EFFECTS OF ERYTHROPOIETIN, ANGIOTENSIN II, AND ANGIOTENSIN-CONVERTING ENZYME INHIBITOR ON ERYTHROID PRECURSORS IN PATIENTS WITH POSTTRANSPLANTATION ERYTHROCYTOSIS

Abstract

Background. 

Angiotensin-converting enzyme inhibitors (ACEI) have become the treatment of choice for posttransplantation erythrocytosis (PTE). Yet the pathogenesis of PTE and the mechanisms of action of ACEI remain unclear. Therefore, we studied the dose response to erythropoietin (Ep), angiotensin II (AII), and the ACEI enalaprilat on the in vitro proliferation of erythroid progenitors in patients with PTE and in controls. We also evaluated ACE polymorphism in the two groups.

Methods. 

Twelve patients with PTE and 12 renal transplant patients without PTE were studied. Erythroid burst-forming units (BFU-E) were isolated from peripheral blood using standard methods. Ep sensitivity was determined for four patients with PTE and three control patients, using 0-3 U/ml Ep. AII dose response was studied in four patients with PTE and five control patients, using AII concentrations of 0-1000 nM. The effect of enalaprilat was studied in eight patients with PTE and eight control patients, using drug concentrations of 0-10 ng/ml. ACE gene insertion/deletion polymorphism was determined by polymerase chain reaction.

Results. 

PTE patients showed a significant shift of the Ep response curve to the left compared with controls, with 50% maximal growth occurring at a lower Ep concentration (0.3 U/ml vs. 0.95 U/ml, P<0.025.) However, there was no difference in the number of BFU-E colonies between PTE patients and controls. AII added to the growth medium produced only minor stimulation in both groups. PTE patients showed significant inhibition of BFU-E growth with 10 ng/ml enalaprilat, but controls showed no inhibition of BFU-E growth with ACEI. There was no difference in ACE polymorphism between PTE and controls.

Conclusions. 

Our data suggest that PTE is associated with increased erythroid progenitor sensitivity to Ep. The effect of ACEI to decrease hematocrit in patients with PTE may be due to inhibition of red cell precursor growth.

Posttransplantation erythrocytosis (PTE^*), first described in 1965 ⁽¹⁾, is a phenomenon that is limited to renal transplant recipients and occurs in 10 to 20% of patients ^(2-4). PTE is commonly defined as persistent elevation of hematocrit ≥51% ^{(2, 4, 5)}. A more rigorous definition requires an absolute increase in red cell mass ⁽³⁾. Other causes of erythrocytosis are excluded by clinical tests. A number of factors have been associated with PTE, including the use of cyclosporine and elevated serum erythropoietin (Ep) levels ^(2-4). The pathogenesis of PTE, however, remains obscure. Several investigators have suggested a variety of possible mechanisms, including abnormal Ep production ^(6-9), abnormal erythroid precursor sensitivity to Ep ^{(10-12, 16)}, abnormal marrow production of angiotensin II (AII), or increased erythroid precursor sensitivity to AII ^{(5, 17)}. The use of angiotensin-converting enzyme inhibitors (ACEI) or a selective AII type 1 (AT₁) receptor antagonist can significantly lower hematocrit by decreasing red cell production in patients with PTE ^{(4, 5)}, further supporting a role for AII. Other growth factors, such as insulin-like growth factor-1 ⁽¹⁸⁾, may also be involved in the pathogenesis of PTE. To further examine the potential mechanisms of PTE and the effect of ACEI, we studied the dose response to Ep, AII, and the ACEI enalaprilat on the in vitro colony formation of erythroid progenitors in patients with PTE and in renal transplant recipients without erythrocytosis.

MATERIALS AND METHODS

Patients. We screened all patients followed in the renal transplant clinic at Montefiore Medical Center and identified 306 patients who had at least four hematocrit measurements and a stable serum creatinine level ≤3 mg/dl over a 12-month period. Forty-nine (16%) patients met criteria for the diagnosis of PTE with a hematocrit ≥51% on three consecutive measurements and an increased red cell mass determined by ⁵¹chromium-labeled red cells. Plasma volume was measured with ¹²⁵I-labeled albumin. Other causes for erythrocytosis, such as volume depletion, malignancy, polycythemia vera, pulmonary disease, or hepatic dysfunction, were excluded by clinical or laboratory testing. At the time of the study, 33 patients with PTE had achieved a normal hematocrit on ACEI therapy. Ten patients refused or were unavailable for the study. The remaining 23 patients agreed to discontinue ACEI and return for monthly monitoring. In 10 patients, PTE recurred 2 to 3 months after stopping ACEI (seven on enalapril, three on lisinopril). These 10 patients were enrolled in the study, in addition to two other patients with newly diagnosed and untreated PTE. Twelve transplant patients with stable hematocrits who did not have PTE and were not on ACEI were selected as controls, with an attempt made to match these patients for age, gender, and serum creatinine levels. The study protocol was approved by the Institutional Review Board of the hospital, and all patients gave written informed consent before entry into the study.

Culture studies with erythroid burst-forming units (BFU-E). Circulating erythroid progenitor cells were isolated using previously described methods ^{(19, 20)}. Sixty milliliters of heparinized peripheral venous blood was obtained from each patient. Mononuclear cell suspensions were prepared by 30 min of centrifugation at 400×g in a Ficoll-Hypaque gradient (density, 1.077 g/ml; Sigma Chemical Company, St. Louis, MO). The cells from the interface were washed twice with alpha medium supplemented with 2% heat-inactivated fetal calf serum (FCS), enumerated, and then diluted with alpha medium to create a stock solution with a final concentration of 5×10⁶ cells/ml. The T cell-depleted mononuclear fraction was prepared by adding 0.1 ml of 2-aminoethylisothio-uronium bromide (AET)-treated sheep red blood cells to each 1.0 ml of stock solution, which was layered over an equal volume of Ficoll-Hypaque. After 30 min of centrifugation, the cells from the interface were washed twice with alpha medium supplemented with 2% FCS, enumerated, and resuspended in 0.9 ml FCS and 0.1 ml dimethyl sulfoxide before cryopreservation at −135°C. At the time of the study, the T cell-depleted mononuclear fraction was resuspended and plated in multiwell tissue culture plates (0.5 ml/well; Thomas Scientific, Swedensboro, NJ) with a final concentration of 2×10⁵ cells/ml in enriched Iscove's methylcellulose (Stem Cell Technologies, Inc., Vancouver, British Columbia). All cultures were prepared in quadruplicate. Each 90 ml of culture medium contained 30 ml of FCS, 10 ml L-glutamine, and the following cytokines (R&D Systems Inc., Minneapolis, MN): 2000 ng of interleukin 3, 5000 ng of stem cell factor, 2000 ng of granulocyte-colony-stimulating factor, 2000 ng of granulocyte/macrophage-colony-stimulating factor, and a concentration of Ep (Amgen, Thousand Oaks, CA) that varied from 0 to 3 U/ml. The culture plates were incubated at 37°C in 5% CO₂ at 100% humidity. Colonies were counted on day 20 using a dissecting or inverted microscope.

Ep sensitivity was defined by the concentration of Ep required to stimulate 50% maximal growth of BFU-E. Ep sensitivity was determined for seven patients (four PTE, three controls) by measuring BFU-E growth in a dose-dependent fashion using Ep in concentrations ranging from 0 to 3 U/ml. The AII dose response was determined for nine patients (four PTE, five controls) using AII (Sigma) concentrations ranging from 0 to 1000 nM. The effect of the ACEI enalaprilat (Merck, West Point, PA) was studied in 16 patients (eight PTE, nine controls) using drug concentrations ranging from 0 to 10 ng/ml. Both AII and enalaprilat studies were performed using a saturating Ep concentration of 3 U/ml. The number of circulating BFU-Es per ml of peripheral blood was determined for 16 patients (eight PTE, eight controls) and was calculated as follows: (Equation) where # colonies represents the number of BFU-E colonies counted on day 20, # cells plated equals the final number of T-depleted mononuclear cells that were plated, # Ficoll cells equals the number of cells recovered after Ficoll separation, # T-depleted cells equals the number of cells recovered after T-cell depletion, WBC equals the total white blood cell count on the day that blood was drawn, and blood vol equals the total volume of patient blood that was processed.

Laboratory tests. Hematocrit was measured using a Sysmex SE 9000 (Toa Medical Electronics Co., Ltd., Kobe, Japan), and serum creatinine level was measured using a Hitachi Auto Analyzer Model 747 (Hitachi Medical Corporation of America, Tarrytown, NY). Angiotensin-converting enzyme levels were measured using a Hitachi Model 911. Ep levels were determined using a chemiluminescence method with a Nichols CLS160 (Nichols Institute Diagnostic, San Juan Capistrano, CA). Specimens for angiotensin-converting enzyme complete blood count and Ep levels were collected simultaneously with blood used for BFU-E culture.

Determination of ACE genotypes. Nuclear DNA was isolated from peripheral leukocytes, and ACE gene insertion-deletion (I/D) polymorphism was determined by polymerase chain reaction using oligonucleotide primers as described by Hunley et al. ⁽²¹⁾. All samples found to be deletion-deletion (D/D) after amplification with conventional primers were reamplified with an insertion-specific primer pair to confirm the result.

Statistics. Statistical analysis was performed using SPSS (SPSS Inc., Chicago, IL). The Student t test was used to compare group means, and chi-square or Fisher's Exact Test was used to compare proportions. A P value of less than 0.05 was considered to indicate statistical significance. All data are presented as mean ± SD.

RESULTS

Table 1 shows clinical and laboratory features of PTE and control patients. There were no significant differences in age, gender, etiology of renal failure, maintenance immunosuppression, transplant duration, or serum creatinine level. Hematocrit was significantly higher in patients with PTE (P<0.001). Although Ep and serum angiotensin-converting enzyme levels tended to be lower in patients with PTE, this did not reach statistical significance. Figure 1 shows the percent maximum BFU-E growth (mean ± SD) for PTE and control patients plotted against a logarithmic scale of Ep concentration. The Ep sensitivity curve was shifted to the left for PTE patients with the 50% maximal growth point (Vmax/2) significantly less than control patients (0.3 vs. 0.95 U/ml, P<0.025). Erythroid progenitors did not grow in the absence of Ep in either PTE or control patients. There was no difference in the number of BFU-E colonies when 3 U/ml Ep was added to the culture medium (54±14.4 versus 58±9.7 colonies/ml, P=NS). Table 2 shows the effect of AII on BFU-E growth. There was only slight stimulation of growth and no difference between PTE and controls.

Table 1: Clinical and laboratory features of PTE and control patients

Figure 1: Ep dose-response curves of 20-day BFU-E of PTE and control patients. The large arrow shows 50% maximal growth (Vmax/2) for PTE patients. The smaller arrow shows Vmax/2 for controls.

Table 2: Effect of AII on BFU-E growth, expressed as percent growth from baseline

The effect of two doses of enalaprilat on BFU-E growth for 16 patients (eight PTE, eight controls) is shown in Figure 2, with the number of BFU-E colonies/ml present at day 20 in the absence of enalaprilat considered baseline, i.e., 100%. Enalaprilat produced a dose-dependent inhibitory effect on BFU-E colony formation in PTE patients but not in controls. The addition of 10 ng/ml enalaprilat significantly inhibited BFU-E colony formation to 62% of baseline in PTE patients, compared with a modest stimulation of BFU-E colony formation to 128% of baseline in control patients (P<0.025). At lower enalaprilat concentrations (2.5 and 5.0 ng/ml), there was a trend toward inhibition of BFU-E colony formation in PTE patients, but this did not reach statistical significance (P=0.07).

Figure 2: Effect of enalaprilat added to the culture medium on BFU-E growth. *P<0.03.

ACE gene (I/D) polymorphism was determined in 11 PTE patients and 9 controls. There was no significant difference in gene frequency between PTE (I=59%, D=41%) and control (I=50%, D=50%). The distribution of ACE genotype was also similar in both groups. The D/D genotype was present in 27%, I/D in 27%, and I/I in 46% of PTE patients; each genotype was present in 33% of control patients (P=NS).

DISCUSSION

Our data support the hypothesis that the pathogenesis of PTE involves increased sensitivity of erythroid progenitors to Ep. Although this mechanism has been postulated by several investigators ^{(10-12, 16)}, to the best of our knowledge this is the first study to establish the Ep sensitivity of circulating BFU-E in renal transplant recipients with and without PTE. In our patients with PTE, the Ep dose-response curve was shifted to the left compared with controls, indicating a reduction in Ep requirement to achieve 50% maximal growth. This result was not due to increased numbers of BFU-E in patients with PTE. A review of the literature revealed nine reports ^{(10-17, 22)}, four in abstract form ^{(13, 14, 16, 22)}, on the effect of Ep on erythroid precursors in patients with PTE. These studies vary considerably in patient selection, controls, criteria for PTE diagnosis, and culture conditions. Four studies used bone marrow isolates ^{(12, 13, 15, 16)} and five used circulating BFU-E ^{(10, 11, 14, 17, 22)}. BFU-E are motile cells and are found in significant numbers in peripheral blood. Although the spectrum of BFU-E in peripheral blood is probably more narrow than that of BFU-E in bone marrow, consisting mostly of early, quiescent BFU-E, both sets of BFU-E show a similar response to different growth factors ⁽²³⁾.

Heilman et al. ⁽¹⁰⁾ were the first to isolate circulating BFU-E from six patients with PTE. Their study demonstrated that Ep was required for BFU-E growth, a finding that has been confirmed by other studies ^{(14, 17)}, including the present report. This requirement may be due to the apparent ability of Ep to prevent the cellular events associated with apoptosis ⁽²⁴⁾. In several studies, however, BFU-E colony formation was observed even in the absence of exogenous Ep ^{(11, 12, 15)}. As these findings were not produced using monocyte or lymphocyte depleted isolates ⁽¹¹⁾, other cellular interactions and cytokine production may have been responsible. T lymphocytes and monocytes have been implicated in the production of factors possessing erythroid burst-promoting activity ^{(25, 26)}. Most reports observed increased BFU-E colony formation in patients with PTE compared with normal or renal transplant controls ^{(11-13, 16, 17)}, but some did not ⁽²²⁾, including the present report. Two groups interpreted their data to suggest there was increased sensitivity of BFU-E to Ep ^{(10-12, 16)}, but three groups reported no increased Ep sensitivity in their studies of cultured BFU-E ^{(13, 17, 22)}. However, these previous studies did not show data in the form of Ep response curves to support their conclusions.

Increased Ep production has been implicated in the pathogenesis of PTE ^(6-9), and the effect of ACEI have been explained on the basis of decreased Ep production ^{(18, 27-30)}. Gaston et al. ⁽⁴⁾, in a recent review of this subject, remarked that although such studies provide support for excessive Ep production in some patients with PTE and suggest that ACEI may decrease Ep production, they do not explain all the data. For example, many patients with PTE have low or even undetectable Ep levels ^(31-33). More important, ACEI often lowers hematocrit in patients with PTE without affecting Ep levels ^{(30, 33, 34)}. In our study, there was no significant difference in serum Ep levels, although there was a trend toward lower levels in PTE patients.

There is a considerable amount of experimental evidence linking the renin-angiotensin system to erythropoiesis. Both renin and angiotensin stimulate Ep production when injected into experimental animals ^(35-38). Gould et al. ⁽³⁹⁾ have previously shown that renin, renin substrate, and AII all correlated with Ep level in a hypoxic rat model. These rats produced a threefold increase in Ep production when renin was injected subcutaneously. This stimulatory effect was abolished and Ep levels became undetectable when rats were pretreated with a single oral dose of an ACEI (SQ 14225), an effect that was reversed with AII administration. In studies involving human subjects, the infusion of AII also increased Ep levels ⁽⁵⁾.

The renin-angiotensin system seems to be involved in erythropoiesis at the progenitor cell level. Mrug et al. ⁽¹⁷⁾ have shown that BFU-E from healthy human subjects express the AT₁ receptor after 6 to 9 days of growth in culture. AII, in the presence of Ep, stimulated BFU-E colony formation when added to the cell culture medium during days 6 to 9, an effect that was eliminated by the addition of 200 nM of losartan. Interestingly, it was noted (but not explained) that the stimulatory effect of AII was minimal when added to the culture medium either before day 6 or after day 9. In our study, AII was added to the culture medium in various concentrations on day 0 and produced an insignificant increase in BFU-E colony formation, confirming the latter observations of Mrug et al ⁽¹⁷⁾.

An important question that has remained unanswered is whether ACEI exert a direct effect on the growth of erythroid precursors. Rostaing et al. ⁽¹⁵⁾ observed BFU-E inhibition of colony formation when the sera from an enalapril-treated patient was added to allogeneic bone marrow cultures. Although Mrug et al. ⁽¹⁷⁾ found that the addition of 200 nM of losartan eliminated the AII-mediated stimulation of human BFU-E colony formation when added on days 6 to 9 of culture, this relatively high concentration of drug did not inhibit baseline colony formation in the presence of Ep ⁽¹⁷⁾. Julian et al. ⁽³²⁾, using an in vitro bioassay for Ep, were unable to demonstrate that enalaprilat (1 to 50 ng/ml) inhibited the incorporation of [³H]thymidine into spleen cell cultures from phenylhydrazine-treated mice in the presence and absence of Ep.

In contrast, our data strongly support the hypothesis that ACEI decrease the hematocrit in patients with PTE by inhibiting the growth of erythroid precursors. Enalaprilat, directly added to cell culture on day 0 at a concentration of 10 ng/ml, inhibited baseline BFU-E colony formation by almost 40% in PTE patients, but there was no evidence of colony inhibition in the cultures from control patients. This striking difference in BFU-E colony formation between PTE patients and controls may be due to an alteration in the cytokine milieu; however, this hypothesis was not tested. In a preliminary report, Carozzi et al. ⁽¹⁶⁾ described a decrease in regulatory cytokine levels and Ep-receptor expression on BFU-E after enalapril treatment in both PTE and control patients.

Recently, a new modulator of erythropoiesis has been described, N-acetyl-seryl-aspartyl-proline, which is an endogenous tetrapeptide that decreases proliferation of red cell precursors including BFU-E ⁽⁴⁰⁾. Angiotensin-converting enzyme inactivates this peptide, and the ACEI captopril has been shown to prevent its degradation ⁽⁴¹⁾. However, there have been no studies of this tetrapeptide in patients with PTE. We speculate that BFU-E from PTE patients may be more susceptible to inhibition of colony formation by ACEI because of decreased cell cycling time, more rapid proliferation, and up-regulation of receptors for the tetrapeptide.

A decrease in the double-deletion (D/D) ACE gene polymorphism has been described in patients with PTE ⁽⁴²⁾. Our data did not confirm this observation, perhaps because of small sample size.

In conclusion, our observations support the hypothesis that increased Ep sensitivity of erythroid progenitors may be responsible for the development of PTE. The effect of ACEI to decrease erythropoiesis in patients with PTE seems to be due to the inhibition of BFU-E growth. Further work is needed to determine whether ACEI affect local AII production or interfere with other regulators of erythropoiesis.