Full text of DOI:10.1053/ajkd.2001.26839

Plasma Levels and Metabolism of AcSDKP in Patients With Chronic
Renal Failure: Relationship With Erythropoietin Requirements
Yannick Le Meur, MD, PhD, Vale´rie Lorgeot, PhD, Lydie Comte, PhD,
Jean-Christophe Szelag, MD, Jean-Claude Aldigier, MD, Claude Leroux-Robert, MD,
and Vincent Praloran, MD, PhD
● N-acetyl-seryl-aspartyl-lysyl-proline (AcSDKP) is a physiological inhibitor of hematopoiesis that is maintained at
stable levels in normal plasma. Its degradation in vivo and in vitro by angiotensin-converting enzyme (ACE)
accounts for the high plasma concentrations of AcSDKP in patients treated with ACE inhibitors. Because ACE
inhibitors can induce anemia in some patients, we measured plasma AcSDKP concentrations in 176 patients with
chronic renal failure: 120 hemodialysis (HD) and 56 nondialysis (nD) patients, 39 of whom were administered ACE
inhibitors. We studied the relationships between AcSDKP levels, hematologic parameters, and recombinant human
erythropoietin (rHuEPO) requirements in these patients. AcSDKP levels were significantly greater in HD (10.3 ؎ 3.9
pmol/mL) and nD (3.1 ؎ 1.8 pmol/mL) patients not administered ACE inhibitors than controls (1.8 ؎ 0.2 pmol/mL). In
all patients, treatment with ACE inhibitors significantly increased these levels fourfold. HD sessions significantly
decreased AcSDKP concentrations by 66% and reduced the predialysis in vitro half-life of AcSDKP (270 ؎ 109
minutes) to values (182 ؎ 67 minutes) not significantly different from those of controls or nD patients. Most HD
patients treated with ACE inhibitors had AcSDKP levels greater than 24 pmol/mL (the greatest concentration found
in other nD and HD patients). Only in this group of patients did weekly doses of rHuEPO correlate with AcSDKP
levels. Our results show that renal function is essential to maintain stable AcSDKP plasma levels, and at high levels,
AcSDKP acts as a uremic toxin causing partial resistance to erythropoietin and inhibiting erythropoiesis.
© 2001 by the National Kidney Foundation, Inc.
INDEX WORDS: N-acetyl-seryl-aspartyl-lysyl-proline (AcSDKP); anemia; chronic renal failure (CRF); hemodialysis (HD);
erythropoietin (EPO); angiotensin-converting enzyme (ACE) inhibitors.
lymphocytes.4-6 This tetrapeptide is produced in
Editorial, p. 649
vitro by multiple cell types and is maintained in
normal plasma at stable concentrations.7,8 Angio-
tensin-converting enzyme (ACE) degrades
-acetyl-seryl-aspartyl-lysyl-proline (AcSDKP)
is a physiological negative regulator of the
AcSDKP both in vitro and in vivo.9-11
N
In healthy volunteers treated for 15 days with
an ACE inhibitor (enalapril, 20 mg/d), plasma
concentrations and daily urinary excretion of
AcSDKP were permanently increased four- to
fivefold.12 Interestingly, this treatment modified
concentrations of hematopoietic progenitors phys-
iologically present in blood.12 High plasma lev-
els ofAcSDKP also have been found in hyperten-
sive patients administered an ACE inhibitor.13
Because the molecular weight of AcSDKP (487
d) is as low as that of creatinine, we measured its
concentration in plasma of patients with chronic
renal failure (CRF) who were or were not admin-
istered anACE inhibitor.14 This preliminary study
showed that patients with CRF had significantly
greater plasmaAcSDKP levels than controls, and
levels were even greater in ACE inhibitor–
treated patients.
proliferation of hematopoietic stem cells. In vivo,
it protects murine hematopoietic stem cells and
spleen colony-forming units from the toxicity of
irradiation or cytosine arabinoside by blocking
their S phase entry.1-3 In vitro, it inhibits the
proliferation of human and murine hematopoi-
etic stem cells and progenitor cells, as well as
From the Service de Ne´phrologie et d’He´modialyse,
Hoˆpital Universitaire Dupuytren; and the Laboratoire
d’He´matologie Expe´rimentale, Faculte´ de Me´decine, Limo-
ges, France.
Received November 29, 2000; accepted in revised form
April 6, 2001.
Y.L.M and V.L. collaborated equally in the work of this
article.
Supported in part by the Conseil Re´gional du Limousin.
Address reprint requests to Vincent Praloran, MD, PhD,
Laboratoire Universitaire d’He´matologie, Bat 1A, Univer-
site´ Victor Segalen Bordeaux 2, 146 rue Le´o Saignat, 33076
Bordeaux cedex, Limoges, France. E-mail: vincent.
praloran@hemato.u-bordeaux2.fr
© 2001 by the National Kidney Foundation, Inc.
0272-6386/01/3803-0008$35.00/0
doi:10.1053/ajkd.2001.26839
Anemia is an almost constant finding in pa-
tients with CRF. Although it is mainly related to
insufficient erythropoietin (EPO) production,15
numerous other factors, such as erythropoietic
inhibitors,16-19 have been shown or suspected to
510
American Journal of Kidney Diseases, Vol 38, No 3 (September), 2001: pp 510-517

RELATIONSHIP OF AcSDKP LEVELS WITH EPO TREATMENT
511
collected in heparinized and EDTA tubes for the determina-
tion ofAcSDKP concentrations and blood cell counts, respec-
tively. Heparinized plasma was rapidly collected after centrif-
ugation (1,500g for 10 minutes at 4°C) of blood samples and
immediately stored at –20°C until use.
have a role. Among them, ACE inhibitors, suc-
cessfully used in the treatment of polycythemia
in kidney graft recipients,20 have been reported
to induce anemia in some hypertensive pa-
tients,21 as well as in some renal transplant recipi-
ents treated for hypertension.22 ACE inhibitors
are also suspected to be responsible for the
anemia and resistance to recombinant human
EPO (rHuEPO) therapy in some hemodialysis
(HD) patients (reviewed in23).
These data led us to measure plasma AcSDKP
levels in patients with CRF who were or were not
administered anACE inhibitor and correlate these
levels with their biological parameters and
rHuEPO requirements.
AcSDKP Assays and Half-Life of
Synthetic AcSDKP
As previously described,7 AcSDKP concentrations were
assessed by a specific competitive enzyme immunoassay
(limit of detection, 0.1 pmol/mL) on the methanol-extracted
plasma fraction. Methanol extraction eliminates high-
molecular-weight molecules that could interfere with the
immunoassay and concentrates low-molecular-weight pep-
tides, including AcSDKP.
To measure the half-life, freshly collected plasma from 13
controls, 20 nD patients, and 13 HD patients (before and
after dialysis), all without treatment with an ACE inhibitor,
was supplemented with synthetic AcSDKP (20 pmol/mL)
and rapidly aliquoted. Methanol (3/1) was immediately
added to one aliquot, whereas the others were incubated at
37°C for 1, 2, and 4 hours before methanol addition and
dosage of the methanol-extracted plasma fraction. The per-
centage of degraded AcSDKP at the different time points
was calculated as follows:
PATIENTS AND METHODS
Patients
One hundred seventy-six patients with CRF and 51 healthy
donors used as controls were included in the study after
obtaining informed consent. Fifty-six patients with CRF did
not require dialysis (nondialysis [nD] patients) and 120
patients underwent HD three times weekly for at least 6
months. Twenty-one of 56 nD patients and 18 of 120 HD
patients were administered various daily doses of an ACE
inhibitor: benazepril, 2.5 mg (n ϭ 1), 5 mg (n ϭ 4), or 10 mg
(n ϭ 3); captopril, 12.5 mg (n ϭ 5), 25 mg (n ϭ 6), or 50 mg
(n ϭ 1); enalapril, 5 mg (n ϭ 4) or 10 mg (n ϭ 2);
perindopril, 1 mg (n ϭ 2), 2 mg (n ϭ 2), or 4 mg (n ϭ 1);
quinapril, 20 mg (n ϭ 1); ramipril, 5 mg (n ϭ 3); or
trandolapril, 0.5 mg (n ϭ 1) or 2 mg (n ϭ 3). Causes of CRF
in nD patients were chronic glomerulonephritis (18 pa-
tients), chronic interstitial nephritis (9 patients), vascular
nephropathy (10 patients), diabetic nephropathy (8 patients),
polycystic disease (3 patients), and undetermined nephropa-
thy (8 patients). Serum creatinine concentrations of nD
patients ranged from 99 to 1,203 ␮mol/L (median, 310
␮mol/L), and creatinine clearances, calculated using Cock-
roft’s formula, ranged from 4 to 82 mL/min (median, 19
mL/min). Creatinine clearance values were not significantly
different between nD patients who were or were not admin-
istered an ACE inhibitor (21 Ϯ 16 versus 25 Ϯ 18 mL/min).
No nD patients were administered rHuEPO. Most HD pa-
tients (92 of 120 patients) were continuously treated with
rHuEPO at a mean dose of 119 IU/kg/wk (range, 20 to 290
IU/kg/wk). Dialysis quality was determined by urea reduc-
tion ratio, calculated as:
100 – (AcSDKP after incubation/
AcSDKP before incubation) ϫ 100
Results were used to calculate the half-life of AcSDKP in
patients and controls, a parameter that reflects plasma ACE
biological activity.
Biological Parameters
A blood cell count and differential were performed on all
patients. Serum ferritin, parathyroid hormone (PTH), and
C-reactive protein (CRP) were measured in HD patients
only. We thus selected 76 HD patients who had stable
rHuEPO requirements and biological parameters for at least
the previous 3 months and hemoglobin (Hb) levels greater
than 10 g/dL, serum ferritin levels greater than 100 ng/mL,
CRP levels less than 20 mg/L, PTH levels less than 400
ng/L, and urea reduction ratios greater than 65% (Table 1).
Among these 76 HD patients, 13 were administered an ACE
inhibitor. Their Hb concentration (11.5 Ϯ 1.08 g/dL) was not
significantly different from that of the 63 other HD patients
not administered an ACE inhibitor (11.64 Ϯ 1.2 g/dL).
(Predialysis blood urea nitrogen concentration –
postdialysis blood urea nitrogen concentration)/
predialysis blood urea nitrogen concentration
Statistics
Results are expressed as mean Ϯ SD. Student’s unpaired
and paired t-tests or Mann-Whitney test was used when
appropriate to compare data. Differences were considered
significant at P less than 0.05. Coefficients of correlation
were calculated using linear regression.
Plasma Samples
Blood drawn by venipuncture in nD patients and controls
and from the arterial needle of the fistula in HD patients was

512
LE MEUR ET AL
Table 1. Major Biological Parameters and rHuEPO Requirements of 76 HD Patients
rHuEPO Requirement
(IU/kg/wk)
Serum Ferritin
(ng/mL)
CRP
(mg/L)
PTH
(ng/L)
Urea Reduction Ratio
(%)
Hb (g/dL)
11.6 Ϯ 1.2
84 Ϯ 64
286 Ϯ 156
12.5 Ϯ 8.5
98 Ϯ 89
72 Ϯ 5.4
NOTE. These patients had stable values for the mentioned biological parameters and rHuEPO treatment for at least the
previous 3 months. Results are expressed as mean Ϯ SD.
RESULTS
Plasma AcSDKP Levels in HD Patients
Before and After HD
Relations Between Plasma AcSDKP Levels and
Other Biological Parameters in nD Patients
Plasma AcSDKP concentrations were signifi-
cantly greater (P Ͻ 0.01) in HD patients before
HD (10.3 Ϯ 3.9 pmol/mL) than in nD patients or
controls and were even greater (P Ͻ 0.01) in
ACE inhibitor–treated HD patients (43.5 Ϯ 32
pmol/mL; Fig 1). In each of the 25 patients
tested, postdialysis levels of AcSDKP decreased
by 47% to 92% (mean, 66% Ϯ 11%) compared
with their predialysis levels. This difference was
significant (P Ͻ 0.01). However, postdialysis
Plasma AcSDKP concentrations were signifi-
cantly greater (P Ͻ 0.01) in nD patients not
administered an ACE inhibitor (3.1 Ϯ 1.8 pmol/
mL) than in controls (1.8 Ϯ 0.2 pmol/mL; Fig 1).
ACE inhibitor–treated nD patients had signifi-
cantly (P Ͻ 0.01) increased levels of AcSDKP
that were fourfold (13.9 Ϯ 9.9 pmol/mL) those
of nD patients not administered an ACE inhibi-
tor. This ACE inhibitor–induced increase was
similar to that we previously evidenced in healthy
volunteers treated with enalapril.12
AcSDKP concentration was inversely corre-
lated to creatinine clearance in patients not admin-
istered an ACE inhibitor (r ϭ –0.58; P Ͻ 0.01;
Fig 2A), but not in those administered an ACE
inhibitor (r ϭ –0.29; P ϭ 0.19; graph not shown).
As previously published,15 we also found a direct
correlation between Hb concentration and creati-
nine clearance in our 56 nD patients (with and
without ACE inhibitor treatment; r ϭ 0.62; P Ͻ
0.01). When these two groups of patients were
analyzed separately (Fig 2B), this correlation
was still significant in both non–ACE inhibitor–
treated and ACE inhibitor–treated patients,
whereas their AcSDKP plasma levels were sig-
nificantly different (Fig 1). Because there was an
inverse correlation between AcSDKP level and
creatinine clearance in nD patients not adminis-
tered an ACE inhibitor, the inverse correlation
that we observed in this group between concen-
trations of AcSDKP and Hb was expected (r ϭ
–0.53; P Ͻ 0.01; Fig 2C).
Between patients with and withoutACE inhib-
itor treatment, no significant differences were
found in mean Hb concentration (10.6 Ϯ 2.7
versus 11.0 Ϯ 2.2 g/dL), total white blood cell
count (5,270 Ϯ 1,370 versus 4,690 Ϯ 2,100 ϫ
109/L), or lymphocyte count (1,840 Ϯ 850 ver-
sus 1,850 Ϯ 1,650 ϫ 109/L), respectively.
Fig 1. Plasma AcSDKP concentrations in controls
(C) and nD or HD patients before dialysis sessions
with or without treatment with an ACE inhibitor (ACEI)
were measured by enzyme immunoassay. Horizontal
bars represent the mean of each group. Differences
between controls and all other groups were significant
(P < 0.01). The significance of other differences is
indicated on the graph.

RELATIONSHIP OF AcSDKP LEVELS WITH EPO TREATMENT
513
Half-Life of Synthetic AcSDKP
The half-life of synthetic AcSDKP in plasma
was similar in nD patients not administered an
ACE inhibitor (135 Ϯ 37 minutes) and controls
(138 Ϯ 40 minutes; Fig 3). Conversely, its half-
life was significantly longer (P Ͻ 0.05) before
dialysis (270 Ϯ 109 minutes) in HD patients not
administered an ACE inhibitor than in controls.
After dialysis, it diminished to levels (182 Ϯ 67
minutes) not significantly different from those of
controls or nD patients (Fig 3). Because we had
previously shown that long-term treatment with
an ACE inhibitor (enalapril) dramatically in-
creased the half-life of AcSDKP in plasma of
healthy controls,12 we did not perform this experi-
ment in the ACE inhibitor–treated nD and HD
patients.
The degradation of AcSDKP by ACE leads to
the formation of the dipeptide Lys-Pro, shown to
be a potential inhibitor ofACE activity.24 Consid-
ering its low molecular weight, the dipeptide
Lys-Pro probably is cleared mainly by the kid-
ney. When its concentration increases between
two dialysis sessions, it could then act as an ACE
inhibitor in predialysis plasma, thus accounting
for the longer half-life of synthetic AcSDKP in
HD patients before dialysis. To show a greater
inhibitory activity of this dipeptide in patients
before dialysis, we prepared methanol-extracted
plasma fractions (that contained AcSDKP and
Fig 2. Correlation between (A) plasma AcSDKP con-
centration and creatinine clearance, (B) creatinine
clearance and Hb concentration in 35 nD patients not
administered an ACE inhibitor (ACEI; ●; solid line) and
21 other nD patients administered an ACE inhibitor (ᮀ;
dashed line), and (C) plasma AcSDKP and Hb concen-
trations in the 35 nD patients not administered an ACE
inhibitor. Creatinine clearance was calculated using
Cockroft’s formula. AcSDKP concentrations were mea-
sured by enzyme immunoassay. Correlation was stud-
ied by linear regression analysis (*P < 0.01).
Fig 3. In vitro half-life of synthetic AcSDKP in
plasma of healthy controls (C), non–ACE inhibitor–
treated nD patients, and HD patients before or after
dialysis. Horizontal bars represent the mean of each
group. Half-life was measured as described in Patients
and Methods.
levels (6.0 Ϯ 1.9 pmol/mL) remained signifi-
cantly greater (P Ͻ 0.01) than those of controls
and nD patients.

514
LE MEUR ET AL
the dipeptide Lys-Pro) from three controls and
three patients before and after dialysis. Serial
dilutions of these fractions were added to a
control sample of plasma containing 20 pmol/mL
of synthetic AcSDKP to study their effect on the
half-life of synthetic AcSDKP (see Patients and
Methods). All methanol-extracted fractions of
controls and HD patients similarly increased the
half-life of synthetic AcSDKP (data not shown).
Relationships Between Plasma AcSDKP
Levels and Weekly rHuEPO
Requirements in HD Patients
In the group of 76 HD patients selected for
their stable biological parameters for at least the
previous 3 months, the correlation, even if statis-
tically significant (r ϭ 0.39; P Ͻ 0.01) was weak
between the weekly requirement of rHuEPO and
AcSDKP concentration. If patients were classi-
fied according to treatment with an ACE inhibi-
tor, this correlation was good (r ϭ 0.69; P Ͻ
0.01) in patients administered an ACE inhibitor
(n ϭ 13), whereas it was absent (r ϭ 0.13; P ϭ
0.32) in those not administered an ACE inhibitor
(n ϭ 63; Fig 4). ACE inhibitor–treated patients
had significantly greater (P Ͻ 0.01) plasma lev-
els of AcSDKP and significantly greater (P Ͻ
0.05) weekly rHuEPO requirements than un-
treated patients (AcSDKP, 38.8 Ϯ 25 versus
10.3 Ϯ 3.6 pmol/mL; rHuEPO, 123 Ϯ 60 versus
77 Ϯ 63 IU/kg/wk); Hb concentrations were not
statistically different. When the 76 HD patients
were divided into two groups according to weekly
rHuEPO requirements (Fig 4, inset), those requir-
ing more than 100 IU/kg/wk (n ϭ 30) had
significantly greater (P Ͻ 0.01) AcSDKP levels
than those requiring less than 100 IU/kg/wk (n ϭ
46; 22.4 Ϯ 21 and 10.4 Ϯ 3.9 pmol/mL, respec-
tively). All patients (with or without ACE inhibi-
tor treatment) administered less than 100 IU/
kg/wk of rHuEPO had AcSDKP levels less than
24 pmol/mL (Fig 4, inset). This group included
only 9% of ACE inhibitor–treated patients. In the
group of patients requiring more than 100 IU/kg/
wk, 30% were treated with an ACE inhibitor.
Among them, 89% had plasma AcSDKP concen-
trations greater than 24 pmol/mL.
Fig 4. Correlations between weekly requirement of
rHuEPO and AcSDKP concentration in 76 HD patients
with stable biological parameters and stable rHuEPO
requirements not administered (●; solid line; n ؍ 63) or
administered an ACE inhibitor (ᮀ; dashed line; n ؍
13). (Inset) The 76 HD patients were classified accord-
ing to weekly rHuEPO requirement: less than 100 IU/
kg/wk (n ؍ 46) and more than 100 IU/kg/wk (n ؍ 30).
The horizontal bar represents the highest value of
AcSDKP for patients with a rHuEPO requirement less
than 100 IU/kg/wk. AcSDKP concentrations were mea-
sured by enzyme immunoassay. Correlation was stud-
ied by linear regression analysis (*P < 0.01).
their S phase entry.1 This 487-d tetrapeptide is
maintained at stable concentrations in normal
plasma.6 Because AcSDKP is physiologically
degraded by the N-terminal active site of ACE10
and partially eliminated in urine,12 its plasma
concentrations are a result of a complex balance
between its production, degradation byACE, and
renal elimination.
Confirming previous preliminary results,14 we
detected a substantial increase in plasma AcSDKP
concentrations in both nD and HD patients. In
nD patients not administered ACE inhibitors, we
found an inverse relationship with creatinine
clearance (r ϭ –0.58; P Ͻ 0.01). The absence of
correlation between these two parameters inACE
inhibitor–treated nD patients was probably caused
by individual variation in type, dose, and metab-
olism of ACE inhibitor (in addition to individual
differences in residual renal clearance of creati-
nine and AcSDKP). Because the in vitro half-life
of synthetic AcSDKP in plasma was identical in
nD patients and controls, we thus confirmed
previous personal studies12,14 establishing that its
DISCUSSION
AcSDKP inhibits the proliferation of hemato-
poietic stem cells and progenitors by blocking

RELATIONSHIP OF AcSDKP LEVELS WITH EPO TREATMENT
515
glomerular filtration has a major role in the
regulation of its plasma level. The significantly
forming unit granulocyte mononocytes) in the
blood of healthy volunteers supports this hypoth-
esis.12
greater levels of AcSDKP in HD as opposed to
nD patients were caused partly by the accumula-
tion of AcSDKP between dialysis sessions. The
longer predialysis half-life of the tetrapeptide
(which returns to normal values after dialysis)
may also be a contributing factor. This longer
half-life reflects the decreased functional activity
of plasma ACE in HD patients. Our hypothesis
that the dipeptide Lys-Pro (a degradation product
of AcSDKP) was responsible for the reduced
activity of ACE in the predialysis plasma of HD
patients was not confirmed by the experimental
method that we used (see Results).
We thus investigated in nD and HD patients
the possible relationships between high plasma
AcSDKP levels (related to CRF itself, an ACE
inhibitor, or both) and Hb concentrations or
rHuEPO requirement. In nD patients, Hb levels
were similar regardless of whether the patients
were treated with an ACE inhibitor, whereas
AcSDKP concentrations were significantly differ-
ent. Moreover, in these patients, the relationship
between Hb concentration and creatinine clear-
ance was not modified by ACE inhibitor treat-
ment. The variability of other factors, such as
PTH level, ferritin level, and residual EPO pro-
duction (an essential factor), in nD patients can
account for these results and suggests that these
factors have a major effect on erythropoiesis.
Conversely, we studied the relationship be-
tween high levels of AcSDKP and EPO require-
ment in a selected population of HD patients in
which these biological factors and rHuEPO treat-
ment were stable for more than 3 months. In this
population, we found a significant correlation
between individual rHuEPO requirement and
AcSDKP level. Furthermore, HD patients requir-
ing smaller doses of rHuEpo (Ͻ100 IU/kg/wk)
had a statistically significant twofold reduction
in AcSDKP levels than those requiring greater
doses (Ͼ100 IU/kg/wk). These results suggest
that at less than a certain threshold (ϳ24 pmol/
mL; Fig 4, inset), which is much greater than in
healthy individuals, AcSDKP does not modify
rHuEPO requirements. Above that threshold, the
significantly greater rHuEPO requirements sug-
gest an inhibitory effect of the tetrapeptide on
erythropoiesis. If verified in a larger number of
patients, these data could account for the absence
of correlation between AcSDKP and Hb levels in
nD patients and HD patients not administered an
ACE inhibitor.
Decreased EPO synthesis is the major patho-
genic factor in the anemia of patients with CRF.15
Although chronic rHuEPO therapy corrects ane-
mia in most patients by increasing the reticulocy-
tosis and life span of red blood cells,15,25 some
patients are resistant to high doses of rHuEPO.26
Hemolysis, occult bleeding and iron deficiency,15
elevated PTH levels,19,27 inflammatory states,28
and inadequate dialysis29,30 are potential causes
of this rHuEPO resistance. Interestingly, treat-
ment with an ACE inhibitor can induce anemia
and neutropenia in some hypertensive patients21,22
and is also suspected to induce rHuEPO resis-
tance in HD patients,23 although some studies are
contradictory.30-33 Finally, treatment with ACE
inhibitors or angiotensin II type 1 receptor antago-
nists has cured polycythemia occurring in some
kidney graft recipients.20,34 These clinical obser-
vations indicate that the renin-angiotensin sys-
tem has a role in the regulation of hematopoiesis;
however, the mechanisms are poorly understood.
ACE inhibitors reduce the production of EPO in
animal models and renal transplant recipients.35-37
It recently was shown that angiotensin II directly
increases in vitro proliferation of erythroid pro-
genitors.38 Because AcSDKP has an inhibitory
effect on the cycling of hematopoietic progeni-
tors,1-5 it is reasonable to believe that its accumu-
lation in plasma of patients with CRF (particu-
larly those administered ACE inhibitors) could
partially account for their deficient erythropoi-
esis. Our recent work showing that the adminis-
tration of an ACE inhibitor significantly reduced
the number of committed hematopoietic progeni-
tors (burst-forming unit–erythroid and colony-
Nevertheless, some HD patients (n ϭ 22; Fig
4, inset) treated with more than 100 IU/kg/wk of
rHuEPO had AcSDKP levels less than 24 pmol/
mL, suggesting that this tetrapeptide is one among
other factors able to induce anemia in CRF. A
threshold level of AcSDKP leading to EPO resis-
tance could explain the apparent contradictory

516
LE MEUR ET AL
ACKNOWLEDGMENT
results of some previous articles studying
rHuEPO resistance in HD patients.30-33 This hy-
pothesis is supported by our two complementary
results showing: (1) all HD patients that had
recieved more than 100 IU/kg/wk of rHuEPO
had more than 24 pmol/mL of AcSDKP; and (2)
a significant correlation of plasma AcSDKP with
rHuEPO treatment was present only in patients
with very high plasma AcSDKP concentrations,
who in most cases were those administered an
ACE inhibitor. In these previous publications,
AcSDKP levels of HD patients were not consid-
ered. The AcSDKP threshold was probably also
reached by only a few patients who were not
identified and thus not included in a specific
group for statistical analysis. It could explain the
apparent discrepancies with our present results.
Our results in HD patients indicate that
AcSDKP is a uremic toxin, similar to other
erythropoietic inhibitors described in the litera-
ture16-18; accumulation can lead to rHuEPO
resistance. Inadequate dialysis in addition to
treatment with an ACE inhibitor permanently
increases plasma and probably bone marrow
AcSDKP levels, which thus could inhibit eryth-
ropoiesis and reduce the response of erythroid
progenitors to usual doses of EPO. Mrug et al38
recently showed that angiotensin II increases
erythropoiesis in vitro. If these results apply to
in vivo situations, they suggest that treatment
with ACE inhibitors could inhibit erythropoi-
esis and reduce EPO sensitivity by two differ-
ent and complementary pathways: increasing
concentrations of AcSDKP and reducing levels
of angiotensin II. This could explain more
appropriately why angiotensin II receptor an-
tagonists exert less of an effect on erythropoi-
esis than ACE inhibitors39; they reduce the
functional activity of angiotensin II on its
target cells without increasing AcSDKP levels
that inhibit cell cycling. Direct evidence to
support these hypotheses is still lacking. This
could be obtained by first measuring in vivo
the percentage of cycling bone marrow hema-
topoietic progenitors (S phase) in animal mod-
els, and later, in nD and HD patients, with or
without ACE inhibitor or angiotensin II recep-
tor antagonist treatment and with high or low
levels of AcSDKP.
The authors thank Dr Richard Smoot for assistance in
manuscript preparation.
REFERENCES
1. Frindel E, Montpezat JP: The physiological role of the
endogenous CFU-S inhibitor acetyl-N-Ser-Asp-Lys-Pro
(Ac-SDKP). Leukemia 3:753-754, 1989
2. Guigon M, Bonnet D: Inhibitory peptides in hematopoi-
esis. Exp Hematol 23:477-481, 1995
3. Frindel E, Guigon M: Inhibition of CFU entry into
cycle by a bone marrow extract. Exp Hematol 5:74-76, 1977
4. Guigon M, Bonnet D, Lemoine F, Obari L, Parmentier
C, Mary JY, Najman A: Inhibition of human bone marrow
progenitors by the synthetic tetrapeptide AcSDKP. Exp
Hematol 18:1112-1115, 1990
5. Bonnet D, Lemoine FM, Pontvert-Delucq S, Baillou
C, Najman A, Guigon M: Direct and reversible inhibitory
effects of the tetrapeptide acetyl-N-Ser-Asp-Lys-Pro (Se-
raspenide) on the growth of human CD34ϩ subpopulation in
response to growth factors. Blood 82:3307-3314, 1993
6. Volkov L, Que´re´ P, Coudert F, Comte L, Antipov Y,
Praloran V: The tetrapeptide AcSDKP, a negative regulator
of cell cycle entry, inhibits the proliferation of human and
chicken lymphocytes. Cell Immunol 168:302-306, 1996
7. Pradelles P, Frobert Y, Creminion C, Liozon E, Masse
A, Frindel E: Negative regulator of pluripotent hemopoietic
stem cell proliferation in human white blood cells and
plasma as analyzed by enzyme immunoassay. Biochem
Biophys Res Commun 170:986-993, 1990
8. Li J, Volkov L, Comte L, Herve P, Praloran V, Char-
bord P: Production and consumption of the tetrapeptide
AcSDKP, a negative regulator of hematopoietic stem cells,
by hematopoietic microenvironmental cells. Exp Hematol
25:140-146, 1997
9. Rousseau A, Michaud A, Chauvet MT, Lenfant M,
Corvol P: The hemoregulatory peptide N-acetyl-Ser-Asp-Lys-
Pro is a natural and specific substrate of the N-terminal
active site of human angiotensin-converting enzyme. J Biol
Chem 270:3656-3661, 1995
10. Wdzieczak-Bakala J, Rousseau A, Lenfant M, Robin-
son S, Grillon C, Riches A: A new physiological role of ACE
in the regulation of hematopoiesis. Stem Cells 13:566, 1995
(abstr)
11. Azizi M, Rousseau A, Ezan E, Guyene TT, Michelet
S, Grognet JM, Lenfant M, Corvol P, Menard J: Acute
angiotensin-converting enzyme inhibition increases the
plasma level of the natural stem cell regulator N-acetyl-seryl-
aspartyl-lysyl-proline. J Clin Invest 97:839-844, 1996
12. Comte L, Lorgeot V, Volkov L, Allegraud A, Aldigier
JC, Praloran V: Effects of the angiotensin-converting en-
zyme inhibitor enalapril on blood haematopoietic progeni-
tors and acetyl-N-Ser-Asp-Lys-Pro concentrations. Eur J Clin
Invest 27:788-790, 1997
13. Azizi M, Ezan E, Nicolet L, Grognet JM, Me´nard J:
High plasma level of N-acetyl-seryl-aspartyl-lysyl-proline:
A new marker of chronic angiotensin-converting enzyme
inhibition. Hypertension 30:1015-1019, 1997
14. Le Meur Y, Aldigier JC, Praloran V: Is AcSDKP level
a reliable marker of chronic angiotensin I-converting en-

RELATIONSHIP OF AcSDKP LEVELS WITH EPO TREATMENT
517
zyme inhibition in hypertensive patients? Hypertension 31:
1201, 1998 (letter)
15. Eschbach JW: The anemia of chronic renal failure:
Pathophysiology and the effects of recombinant erythropoi-
etin. Kidney Int 35:134-148, 1989
16. Cassidy MJD, De Jager C, Ebrahim O, Camachio P,
Robson S: Peripheral blood mononuclear cells from patients
with chronic renal failure release factors which suppress
erythropoietin secretion in vitro. Nephrol Dial Transplant
9:775-779, 1994
17. Kushner D, Beckman B, Fisher JW: Do polyamines
play a role in the pathogenesis of the anemia of end-stage
renal disease? Kidney Int 36:171-174, 1989
18. McGonicle RJS, Husserl F, Wallin JD, Fisher JW:
Hemodialysis and continuous peritoneal dialysis effects on
erythropoiesis in renal failure. Kidney Int 25:430-436, 1984
19. Meytes D, Bogin E, Dukes PP, Massry SG: Effects of
parathyroid hormone on erythropoiesis. J Clin Invest 67:
1263-1269, 1981
20. Conlon PJ, Farrell J, Donohoe J, Carmody M, Walshe
JJ: The beneficial effect of enalapril on erythrocytosis after
renal transplantation. Transplantation 56:217-219, 1993
21. Griffing GT, Melby JC: Enalapril (MK-421) and
the white cell count and haematocrit. Lancet 1:1361, 1982
(letter)
22. Vlahakos DM, Canzanello VJ, Madaio MP, Madias
NE: Enalapril-associated anemia in renal transplant recipi-
ents treated for hypertension. Am J Kidney Dis 17:199-205,
1991
23. Macdougall IC: The role of ACE inhibitors and
angiotensin II receptor blockers in the response to epoetin.
Nephrol Dial Transplant 14:1836-1841, 1999
24. Rieger KJ, Seaz-Servent N, Papet MP, Wdzieczak-
Bakala J, Morgat JL, Thierry J, Voelter W, Lenfant M:
Involvement of human plasma angiotensin I-converting en-
zyme in the degradation of the haemoregulatory peptide
N-acetyl-seryl-aspartyl-lysyl-proline. Biochem J 296:373-
378, 1993
25. Polenakovic M, Sikole A: Is erythropoietin a survival
factor for red blood cells? J Am Soc Nephrol 7:1178-1182,
1996
26. Valderrabano F: Erythropoietin in chronic renal fail-
ure. Kidney Int 50:1373-1391, 1996
27. Rao DS, Shih MS, Mohini R: Effect of serum parathy-
roid hormone and bone marrow fibrosis on the response to
erythropoietin in uremia. N Engl J Med 328:171-175, 1993
28. Barany P, Divino Fihlo JC, Bergstro¨m J: High C-
reactive protein is a strong predictor of resistance to erythro-
poietin in hemodialysis patients. Am J Kidney Dis 29:565-
568, 1997
29. Ifudu O, Feldman J, Friedman EA: The intensity of
hemodialysis and the response to erythropoietin in patients
with end-stage renal disease. N Engl J Med 334:420-425,
1996
30. Walter J: Does captopril decrease the effect of human
recombinant erythropoietin in haemodialysis patients? Neph-
rol Dial Transplant 8:1428, 1993 (letter)
31. Hirakata H, Onayama K, Iseki K, Kumagai H, Fujimi
S, Omae T: Worsening of anemia induced by long-term use
of captopril in hemodialysis patients. Am J Nephrol 4:355-
360, 1984
32. Abu-Alfa AK, Cruz D, Perazella MA, Mahnensmith
RL, Simon D, Bia MJ: ACE inhibitors do not induce
recombinant human erythropoietin resistance in hemodialy-
sis patients. Am J Kidney Dis 35:1076-1082, 2000
33. Albitar S, Genin R, Fen-Chong M, Serveaux MO,
Bourgeon B: High dose enalapril impairs the response to
erythropoietin treatment in hemodialysis patients. Nephrol
Dial Transplant 13:1206-1210, 1998
34. Ducloux D, Saint-Hillier Y, Chalopin JM: Effect of
losartan on haemoglobin concentration in renal transplant
recipients—A retrospective analysis. Nephrol Dial Trans-
plant 12:2683-2686, 1997
35. Gould AB, Goodman S, DeWolf R, Onesti G, Swartz
C: Interrelation of the renin system and erythropoietin in
rats. J Lab Clin Med 96:523-534, 1980
36. Fried W, Barone-Valeras J, Barone T, Anagnostou A:
Effect of angiotensin infusion on extrarenal erythropoietin
production. J Lab Clin Med 99:520-525, 1982
37. Gossmann J, Thu¨rman P, Bachmann T, Weller S,
Kachel HG, Schoeppe W, Scheuermann EH: Mechanism of
angiotensin converting enzyme inhibitor-related anemia in
renal transplant recipients. Kidney Int 50:973-978, 1996
38. Mrug M, Stopka T, Julian BA, Prchal JF, Prchal JT:
Angiotensin II stimulates proliferation of normal early ery-
throid progenitors. J Clin Invest 100:2310-2314, 1997
39. Schiffl H, Lang SM: Angiotensin-converting enzyme
inhibitors but not angiotensin II AT 1 receptor antagonists
affect erythropoiesis in patients with anemia of end-stage
renal disease. Nephron 81:106-108, 1999