMP 6.03 for Macintosh OS-X 10.5 (SAS
Institute Inc, Cary, NC, USA). A P-value of less than .05 was
considered statistically significant.
3.RESULTS
RelationshipbetweenADPRCAandclinicalparameters
In total, sixty Japanese diabetic patients (31 males and 29
females) and fifteen nondiabetic controls (6 males and 9
females) were enrolled in this study. The baseline char-
acteristics of the subjects are shown in Table 1. ADPRCA
showed statistically significant negative correlation with
the level of HbA1c (P = .040, R2= 0.073, Figure 1). As
shown in Table 2, no significant correlation was observed
between ADPRCA and other clinical parameters (level of
fasting plasma glucose, systolic and diastolic blood pressure,
serum total cholesterol, logarithm of serum TG, HDL-C,
LDL-C, age, duration of diabetes, BMI, gender difference,
type of diabetes (Figure 2(a)), or medication for diabetes
(Figure 2(b))).
ADPRCAinindividualswithdiabeticcomplications
Figure 3 shows the relationship between ADPRCA and
the diabetic vascular complications. ADPRCA was signif-
icantly lower in subjects with nephropathy than those
1.5
2
2.5
3
3.5
4
Ln ADPRC
45678910 1112
HbA1c
Figure 1: ADP-ribosyl cyclase activities and the level of HbA1c.
StatisticallysignificantrelationshipwasobservedbetweenADPRCA
and HbA1c (P = .040, R2= 0.073, Ln(ADPRC) =3.4277567–
0.093904∗HbA1c).
without (P = .0198) and nondiabetic controls (P = .0332)
(Figure 3(a)). Subjects with other complications also showed
similar tendencies; however, no significant correlations
betweenADPRCAandretinopathy,neuropathy,ormacroan-
giopathy were observed (Figures 3(b)–3(d)). Although
HbA1c level showed relationship with ADPRCA, no signif-
icant difference was observed in ADPRCA between subjects
with any diabetic complications (data not shown).
As HbA1c was significantly related to ADPRCA, we
examined whether HbA1c contributed to the relationship
betweenADPRCAandnephropathy.TheresultsofANCOVA
showed that the difference just failed to meet statistical
significance after adjusting for HbA1c (Table 3). Next,
to examine whether ADPRCA could be an independent
contributor to the existence of nephropathy, we performed
logisticanalysesadjustedforsystolicbloodpressure,diastolic
blood pressure, TG, HDL, LDL, gender, type of diabetes,
duration of diabetes, medication for diabetes, HbA1c, and
BMI. Systolic blood pressure and ADPRCA contributed
significantly to the existence of nephropathy (Table 4).
4.DISCUSSION
The main finding of this study was that diabetic subjects
with nephropathy showed decreased ADPRCA. However,
PBMCs in proinflammatory states like diabetic vasculopathy
might be relating to increased ADPRCA as several cytokines
including IL-8, IFN-gamma upregulate intracellular CD38
activity [17], our results interestingly showed decreased
ADPRCA in PBMCs. Logistic analysis revealed that only
systolic blood pressure and ADPRCA, but not HbA1c, were
significantly related to the incidence of nephropathy. There-
fore, contribution of HbA1c to the relationship between
ADPRCA and nephropathy should be considered small
in extent. ADPRCA’s correlation with nephropathy seems
reasonable.

Page 4

4Experimental Diabetes Research
1.5
2
2.5
3
3.5
4
Ln ADPRC
CTL Type 1 DM Type 2 DM
N.S.
N.S.N.S.
(a)
1.5
2
2.5
3
3.5
4
Ln ADPRC
DietInsulin OHA
N.S.
N.S.N.S.
(b)
Figure 2: (a) ADP-ribosyl cyclase activities and type of diabetes. No significant relationship was observed between ADPRCA and type of
diabetes. (b) ADP-ribosyl cyclase activities and medication for type 2 diabetes. No significant relationship was observed between ADPRCA
and medication for type 2 diabetes. CTL: control; OHA: oral hypoglycemic agent.
1.5
2
2.5
3
3.5
4
Ln ADPRC
CTLNephropathy (−)Nephropathy (+)
P = .0332
N.S. (P = .7942)
P = .0198
ANOVA: F-value = 3.5126, P-value = .0352
(a)
1.5
2
2.5
3
3.5
4
Ln ADPRC
CTLRetinopathy (−)Retinopathy (+)
N.S. (P = .1334)
N.S. (P = .5481)N.S. (P = .2543)
ANOVA: F-value = 1.2896, P-value = .2818
(b)
1.5
2
2.5
3
3.5
4
Ln ADPRC
CTL Neuropathy (−)Neuropathy (+)
N.S. (P = .1862)
N.S. (P = .4877)N.S. (P = .4427)
ANOVA: F-value = 0.9207, P-value = .4030
(c)
1.5
2
2.5
3
3.5
4
Ln ADPRC
CTLMacroangiopathy (−) Macroangiopathy (+)
ANOVA: F-value = 2.3347, P-value = .1043
(d)
N.S. (P = .0526)
N.S. (P = .5814) N.S. (P = .0704)
Figure 3: Relationship between logarithm of ADP-ribosyl-cyclase activity Ln(ADPRCA) and diabetic vascular complications. Figures show
relationship between nondiabetic control, subjects with each complication, and those without. (a) Subjects with nephropathy showed lower
ADPRCA than those without nephropathy (P = .0198) and nondiabetic controls (P = .0332). (b)–(d) No significant difference in ADPRCA
was observed in subjects with retinopathy, neuropathy, and macroangiopathy. CTL: control.

Page 5

Michio Ohtsuji et al.5
Table 1: Baseline characteristics of subjects. Results are expressed as mean ± S.D.
Type 1 diabetesType 2 diabetes
50
26/24
63.0 ±12.0
22.9 ±3.9
16.0 ±9.7
164 ± 54
7.6 ±1.3
Nondiabetic control
15
6/9
51.0 ±22.0
22.1 ±2.4
—
81 ± 12
4.6 ±0.7
Number of subjects
Gender (male/female)
Age (years)
BMI (kg/m2)
Duration of diabetes (years)
Fasting plasma glucose (mg/dl)
HbA1c (%)
Medication for diabetes
Diet alone
Oral hypoglycemic agents
Insulin
Diabetic complications
Nephropathy
Retinopathy
Neuropathy
Macroangiopathy
ADP-ribosyl cyclase activity (nmol/min/mg protein)
10
5/5
36.0 ± 14.2
21.4 ±2.6
11.0 ±9.2
244 ± 130
8.0 ± 1.5
—
—
10
4
30
16
—
—
—
3
4
4
0
20
22
25
18
16.9 ±7.516.6 ±7.6 17.8 ±4.6
Table 2: Correlation of ADP-ribosyl-cyclase activity with subject
parameters (calculated with logarithm-transformed TG).
Factor
Fasting plasma glucose
HbA1c
Systolic blood pressure
Diastolic blood pressure
Total cholesterol (TC)
HDL-C
LDL-C
Triglyceride (TG)
Duration of diabetes
Age
BMI
Gender
R2
P-value
NS
.040
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
0.0120
0.0731
0.0133
0.0301
0.0424
0.0042
0.0258
0.0588
0.0150
0.0367
0.0174
0.0151
Table 3: Comparison of ADPRCA influence on nephropathy,
adjusted for HbA1c.
Factor
Nephropathy (−)
Nephropathy (+)
Average
2.799
2.576
S.E.
0.076
0.095
95% C.I.
2.646–2.951
2.386–2.765
P-value
.075
First, to discuss the role of cADPR-mediated signals in
PBMCs, since ADPRCA could be stimulated by angiotensin-
II [37], kidney tissue with diabetic nephropathy could show
increased ADPRCA. Recent report by Kim et al. showed
increased ADPRCA in the kidney of STZ-induced diabetes
mice[18].AswemeasuredADPRCAinPBMCs,thisdiscrep-
ancy could be acceptable. Interestingly, the roles of CD38 on
PBMCs’effectonvascularcomplicationsarebidirectional.In
Table 4: Logistic analysis between nephropathy and parameters
in diabetic subjects. Carrier of complication = 1, noncarrier = 0;
medication: insulin =2, OHA = 1, diet = 0; gender: male = 1,
female = 0; type of diabetes: type 2 =1, type 1 = 0.
Factor
Systolic blood pressure
ADP-ribosyl-cyclase activity
Regression coefficient
−0.09115
2.61758
R2
P-value
.0140
.0229
6.040
3.707
PBMCs, binding of agonistic anti-CD38 antibodies, which
stimulate ADPRCA, induces release of proinflammatory
cytokines including IL-1, IL-6, and TNF-alpha over the short
term [40]. Cytokine release could make an important con-
tribution to inflammation responsible in the early stages of
diabeticvascularcomplications.Ontheotherhand,agonistic
CD38 ligation inhibits cell growth and induces apoptosis
in B-cell precursors [41] mediating phosphatidylinositol 3-
kinase signaling [42], although having stimulatory effects
on mature lymphocytes. The suppressive effect mediated by
CD38 was also observed in experiments with patient-derived
myeloid leukemia cells and with the murine cell line [43]. In
addition, CD38 expression has been reported in circulating
monocytes but not in resident macrophages and dendritic
cells [44, 45]. Differentiation of monocytes to macrophages
resultedinthedownregulationofsurfaceexpressionofCD38
[46]. CD38 is strongly expressed in lymphocyte precursors,
declined during differentiation, and then upregulated again
in mature plasma cells [47]. CD157 was suggested to display
a similar expression tendency in myeloid cells [48]. Since
hyperglycemia directly enhances protein ADP-ribosylation
in cultured neuroblastoma cells [49], resulting in increased
ADPRCA in diabetic subjects, we speculate that decreased
ADPRCA in PBMCs could reflect decreased suppressive

Page 6

6Experimental Diabetes Research
effects of CD38 and CD157 and increased numbers of
differentiated cells.
Second, let us consider the agonistic effects of autoan-
tibodies against CD38, which is shown to exert insulin
secretion from cultured human islets [50] through ADPRCA
activation. In humans, the majority of anti-CD38 autoanti-
bodies (∼60%) display agonistic properties [51, 52], which
demonstrate the capability to trigger Ca2+release in lym-
phocytic cell lines [53]. In agreement with these functional
features, the presence of anti-CD38 autoantibodies in type
2 diabetic patients was associated with significantly higher
levels of fasting plasma C-peptide and insulin, as compared
with anti-CD38 negative subjects. Thus, anti-CD38 autoim-
munity might indicate a relative protection against beta-
cell failure and a lower risk of insulin requirement [52, 54,
55]. Previous reports on the clinical characteristics of anti-
CD38 autoantibodies carriers have not gone into depth on
diabetic complications, although the possible exacerbation
of diabetic complications by the agonistic effect of anti-
CD38 autoantibodies on PBMCs was noted by Mallone et
al. [51]. However, several discussions should be needed for
the relationship between CD38 autoantibodies and diabetic
vasculopathy as agonistic CD38 autoantibodies possibly
stimulate both insulin rsecretion resulting in hyperinsuline-
mia, the prominent risk factor for diabetic macroangiopathy,
and angiotensin-II induced renal artery contraction. We just
speculate that those with anti-CD38 autoantibodies should
show lower frequencies of diabetic nephropathy, and other
vascular complications in long-term follow-up studies due
to better glycemic control and increased ADPRCA reflecting
less maturated PBMCs.
Third, as all of our subjects were diabetic, it behooves
us mention the Okamoto model of diabetes pathogenesis.
CD38-related signal has been well documented in dia-
betes mellitus [18, 56–58]. Our results showing decreased
ADPRCA being significantly correlated to increased HbA1c
is compatible with previous studies reflecting to some degree
the Okamoto model [59–61]. Interestingly, in our study,
no significant relationship was observed between ADPRCA
and fasting plasma glucose. We suggest that ADPRCA
might be more related to postprandial than fasting plasma
glucose level, the former being strongly responsive to insulin
secretion [27, 29].
Finally, the clinical significance of our results: CD38
expression, representative of ADP-ribosyl cyclase, is already
used clinically as a prognostic marker for HIV infected
subjects [62, 63] and chronic lymphoid leukemia (CLL)
[13, 56, 64]. Since other ADP ribosyl cyclases including
CD157 reside on the cell surface of PBMCs, restricting
examination to CD38 would not entirely cover cADPR-
mediated signaling impairment. Measuring ADPRCA would
thus be a better approach in flagging up groups at risk
for diabetic nephropathy from among general diabetic
subjects.WeenvisionADPRCAmeasurementbeingsimilarly
introduced into clinical settings for detecting subjects at
high-risk for diabetic nephropathy and thus prompting early
intervention to prevent ESRD.
This study had a number of limitations. First, we did
not have data on anti-CD38 autoantibodies. ADPRCA in
each subject is partly determined by preexisting factors inde-
pendentofdiabetes,includingCD38geneticpolymorphisms
and autoantibodies against CD38. We speculate that higher
ADPRCA values might reflect possession of anti-CD38
autoantibodies. Second, we did not examine any markers of
lymphocyte or monocyte maturation/differentiation. Both
should be examined in further studies.
In conclusion, our findings suggest that decreased
ADPRCA in PBMCs is significantly correlated with diabetic
nephropathy and that measurement of ADPRCA in PBMCs
could serve as an important marker associated with diabetic
nephropathy. Further prospective studies should be intro-
duced to clarify the mechanism and predictive significance
of decreased ADPRCA in diabetic complications.
ACKNOWLEDGMENTS
The authors would like to thank Mrs. Reiko Ikeda for
her technical assistance in data analyses and comments
regarding this manuscript. K. Yagi and M. Shintaku-Kubota
contributed equally to this work.
REFERENCES
[1] M. Brownlee, “Biochemistry and molecular cell biology of
diabetic complications,” Nature, vol. 414, no. 6865, pp. 813–
820, 2001.
[2] W. F. Keane, K. Kurokawa, P. A. Lyle, et al., “Treatment
of type 2 diabetic patients with kidney disease with AT1-
receptor antagonists: lessons from recent trials,” Clinical and
Experimental Nephrology, vol. 6, no. 4, pp. 175–181, 2002.
[3] T. Usami, K. Koyama, O. Takeuchi, K. Morozumi, and G.
Kimura, “Regional variations in the incidence of end-stage
renal failure in Japan,” The Journal of the American Medical
Association, vol. 284, no. 20, pp. 2622–2624, 2000.
[4] K. Kurokawa, J. C. N. Chan, M. E. Cooper, W. F. Keane,
S. Shahinfar, and Z. Zhang, “Renin angiotensin aldosterone
system blockade and renal disease in patients with type 2
diabetes: a subanalysis of Japanese patients from the RENAAL
study,” Clinical and Experimental Nephrology, vol. 10, no. 3,
pp. 193–200, 2006.
[5] M.-L. Gross, R. Dikow, and E. Ritz, “Diabetic nephropathy:
recent insights into the pathophysiology and the progression
of diabetic nephropathy,” Kidney International, vol. 67, sup-
plement 94, pp. S50–S53, 2005.
[6] H. Abrahamian, G. Endler, M. Exner, et al., “Association of
low-grade inflammation with nephropathy in type 2 diabetic
patients: role of elevated CRP-levels and 2 different gene-
polymorphisms of proinflammatory cytokines,” Experimental
and Clinical Endocrinology & Diabetes, vol. 115, no. 1, pp. 38–
41, 2007.
[7] B. A. Young, R. J. Johnson, C. E. Alpers, et al., “Cellular
events in the evolution of experimental diabetic nephropathy,”
Kidney International, vol. 47, no. 3, pp. 935–944, 1995.
[8] C.-G. Ihm, J.-K. Park, S.-P. Hong, et al., “A high glucose con-
centration stimulates the expression of monocyte chemotactic
peptide 1 in human mesangial cells,” Nephron, vol. 79, no. 1,
pp. 33–37, 1998.
[9] P. Cortes, X. Zhao, B. L. Riser, and R. G. Narins, “Role of
glomerular mechanical strain in the pathogenesis of diabetic
nephropathy,” Kidney International, vol. 51, no. 1, pp. 57–68,
1997.

Page 7

Michio Ohtsuji et al.7
[10] J. F. Navarro-Gonz´ alez and C. Mora-Fern´ andez, “The role of
inflammatory cytokines in diabetic nephropathy,” Journal of
the American Society of Nephrology, vol. 19, no. 3, pp. 433–442,
2008.
[11] S. Okada, K. Shikata, M. Matsuda, et al., “Intercellular
adhesion molecule-1-deficient mice are resistant against renal
injuryafterinductionofdiabetes,”Diabetes,vol.52,no.10,pp.
2586–2593, 2003.
[12] H. Sugimoto, K. Shikata, K. Hirata, et al., “Increased
expression of intercellular adhesion molecule-1 (ICAM-1) in
diabeticratglomeruli:glomerularhyperfiltrationisapotential
mechanismofICAM-1upregulation,”Diabetes,vol.46,no.12,
pp. 2075–2081, 1997.
[13] H. C. Lee, “Physiological functions of cyclic ADP-ribose
and NAADP as calcium messengers,” Annual Review of
Pharmacology and Toxicology, vol. 41, pp. 317–345, 2001.
[14] M. Howard, J. C. Grimaldi, J. F. Bazan, et al., “Formation
and hydrolysis of cyclic ADP-ribose catalyzed by lymphocyte
antigen CD38,” Science, vol. 262, no. 5136, pp. 1056–1059,
1993.
[15] S. Takasawa, A. Tohgo, N. Noguchi, et al., “Synthesis and
hydrolysis of cyclic ADP-ribose by human leukocyte antigen
CD38 and inhibition of the hydrolysis by ATP,” The Journal of
Biological Chemistry, vol. 268, no. 35, pp. 26052–26054, 1993.
[16] E. Zocchi, L. Franco, L. Guida, et al., “A single protein
immunologically identified as CD38 displays NAD+glycohy-
drolase, ADP-ribosyl cyclase and cyclic ADP-ribose hydrolase
activities at the outer surface of human erythrocytes,” Bio-
chemical and Biophysical Research Communications, vol. 196,
no. 3, pp. 1459–1465, 1993.
[17] F. Malavasi, S. Deaglio, A. Funaro, et al., “Evolution and
function of the ADP ribosylcyclase/CD38 gene family in
physiology and pathology,” Physiological Reviews, vol. 88, no.
3, pp. 841–886, 2008.
[18] S.-Y. Kim, R. Gul, S.-Y. Rah, et al., “Molecular mechanism of
ADP-ribosyl cyclase activation in angiotensin II signaling in
murine mesangial cells,” American Journal of Physiology, vol.
294, no. 4, pp. F982–F989, 2008.
[19] M. J. Berridge, P. Lipp, and M. D. Bootman, “The versatility
and universality of calcium signalling,” Nature Reviews Molec-
ular Cell Biology, vol. 1, no. 1, pp. 11–21, 2000.
[20] E. Carafoli, L. Santella, D. Branca, and M. Brini, “Generation,
control, and processing of cellular calcium signals,” Critical
Reviews in Biochemistry and Molecular Biology, vol. 36, no. 2,
pp. 107–260, 2001.
[21] H. Higashida, M. Hashii, S. Yokoyama, N. Hoshi, K. Asai, and
T. Kato, “Cyclic ADP-ribose as a potential second messenger
for neuronal Ca2+signaling,” Journal of Neurochemistry, vol.
76, no. 2, pp. 321–331, 2001.
[22] A. H. Guse, “Cyclic ADP-ribose (cADPR) and nicotinic acid
adeninedinucleotidephosphate(NAADP):novelregulatorsof
Ca2+-signalingandcellfunction,” CurrentMolecularMedicine,
vol. 2, no. 3, pp. 273–282, 2002.
[23] H. C. Lee, Cyclic ADP-Ribose and NAADP: Structure,
Metabolism and Functions, Springer, Dordrecht, The Nether-
lands, 2001.
[24] F.-X. Boittin, M. Dipps, N. P. Kinnear, A. Galione, and A.
M. Evans, “Vasodilation by the calcium-mobilizing messenger
cyclic ADP-ribose,” The Journal of Biological Chemistry, vol.
278, no. 11, pp. 9602–9608, 2003.
[25] D. A. Deshpande, T. F. Walseth, R. A. Panettieri, and M. S.
Kannan, “CD38/cyclic ADP-ribose-mediated Ca2+signaling
contributes to airway smooth muscle hyper-responsiveness,”
The FASEB Journal, vol. 17, no. 3, pp. 452–454, 2003.
[26] T.-S. Nam, H. C. Sung, S.-Y. Rah, et al., “Discovery of
a small-molecule inhibitor for kidney ADP-ribosyl cyclase:
implication for intracellular calcium signal mediated by cyclic
ADP-ribose,” Experimental and Molecular Medicine, vol. 38,
no. 6, pp. 718–726, 2006.
[27] B. J. Kim, K. H. Park, C. Y. Yim, et al., “Generation of nicotinic
acidadeninedinucleotidephosphateandcyclicADP-riboseby
glucagon-like peptide-1 evokes Ca2+signal that is essential for
insulin secretion in mouse pancreatic islets,” Diabetes, vol. 57,
no. 4, pp. 868–878, 2008.
[28] E. G. Teggatz, G. Zhang, A. Y. Zhang, et al., “Role of cyclic
ADP-ribose in Ca2+-induced Ca2+release and vasoconstric-
tion in small renal arteries,” Microvascular Research, vol. 70,
no. 1-2, pp. 65–75, 2005.
[29] S. Bruzzone, A. De Flora, C. Usai, R. Graeff, and H.
C. Lee, “Cyclic ADP-ribose is a second messenger in the
lipopolysaccharide-stimulated proliferationof humanperiph-
eral blood mononuclear cells,” The Biochemical Journal, vol.
375, part 2, pp. 395–403, 2003.
[30] J.R.Gavin,K.G.M.M.Alberti,M.B.Davidson,etal.,“Report
of the expert committee on the diagnosis and classification of
diabetesmellitus,”DiabetesCare,vol.20,no.7,pp.1183–1197,
1997.
[31] K. G. M. M. Alberti and P. Z. Zimmet, “Definition, diagnosis
and classification of diabetes mellitus and its complications—
part 1: diagnosis and classification of diabetes mellitus.
ProvisionalreportofaWHOconsultation,”DiabeticMedicine,
vol. 15, no. 7, pp. 539–553, 1998.
[32] M. Buysschaert, A.-S. Dramais, P. E. Wallemacq, and M.
P. Hermans, “Hyperhomocysteinemia in type 2 diabetes:
relationship to macroangiopathy, nephropathy, and insulin
resistance,”DiabetesCare,vol.23,no.12,pp.1816–1822,2000.
[33] A. Bøyum, “Isolation of mononuclear cells and granulocytes
from human blood. Isolation of monuclear cells by one cen-
trifugation, and of granulocytes by combining centrifugation
andsedimentationat1g,”ScandinavianJournalofClinicaland
Laboratory Investigation. Supplement, vol. 97, pp. 77–89, 1968.
[34] A. Bøyum, “Isolation of leucocytes from human blood.
Further observations. Methylcellulose, dextran, and ficoll
as erythrocyteaggregating agents,” Scandinavian Journal of
Clinical and Laboratory Investigation. Supplement, vol. 97, pp.
31–50, 1968.
[35] A. Bøyum, “Isolation of lymphocytes, granulocytes and
macrophages,” Scandinavian Journal of Immunology. Supple-
ment, vol. 5, supplement 5, pp. 9–15, 1976.
[36] H. Higashida, A. Egorova, C. Higashida, et al., “Sympathetic
potentiation of cyclic ADP-ribose formation in rat cardiac
myocytes,” The Journal of Biological Chemistry, vol. 274, no.
47, pp. 33348–33354, 1999.
[37] H. Higashida, J.-S. Zhang, M. Hashii, M. Shintaku, C.
Higashida, and Y. Takeda, “Angiotensin II stimulates cyclic
ADP-ribose formation in neonatal rat cardiac myocytes,” The
Biochemical Journal, vol. 352, part 1, pp. 197–202, 2000.
[38] R. M. Graeff, T. F. Walseth, K. Fryxell, W. D. Branton, and
H. C. Lee, “Enzymatic synthesis and characterizations of
cyclic GDP-ribose. A procedure for distinguishing enzymes
with ADP-ribosyl cyclase activity,” The Journal of Biological
Chemistry, vol. 269, no. 48, pp. 30260–30267, 1994.
[39] R. M. Graeff, L. Franco, A. De Flora, and H. C. Lee, “Cyclic
GMP-dependent and -independent effects on the synthesis
of the calcium messengers cyclic ADP-ribose and nicotinic
acidadeninedinucleotidephosphate,”TheJournalofBiological
Chemistry, vol. 273, no. 1, pp. 118–125, 1998.

Page 8

8Experimental Diabetes Research
[40] C. M. Ausiello, A. La Sala, C. Ramoni, F. Urbani, A. Funaro,
and F. Malavasi, “Secretion of IFN-γ, IL-6, granulocyte-
macrophage colony-stimulating factor and IL-10 cytokines
after activation of human purified T lymphocytes upon CD38
ligation,” Cellular Immunology, vol. 173, no. 2, pp. 192–197,
1996.
[41] M.-A. Kumagai, E. Coustan-Smith, D. J. Murray, et al.,
“Ligation of CD38 suppresses human B lymphopoiesis,” The
Journal of Experimental Medicine, vol. 181, no. 3, pp. 1101–
1110, 1995.
[42] A. Kitanaka, C. Ito, H. Nishigaki, and D. Campana, “CD38-
mediated growth suppression of B-cell progenitors requires
activation of phosphatidylinositol 3-kinase and involves its
association with the protein product of the c-cbl proto-
oncogene,” Blood, vol. 88, no. 2, pp. 590–598, 1996.
[43] E. Todisco, T. Suzuki, K. Srivannaboon, et al., “CD38 ligation
inhibitsnormalandleukemicmyelopoiesis,”Blood,vol.95,no.
2, pp. 535–542, 2000.
[44] M.-T. Zilber, S. Gregory, R. Mallone, et al., “CD38
expressed on human monocytes: a coaccessory molecule in
the superantigen-induced proliferation,” Proceedings of the
National Academy of Sciences of the United States of America,
vol. 97, no. 6, pp. 2840–2845, 2000.
[45] T. Musso, S. Deaglio, L. Franco, et al., “CD38 expression and
functional activities are up-regulated by IFN-γ on human
monocytes and monocytic cell lines,” Journal of Leukocyte
Biology, vol. 69, no. 4, pp. 605–612, 2001.
[46] M. Pfister, A. Ogilvie, C. P. da Silva, A. Grahnert, A. H. Guse,
and S. Hauschildt, “NAD degradation and regulation of CD38
expression by human monocytes/macrophages,” European
Journal of Biochemistry, vol. 268, no. 21, pp. 5601–5608, 2001.
[47] M. Alessio, S. Roggero, A. Funaro, et al., “CD38 molecule:
structural and biochemical analysis on human T lymphocytes,
thymocytes,andplasmacells,”TheJournalofImmunology,vol.
145, no. 3, pp. 878–884, 1990.
[48] K. M. McNagny, P.-A. Cazenave, and M. D. Cooper, “BP-
3 alloantigen. A cell surface glycoprotein that marks early B
lineage cells and mature myeloid lineage cells in mice,” The
Journal of Immunology, vol. 141, no. 8, pp. 2551–2556, 1988.
[49] A. M. Di Giulio, E. Lesma, E. Germani, and A. Gorio, “Inhibi-
tion of high glucose-induced protein mono-ADP-ribosylation
restores neuritogenesis and sodium-pump activity in SY5Y
neuroblastoma cells,” Journal of Neuroscience Research, vol. 57,
no. 5, pp. 663–669, 1999.
[50] C. Pupilli, S. Giannini, P. Marchetti, et al., “Autoantibodies
to CD38 (ADP-ribosyl cyclase/cyclic ADP-ribose hydrolase)
in Caucasian patients with diabetes: effects on insulin release
from human islets,” Diabetes, vol. 48, no. 12, pp. 2309–2315,
1999.
[51] R. Mallone, E. Ortolan, G. Baj, et al., “Autoantibody response
toCD38inCaucasianpatientswithtype1andtype2diabetes:
immunological and genetic characterization,” Diabetes, vol.
50, no. 4, pp. 752–762, 2001.
[52] A. Antonelli and E. Ferrannini, “CD38 autoimmunity: recent
advances and relevance to human diabetes,” Journal of
Endocrinological Investigation, vol. 27, no. 7, pp. 695–707,
2004.
[53] A. Antonelli, G. Baj, P. Marchetti, et al., “Human anti-
CD38 autoantibodies raise intracellular calcium and stimulate
insulin release in human pancreatic islets,” Diabetes, vol. 50,
no. 5, pp. 985–991, 2001.
[54] A. Antonelli, T. Tuomi, M. Nannipieri, et al., “Autoimmunity
to CD38 and GAD in type I and type II diabetes: CD38 and
HLAgenotypesandclinicalphenotypes,”Diabetologia,vol.45,
no. 9, pp. 1298–1306, 2002.
[55] R. Mallone, E. Ortolan, S. Pinach, et al., “Anti-CD38
autoantibodies: characterisation in new-onset type I diabetes
and latent autoimmune diabetes of the adult (LADA) and
comparisonwithotherisletautoantibodies,”Diabetologia,vol.
45, no. 12, pp. 1667–1677, 2002.
[56] S. Takasawa, K. Nata, H. Yonekura, and H. Okamoto, “Cyclic
ADP-ribose in insulin secretion from pancreatic β cells,”
Science, vol. 259, no. 5093, pp. 370–373, 1993.
[57] H. Okamoto, S. Takasawa, and K. Nata, “The CD38-cyclic
ADP-ribose signalling system in insulin secretion: molecular
basis and clinical implications,” Diabetologia, vol. 40, no. 12,
pp. 1485–1491, 1997.
[58] N.-H. An, M.-K. Han, C. Um, et al., “Significance of ecto-
cyclase activity of CD38 in insulin secretion of mouse
pancreatic islet cells,” Biochemical and Biophysical Research
Communications, vol. 282, no. 3, pp. 781–786, 2001.
[59] H.OkamotoandS.Takasawa,“Recentadvancesinphysiologi-
calandpathologicalsignificanceofNAD+metabolites:rolesof
poly(ADP-ribose) and cyclic ADP-ribose in insulin secretion
and diabetogenesis,” Nutrition Research Reviews, vol. 16, no. 2,
pp. 253–266, 2003.
[60] H. Okamoto and S. Takasawa, “Recent advances in the
Okamoto model: the CD38-cyclic ADP-ribose signal system
and the regenerating gene protein (Reg)-Reg receptor system
in β-cells,” Diabetes, vol. 51, supplement 3, pp. S462–S473,
2002.
[61] H. Okamoto, S. Takasawa, K. Nata, I. Kato, A. Tohgo, and N.
Noguchi, “Physiological and pathological significance of the
CD38-cyclic ADP-ribose signaling system,” in Human CD38
and Related Molecules, K. Mehta and F. Malavasi, Eds., vol.
75 of Chemical Immunology, pp. 121–145, S. Karger, Basel,
Switzerland, 2000.
[62] Z. Liu, W. G. Cumberland, L. E. Hultin, H. E. Prince, R.
Detels, and J. V. Giorgi, “Elevated CD38 antigen expression
on CD8+T cells is a stronger marker for the risk of chronic
HIV disease progression to AIDS and death in the multicenter
AIDS Cohort study than CD4+cell count, soluble immune
activation markers, or combinations of HLA-DR and CD38
expression,” Journal of Acquired Immune Deficiency Syndromes
and Human Retrovirology, vol. 16, no. 2, pp. 83–92, 1997.
[63] J. V. Giorgi, H.-N. Ho, K. Hirji, et al., “CD8+lymphocyte
activation at human immunodeficiency virus type 1 sero-
conversion: development of HLA-DR+CD38- CD8+cells
is associated with subsequent stable CD4+cell levels,” The
Journal of Infectious Diseases, vol. 170, no. 4, pp. 775–781,
1994.
[64] S. Deaglio, S. Aydin, T. Vaisitti, L. Bergui, and F. Malavasi,
“CD38 at the junction between prognostic marker and
therapeutic target,” Trends in Molecular Medicine, vol. 14, no.
5, pp. 210–218, 2008.