Top Banner
Hydroxyurea Affects Cell Morphology, Cation Transport, and Red Blood Cell Adhesion in Cultured Vascular Endothelial Cells By Norma C. Adragna, Peter Fonseca, and Peter K. Lauf Hydroxyurea (HU) significantly increases fetal hemoglobin (Hb) production and concomitantly affects passive erythro- cyte K transport and cell volume in patients homozygous for Hb S, thus decreasing disease severity. Red blood cells (RBCs) with Hb S display a greater adherence t o vascular endothelial cells (VECs) than do Hb A cells, thus increasing the probability of vaso-occlusive crisis. The effect of HU on the structure and function of VECs is still unknown. In the present study, HU significantly changed, in a dose-depen- dent manner, the morphology and monovalent cation com- position of cultured VECs after incubation in normal culture medium for up to 10 days in the absence and presence of 0.3 (therapeutic dose) and 3.0 (toxic dose) mmol/L HU. Treated cells showed significant morphologic changes such as an increase in apparent cell size and the formation H YDROXYUREA (HU) has been shown to increase fetal hemoglobin (Hb F) production significantly in patients with sickle cell anemia.’” Epidemiologic studies show that high levels of Hb F are associated with reduced disease ~everity~’~?’ and potentially may decrease the num- ber of vaso-occlusive c r i s e ~ . ~ . ~ Administration of HU topa- tients with severe cases of sickle cell anemia increases Hb F, F reticulocyte count, Hb F red blood cells (RBCs), mean corpuscular volume, RBC survival, oxygen affinity, and to- tal RBC cation content, with additional improvement of RBC deformability and its hydration status.’” In contrast, mean corpuscular Hb concentration, hemolysis, percent of irreversibly sickled RBCs, and, notably, KC3 cotransport are decreased by treatment with the drug.4~~ HU, an antitumor agent, inhibits DNA synthesis by de- stroying the tyrosyl free radical of the M2 subunit of the enzyme ribonucleotide reductase. This enzyme catalyzes the formation of deoxyribonucleotides from ribonucleoside diphosphate precursors and is a rate-limiting step in DNA 0.25 r .- U) g 0.20 z 00 0.10 E 0.15 c Q) C C .- B 0.05 a 2 0.00 CONT 0.3HU 3H U Fig 1. Protein content per dish of bovineVEC. line BFAE-39 as a function of HU concentration after 3 days of culture. HU was added to the culture medium during cell seeding. Results are protein contents per dish of three independent experiments (CONT = con- trol; 0.3 HU = 0.3 mmol/L HU; 3 HU = 3 mmol/L HU) with cells at passages 3 and 6. Values are expressed as X f SD (n = 3 dishes per condition). of multinucleated giant cells. The protein content per dish decreased by 50% and 80% at 0.3 and 3.0 mmd/L HU, re- spectively, accompanied by an increase in cell Na (maxi- mum, -200%) and cell K (maximum, -50%) contents at about days 4 to 6 and 8 to 10. respectively. In addition. HU decreased RBC adherencetoVECs in experiments with 6’Cr-loaded Hb A or Hb S RBCs. The HU-induced changes in VEC morphology, cation composition, and RBC adher- ence may be caused or accompanied by alterations in cell membrane permeability, transformation of endothelial cells, or decreased number/density of VEC adhesion mole- cules. Precise mechanisms of the HU effects warrant fur- ther investigation in light of the reported beneficial effects of HU in the treatment of sickle cell anemia. Q 1994 by The American Society of Hematology. synthesk8 In addition, DNA synthesis has been associated with an increaseinNainflux, Na/K pump activity, Ca efflux, and intracellular pH (pHi) in cultured fibroblasts.’ Furthermore, HU ( l mmol/L) has been used to synchronize cultured bovine aortic endothelial cells.” However, the effect of HU on DNA synthesis and cell cycle progression depends on its concentration, treatment times, and other unknown factors.’ I The mechanism of vaso-occlusion involves blood cells, vascular endothelial cells (VECs), and plasma factors. Re- cent studies have demonstrated that RBCs with Hb S adhere better to endothelial cells than do RBCs with Hb and that this effect is increased in the presence of autologous plasma’”’6 or autologous platelets.16 Adhesion of erythro- cytes and other blood cells appears to induce local hypoxia and stasis, which lead to vaso-occlusion when accompanied by vasoconstriction. In addition, hypoxia induces the pro- duction of platelet-derived growth factor (PDGF) and endo- thelin, two potent vasoconstrictors secreted by VECs.”~’* In light of the above information, the present study was designed to assess the effect of HU on VEC function, which, as in other cells (RBCsand fibroblasts), should be reflected From the Departments of Pharmacology and Toxicology, Sur- gery, and Physiology and Biophysics, Wright State University, School ofMedicine, Dayton, OH. Submitted November 25, 1992: accepted September 1. 1993. Supported in part by Wright State University (WSU) Grant No. BRSG 661 797 and the WSU Department ofSurgery. Previously presented in abstract form at the American Heart As- sociation Scientijic Conference on Functional and Structural Mech- anisms of Vascular Control, Snowbird, UT, January 29 to February l, 1992. Address reprint requests to Norma C. Adragna, PhD, Department of Pharmacology and Toxicology, Wright State University, School ofMedicine, Dayton, OH 45401-0927. The publication costs of this article were defayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C.section 1734 solely to indicate this facl. 0 1994 by The American Society of Hematology. 0006-4971/94/8302-0023$3.00/0 B/&, Vol83, No 2 (January 15). 1994: pp 553-560 553 For personal use only. on December 20, 2018. by guest www.bloodjournal.org From
9

Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

Feb 09, 2022

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

Hydroxyurea Affects Cell Morphology, Cation Transport, and Red Blood Cell Adhesion in Cultured Vascular Endothelial Cells

By Norma C. Adragna, Peter Fonseca, and Peter K. Lauf

Hydroxyurea (HU) significantly increases fetal hemoglobin (Hb) production and concomitantly affects passive erythro- cyte K transport and cell volume in patients homozygous for Hb S, thus decreasing disease severity. Red blood cells (RBCs) with Hb S display a greater adherence to vascular endothelial cells (VECs) than do Hb A cells, thus increasing the probability of vaso-occlusive crisis. The effect of HU on the structure and function of VECs is still unknown. In the present study, HU significantly changed, in a dose-depen- dent manner, the morphology and monovalent cation com- position of cultured VECs after incubation in normal culture medium for up to 10 days in the absence and presence of 0.3 (therapeutic dose) and 3.0 (toxic dose) mmol/L HU. Treated cells showed significant morphologic changes such as an increase in apparent cell size and the formation

H YDROXYUREA (HU) has been shown to increase fetal hemoglobin (Hb F) production significantly in

patients with sickle cell anemia.’” Epidemiologic studies show that high levels of Hb F are associated with reduced disease ~everity~’~?’ and potentially may decrease the num- ber of vaso-occlusive c r i s e ~ . ~ . ~ Administration of HU to pa- tients with severe cases of sickle cell anemia increases Hb F, F reticulocyte count, Hb F red blood cells (RBCs), mean corpuscular volume, RBC survival, oxygen affinity, and to- tal RBC cation content, with additional improvement of RBC deformability and its hydration status.’” In contrast, mean corpuscular Hb concentration, hemolysis, percent of irreversibly sickled RBCs, and, notably, KC3 cotransport are decreased by treatment with the drug.4~~

HU, an antitumor agent, inhibits DNA synthesis by de- stroying the tyrosyl free radical of the M2 subunit of the enzyme ribonucleotide reductase. This enzyme catalyzes the formation of deoxyribonucleotides from ribonucleoside diphosphate precursors and is a rate-limiting step in DNA

0.25 r .- U)

g 0.20 z 00 0.10 E 0.15 c Q)

C

C .- B 0.05 a 2

0.00 CONT 0.3HU 3H U

Fig 1. Protein content per dish of bovineVEC. line BFAE-39 as a function of HU concentration after 3 days of culture. HU was added to the culture medium during cell seeding. Results are protein contents per dish of three independent experiments (CONT = con- trol; 0.3 HU = 0.3 mmol/L HU; 3 HU = 3 mmol/L HU) with cells at passages 3 and 6. Values are expressed as X f SD (n = 3 dishes per condition).

of multinucleated giant cells. The protein content per dish decreased by 50% and 80% at 0.3 and 3.0 mmd/L HU, re- spectively, accompanied by an increase in cell Na (maxi- mum, -200%) and cell K (maximum, -50%) contents at about days 4 to 6 and 8 to 10. respectively. In addition. HU decreased RBC adherence toVECs in experiments with 6’Cr-loaded Hb A or Hb S RBCs. The HU-induced changes in VEC morphology, cation composition, and RBC adher- ence may be caused or accompanied by alterations in cell membrane permeability, transformation of endothelial cells, or decreased number/density of VEC adhesion mole- cules. Precise mechanisms of the HU effects warrant fur- ther investigation in light of the reported beneficial effects of HU in the treatment of sickle cell anemia. Q 1994 by The American Society of Hematology.

synthesk8 In addition, DNA synthesis has been associated with an increase in Na influx, Na/K pump activity, Ca efflux, and intracellular pH (pHi) in cultured fibroblasts.’ Furthermore, HU ( l mmol/L) has been used to synchronize cultured bovine aortic endothelial cells.” However, the effect of HU on DNA synthesis and cell cycle progression depends on its concentration, treatment times, and other unknown factors.’ I

The mechanism of vaso-occlusion involves blood cells, vascular endothelial cells (VECs), and plasma factors. Re- cent studies have demonstrated that RBCs with Hb S adhere better to endothelial cells than do RBCs with Hb and that this effect is increased in the presence of autologous plasma’”’6 or autologous platelets.16 Adhesion of erythro- cytes and other blood cells appears to induce local hypoxia and stasis, which lead to vaso-occlusion when accompanied by vasoconstriction. In addition, hypoxia induces the pro- duction of platelet-derived growth factor (PDGF) and endo- thelin, two potent vasoconstrictors secreted by VECs.”~’*

In light of the above information, the present study was designed to assess the effect of HU on VEC function, which, as in other cells (RBCs and fibroblasts), should be reflected

From the Departments of Pharmacology and Toxicology, Sur- gery, and Physiology and Biophysics, Wright State University, School of Medicine, Dayton, OH.

Submitted November 25, 1992: accepted September 1. 1993. Supported in part by Wright State University (WSU) Grant No.

BRSG 661 797 and the WSU Department ofSurgery. Previously presented in abstract form at the American Heart As-

sociation Scientijic Conference on Functional and Structural Mech- anisms of Vascular Control, Snowbird, UT, January 29 to February l , 1992.

Address reprint requests to Norma C. Adragna, PhD, Department of Pharmacology and Toxicology, Wright State University, School ofMedicine, Dayton, OH 45401-0927.

The publication costs of this article were defayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this facl.

0 1994 by The American Society of Hematology. 0006-4971/94/8302-0023$3.00/0

B/&, Vol83, No 2 (January 15). 1994: pp 553-560 553

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom

Page 2: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

554 ADRAGNA, FONSECA. AND LAUF

Fig 2. Effect of HU on VEC morphology. (A) Control cells at passage 3 after 10 days in cul- ture and in the absence of HU were approximately 90% con- fluent, with some cells un- dergoing division. Aging of the cultures was manifested by the presence of vacuoles. (B) Cells were cultured for 10 days and treated for 7 days with 0.3 mmol/L HU; cultures are at ap- proximately 60% confluence. Note the larger cell size, more binucleated cells, and the larger and higher number of vacuoles than in control cells.

in morphologic and biophysical properties, as well as in RBC adhesion.

MATERIALS AND METHODS

Materials. The endothelial cell line BFAE-39 originated from bovine thoracic aorta of normal 4- to 5-month-old male fetuses(AG 07680, NIA Aging Cell Repository, Coriell Institute for Medical Re- search, Camden, NJ). Media and chemicals used were as follows: F- 12 nutrient mixture (Ham, HF12). cell culture media supplements, trypsin, penicillin, streptomycin, amphotericin B, and EDTA

(GIBCO, Grand Island, NY); defined fetal bovine serum (Hyclone Laboratories. Logan, UT); bovine serum albumin (catalog no. A7030). tris(hydroxymethy1)aminomethane base (Tris), and HU (Sigma Chemical, St Louis. MO); and Dulbecco's phosphate- buffered saline and Hanks balanced salt solution without calcium and magnesium (Hanks medium [HMI, Whittaker, M.A. Biopro- ducts, Walkersville, MD).

Cell culiwe. Bovine aortic endothelial cells were propagated in HF12 (passages 3 through 6, cumulative population doubling [CPD] 30) supplemented with antibiotics and 10% fetal bovine serum. Cells were subcultured with trypsin-EDTA asdescribed else-

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom

Page 3: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

HYDROXYUREA EFFECTS ON VASCULAR ENDOTHELIUM 555

Fig 2. (Cont'd) (C) Cells were cultured for 10 days and treated for 7 days with 3.0 mmol/L HU; cultures are at ap- proximately 30% to 40% ron- fluence. Note the larger cell size and higher number of binucle- ated and trinucleated cells a s compared with A and B; how- ever, t h e number of vacuoles was smaller than in B. (Original magnification X 200; Nikon Di- Phot microscope; Nikon, Tokyo, Japan.)

where,I9 and passaged at a 1:2 or 1:5 ratio in 81-cm2 flasks. For cation transport measurements, cells were cultured in 35-mm di- ameter dishes, fed every other day, and used after 2 to I I days in culture. The splitting ratios were one 25-cmz flask into I5 dishes or one 81cm2 flask into 50 dishes; three dishes per point were used. For experiments on RBC-endothelial cell adhesion, cells were seeded in six-well plates of 35-mm diameter (Costar, Cambridge, MA) and used after confluence (between 5 and I2 days after seed- ing).

HU treatment. HU was added to the culture medium either during or 2 days after cell seeding, at final concentrations of 0.3 or 3.0 mmol/L to simulate therapeutic and toxic doses, respectively.

Endothelial cell cation and prolein determinalions. Na and K levels were determined by atomic absorption spectrophotometry (Perkin Elmer 5000; Perkin Elmer-Cetus, Norwalk, CT), and pro- tein content was determined by the method of Lowry et al, as de- scribed previously." RBC adhesion to VECs. The methods of Hebbel et allz and

Wautier et all3 were used with some modifications, as described be- low, to determine RBC adhesion to VECs. RBC loading with "Cr. Blood was obtained by venipuncture

into heparinized vacuum tubes from two healthy individuals (ho- mozygous for Hb A) and from two individuals with sickle cell dis- ease (homozygous for Hb S . courtesy of Dr D. Rucknagel, Compre- hensive Sickle Cell Center, University of Cincinnati, College of Medicine, Cincinnati, OH) and processed within I to 2 hours. After centrifugation for I O minutes at 3 ,000~ and 4°C. the whole-blood sample was divided into three fractions, plasma, buffy coat resus- pended in plasma, and packed RBCs. The first two fractions were maintained on ice while the RBCs were washed three times with an isotonic phosphate-buffered saline solution containing the follow- ing (mmol/L): 137 NaCI, 2.7 KCI, 0.9 CaCIz, 0.5 MgCIz, 8.1 NaZHPO4, and 1.5 KHZPO4. After the last wash, packed cells were resuspended at approximately 50% hematocrit in phosphate- buffered saline and labeled for 40 minutes at 37°C with S'Cr(sodium chromate, Amersham, Arlington Heights, IL; 50 pCi/mL packed RBCs). The labeled RBCs were washed an additional three times

with HM. divided into three aliquots, washed once more, and re- suspended to hematocrit 25% (vol/vol) in either HM containing 0.5% bovine serum albumin. plasma alone, or plasma containing the buffy coat. RBC adhesion. Because adherence of endothelial cells appears

to be independent of the species.14 we used VECs of bovine origin (cell line BFAE-39). Endothelial cells cultured in six-well plates to confluence and in the presence or absence of HU were washed twice in HM with 0.5% bovine serum albumin. Labeled RBCs (0.7 mL of 25% [vol/vol] suspension) were layered on duplicate wells and incubated for 30 minutes at 37°C. After incubation, nonadhered cells were removed by aspiration. and the cultures were washed five times with 0.7-mL aliquots of HM with albumin and once with distilled water to remove residual radioactivity (from adhered RBCs) in the plates. Radioactivity from each sample was measured in a Packard Auto-Gamma 5650 counter (Packard Instrument, Downers Grove, IL). Results were expressed as the percentage of total radioactivity that remained in the cultures after each wash, as described by Wautier et al."

Statis~ics. Results are expressed as the mean f standard devia- tion ( X 2 SD). Statistical significance was assessed by two-sample paired or unpaired I test analysis with computer software.

RESULTS

Exposure qf endothelial cells to HU before and after cell auachment. Seeding of VECs in the presence of HU de- creased the protein content per dish after 3 days of culture in a dose-dependent manner (Fig 1) with no effect on Na and K contents (data not shown). Figure I shows three in- dependent experiments, two of them on VECs at passage 6 and one at passage 3. Under these conditions, HU caused, again in a dose-dependent manner, significant changes in cell volume and extensive vacuolization associated with ini- tial cell death (data not shown). To determine whether HU interfered with cell attachment, growth, or both, the effect

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom

Page 4: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

556

of HU on cell morphology was investigated after cell attach- ment. Figure 2A through C shows the effect of HU on cell morphology when added 2 days after seeding. Cells at pas- sage 3 or 6 were incubated for 10 days in the absence (con- trol) and presence of 0.3 and 3.0 mmol/L HU, respectively. HU increased cell volume and caused appearance of binu- cleated and trinucleated cells with extensive vacuolization (both at passages 3 and 6) in a dose-dependent manner (Fig 2B and C).

Figure 3 shows a representative experiment demonstrat- ing the effect of HU, added 2 days after seeding, on the pro- tein content per dish at various days in culture and in the presence or absence of HU. The drug decreased the protein content per dish after 24 hours of treatment (day 3 in cul- ture), an effect that became dose-dependent after 48 hours (day 4 in culture), reaching a maximum after 9 days of treat- ment. Thus, the effect of HU on cell growth was not medi- ated by inhibition of cell attachment.

Under experimental conditions identical to those ofFig 3, HU increased the Na content of VECs in a dose-dependent manner between 4 and 8 days in culture and partially de- creased the peak of Na content characteristic for VECs at about 10 to I 1 days of culture19 (Fig 4A). Figure 4B shows the typical decrease of cell K content after seeding," which was prevented by HU in a dose-dependent manner after day 4 in culture, ie, 2 days after HU treatment was started. The effect of HU on total cation content (a measure of VEC vol- ume) was similar to that on K, the major osmolyte in these cells (Fig 4C).

A clear insight into the dynamics of ionic shifts can be obtained by plotting the HU-related changes as a function of time. Figure 5A through C shows the mean values of HU- dependent Na, K, and total cation changes, respectively, versus days in culture from three independent experiments. The effect of HU on Na content was transient, with a maxi- mum increase of about 70% at 0.3 mmol/L and 150% at 3.0 mmol/L HU between 4 and 7 days in culture (Fig 5A). The HU-induced increase of K (Fig 5B) and total cation contents (Fig 5C) occurred after day 4, with a maximum (up to 40% to 50% at 3.0 mmol/L HU) after day 8. Hence, HU caused statistically significant and sequential changes in Na and K contents of cultured VECs.

RBC adhesion to HU-treated VECs. Figure 6 plots the percentage of 5'Cr radioactivity associated with VECs in a wash-out experiment to determine Hb A RBC adhesion to untreated and HU-treated VECs (0.3 mmol/L HU for 7 days) for three conditions in which cells were incubated in either HM, autologous plasma, or autologous plasma con- taining the buffy coat (for further details, see legend to Fig 6). The rationale for using autologous plasma containing white blood cells (in the buffy coat) was to test for competi- tion between these and VECs for RBC adherence. No difference was observed in RBC adhesion to either treated or untreated VECs in buffered saline solution, whereas a slight and probably nonsignificant decrease was observed in the adhesion of RBCs to HU-treated VECs in autologous plasma and plasma containing white blood cells. However, in comparison to plasma alone, the presence of white blood cells in autologous plasma appeared to produce a small but

ADRAGNA, FONSECA, AND LAUF

significant difference (P < .05, n = 3 independent experi- ments) in RBC adhesion, which disappeared as the washing procedure continued.

Table l summarizes the fractional 5'Cr radioactivity re- maining in VECs during wash-out experiments to deter- mine the effect of HU on Hb A and S RBC adhesion. Mean values per wash were calculated from three experiments with Hb A and two experiments with Hb S RBCs. On aver- age, incubation in plasma left higher counts along the wash- out curve, although statistical significance was reached only for the first two washes both in control and HU-treated VECs.

Determination of the radioactivity of hemolysates (ob- tained by addition of distilled water to the dishes after the five washes were completed) is another means to estimate VEC-bound RBC "Cr. Figure 7 summarizes the hemolysate counts for the first four conditions described in Fig 6 for two subjects with Hb A and two subjects with Hb S. For the first subject with Hb A (A l), HU treatment of VECs in culture lasted 3 days [A1(3d)]. For the second Hb A subject (A2), the experiments were performed after 4 [A2(4d)] and 7 [A2(7d)] days of treatment with HU. Note that at day 4 [A2(4d)] no HU-induced decrease was observed for cells in- cubated in HM and only a small decrease ( 1 8%) was shown by those incubated in autologous plasma, whereas at day 7 [A2(7d)], HU decreased by 30% and 50% the hemolysate counts in cells incubated in HM and plasma, respectively.

In light of the above findings, VECs were treated for 7 days with 0.3 mmol/L HU to study Hb S RBC adhesion. For the first Hb S subject (SI), HU decreased the counts by 0% and 20% for cells incubated in HM and plasma, respec- tively, and for the second Hb S subject (S2), by 29% and 58%. Despite the small number of samples, the difference in hemolysate radioactivity between control and HU-treated endothelial cells incubated in plasma was significant (P < .05, n = 5) .

Thus, the HU-induced decrease in hemolysate counts when cells were incubated in either HM or autologous plasma showed time and interindividual dependence. The time dependence is clearly shown for subject A2 when com-

a' 0.00 ' I

0 2 4 6 8 10 12 Days in Culture

Fig 3. Effect of two HU concentrations on VEC protein content per dish (at passage 6) as a function of days in culture. Control (D); 0.3 mmol/L HU (A); and 3.0 mmd/L HU (0). Results are expressed as X 2 SD; errors not shown when smaller than symbols.

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom

Page 5: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

HYDROXYUREA EFFECTS ON VASCULAR ENDOTHELIUM

400 T I 3 4 A I

5 j j 200

- v)

""""""

n .- C 0 , A

a a € E .2 1: 2 2000

c Q v +-a 1500 c a, C +-a

1000

CI a, g 2500 -

-

4

, B

1000 0 2 4 6 8 10 12

Days in Culture

Fig 4. Effect of HU on cation content in bovine aortic endothelial cells after several days in culture. (A) Na; (B) K; (C) total cations. Cell passage number, symbols, and expression of results are as in Fig 3.

paring the HU-induced decrease in RBC adhesion in HU- treated VECs at days 4 and 7; the longer the treatment, the higher the HU-induced decrease. In addition, the length of treatment appeared to affect in a similar fashion the HU- induced decrease in RBC adherence and the changes ob- served in cation transport (compare Figs 4, 5, and 7). The interindividual dependence is clearly shown when compar- ing the HU-induced decrease after 7 days of VEC treatment for subjects A2, S I , and S2.

Our results also show that the HU-induced decrease was independent of the Hb type. A greater decrease occurred in cells incubated in plasma, suggesting a decreased RBC ad- hesion to HU-treated VECs. Furthermore, the total number of counts remaining after the last wash increased mainly in untreated VECs as a function of days in culture,

In three independent experiments with Hb A RBCs, the presence of white blood cells in autologous plasma de- creased the hemolysate radioactivity from 0.82% * 0.34%

557

to 0.34% & 0.16% in control VECs and from 0.56% -t 0.17% to 0.28% k 0.21% in HU-treated VECs (P < .05) suggesting competition between VECs and white blood cells for RBC adherence rather than between VECs and platelets, since au- tologous platelets from Hb S but not from Hb A increase RBC adhesion.I6

DISCUSSION

This study reports for the first time prominent effects ol therapeutic and toxic doses of HU on the morphology and steady-state ionic composition of cultured VECs. The HU- induced formation of giant, multinuclear VECs with re- duced confluency (Fig 2) and hence decreased protein content per dish (Fig 3 ) is consistent with the fact that HU inhibits DNA synthesis' and that large RBCs appear in the circulation after HU treatment of Hb S patient^.^,^ Our data rule out interference by HU solely on cell attachment (Fig

200-A 1 50

40 B A

-10; 3 I 6 9 12

Days in Culture

Fig 6. HU-dependent cation changesin bovine aortic endothelial cells as a function of days in culture. (A) Na: (B) K; (C) total cations. Cells were at passages 3 through 6. (A) 0.3 mmol/L HU; (0) 3.0 mmol/L HU. Results are expressed as X f SD; means represent the percent change in three independent experiments with tripli- cate samples each.

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom

Page 6: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

558 ADRAGNA. FONSECA. AND LAUF

0 1 2 3 4 5 6 Wash Number

Fig 6. Adhesion of 6'Cr-labeled RBCs from a Hb A donor to cul- tured bovine aortic endothelial cells (cell line BFAE-39). Results are expressed as the mean (n = 2) percentage of radioactivity remain- ing after each wash; HU concentration was 0.3 mmol/L. Cells were cultured for 10 days and treated with HU for 7 days. (1) Untreated VECs in the presence of Hanks balanced salt solution (U); (2) HU- treated VECs in Hanks (A); (3) untreated VECs in autologous plasma (0); (4) HU-treated VECs in autologous plasma (m); (5) untreated VECs in autologous plasma containing white blood cells (A); (6) HU- treated VECs in autologous plasma containing white blood cells (e).

3). The substantial changes in cellular morphology may re- flect endothelial transformation, failure of cell division (en- domitotic reduplication?), cell fusion, or a combination of these effects, and may be at the root of or causal to activa- tion/inhibition of ion transport pathways as evident from the cellular cation changes specifically due to HU treatment (Fig 5).

HU altered the Na and K composition of cultured VECs when added after cell attachment (Figs 4A through C and 5A through C), but not when added during cell seeding (data not shown). At first glance, these results appear to indicate that HU affects cation content differently depending on the time of exposure to the drug, ie, before or after cell attach- ment occurred. However, when cells were exposed to HU during seeding, the changes in both protein and cation content were studied only after 3 days in culture (Fig 1; other results not shown). At this time point, Fig 4A through C also shows no statistical difference in cation content be- tween HU-treated and untreated cells, while the effect ap- pears beyond 4 days in culture. Thus, the changes in cation content were not a consequence of HU effect(s) on cell at- tachment, but rather oftime- and concentration-dependent effect(s) on cellular ionic transport.

The increase in Na content was transient and lasted less than 72 hours (Figs 4A and SA). This phenomenon may underlie an increase in the passive permeability of VECs to Na. Since this effect preceded the changes in K content, it is possible that the increased cellular Na content resulted in (1) activation of already-existing Na/K pumps, ( 2 ) synthesis of new pump molecules, (3) a sum of these two effects, or (4) a mediated passive Na entry (through Na-K-Cl cotransport or Na/H exchange?). Such mechanisms should increase cel- lular K levels. However, if this was the only mechanism of increase in K content, the total cation content or cell vol- ume would be expected to remain the same or decrease since

the pump operates at a 3 Na:2 K stoichiometry. Thus, these results suggest that in addition to direct or indirect effects on the Na/K pump, HU affects K content either by inhibiting other pathway(s) for K exit or by stimulating alternative routes for K influx. The first possibility seems more likely in light of the effects of HU on K-Cl cotransport and cell vol- ume in RBCS.~ Although at this point K-Cl cotransport has not been shown in VECs, Fig 4B and unpublished evidence from this laboratory points in this direction, since HU in- hibited in a dose-dependent manner and as a function of time in culture the normal decrease in K content reported earlier." This effect of HU on K content was accompanied by an increase of cation content and thus of cell volume (Fig 4C and K ) . which resembles the effects of the drug in RBCs of patients under treatment.''"3

HU significantly decreased RBC adherence to VECs dur- ing analysis of the radioactivity remaining in the hemoly- sates after the fifth wash (Fig 7). Furthermore, in 51Cr wash- out experiments (Fig 6 and Table l ) , adhesion of RBCs to VECs was higher in the presence of plasma than in HM both with or without HU treatment. However, statistical signifi- cance was obtained only for the two initial washes, whereas the HU effect was statistically nonsignificant. This result may be accounted for by several factors. First, the high num- ber of counts in the wash-out experiments produced a larger error than the one inherent to the small number of counts extracted in the hemolysate, with the consequent effect on statistical significance. Second, the effect of HU was time- dependent; however, selection for computation of the ex- periments performed after 7 days of treatment with HU did not improve the statistical significance in the wash-out curves. Third, the effect of HU on RBC adherence showed interindividual variability, suggesting that RBC-VEC adhe- sion is a specific process.

We did not confirm previous findings of increased Hb S RBC adherence to V E C S , ' ~ " ~ which may be due to the small sample size and interindividual variability.16 The use of bo- vine instead of human serum albumin during the incuba- tions with HM could explain the differences observed be- tween this medium and the conditions using human

Table 1. Effect of HU Treatment on "Cr-Labeled RBC Adherence to VECs

Medlurn Wash No. Control (%) HU (%)

HM 1 2 3 4 5

Plasma 1 2 3 4 5

8.5 f 2.5 4.9 f 1.8 3.4 t 1.8 2.6 ? 1.9 2.2 f 1.8

12.8 f 3.3' 8.7 f 2.9t 6.3 i- 2.4 5.2 k 2.3 4.5 f 2.2

9.1 f 2.6 4.7 f 1.7 3.3 2 1.9 2.7 f 1.9 2.3 f 1.9

14.6 k 4.2t 9.5 k 2.4t 7.3 t 2.6 6.4 ? 2.9 5.8 f 3.2

Results are presented as fractional 5'Cr radioactivity remaining in VECs after each wash (n = 5, x f SD).

* P c .01 with respect to HM. t P c .05 with respect to HM.

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom

Page 7: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

HYDROXYUREA EFFECTS ON VASCULAR ENDOTHELIUM 559

tance not only for sickle cell anemia, but also for chemo- therapy, since cell enlargement seems to be a generalized phenomenon induced by HU and also by other drugs. HU is not only a cytostatic drug, but also a carcinogen. Knowing the mechanism of toxicity of the drug at both cellular and molecular levels could provide relevant information for the treatment of sickle cell anemia and of cancer as well.

The decrease in RBC adhesion could result either from cell enlargement by decreasing the relative density of adhe- sion molecules anchoring the RBC, from a drug-induced concomitant decrease in the absolute number of adhesion molecules, or from HU effects on their normal function. A combination of all these factors is also possible. These fac- tors may appear as a consequence of ( I ) changes in mem- brane potential affecting the assembly of the adhesion com- plexes that insert in the membrane, (2) altered surface potential, local pH, ionic microenvironment, or cell swell- ing, or (3) any other change affecting production of second messengers. All of these phenomena can be reflected in an altered ion transport activity of the cell. Understanding the mechanism of decreased adhesion can provide significant information on the process of RBC-endothelial cell adhe- sion that is at the root of the vaso-occlusive process, which is of high significance in sickle cell anemia.

In conclusion, our results show that HU induced signifi- cant morphologic and membrane transport changes in cul- tured VECs. In addition, this study provokes interesting questions that could stimulate further studies to determine the effect of HU on VEC function in vivo.

c 1.20

8 0.90 .- > c

0 U .-

2 0.60 2 0.30

0.00 v)

Al(3d) A2(4d) A2(7d) S1 (7d) S2(7d)

Fig 7. Adhesion ol "Cr-labeled RBCs from two Hb A (Al, A 2 1 and two Hb S (Sl, 82) donors to cultured bovine aortic endothelial cdls after incubation in media of different composition. Results are expressed as the mean (n = 2) percentage radioactivity extracted after hemolyzing the cells in distilled water (see the Methods). Cells were cultured from 5 to 10 days and treated with 0.3 moi/L HU from 3 to 7 days, as indicated in parentheses in the figure. Untreated endothelial cells incubated in HM (m); HU-treated cells in HM I control celb incubated in autologous plasma (a); HU-treated cells in plasma (HI.

plasma. However, based on the finding that bovine and hu- man VECs behave similarly for RBC adhesion,14 it is to be expected that the differences observed between HM and plasma are due to the presence of other plasma components, ie, Ca, growth factors, vasoactive agents, and hormones.12"* Our results instead confirm the suggestion by Narla and Ev- a n ~ ' ~ of a lack of an effect of plasma-free suspending me- dium on RBC-VEC adhesion, which was shown by Hebbel et all2 and Antonucci et all6 to be due to a calcium effect. Furthermore, our results on RBC adherence to HU-treated VECs are commensurate with in vivo studies indicating a beneficial effect of HU on the mechanism of vaso-occlusion.

Our results on cell morphology thus appear to indicate that by inhibiting DNA synthesis, HU provoked initial cell death. However, by an as-yet-unknown mechanism (trans- formation? adaptation?), the surviving cells were able to multiply and grow, although a higher percentage of binucle- ated and trinucleated cells was observed. If massive VEC death or cell injury also occurred in vivo, this would be an- ticipated to initiate clotting, which is especially disadvanta- geous for sickle cell anemia. However, the fact that the in- crease in Na content was transient and that the cells, rather than losing K, increased their cation content after 4 days of treatment (Figs 4 and 5) is an indicator of good cell health, since dying cells cannot maintain their ionic gradients. Fur- thermore, HU has been shown to heal chronic leg ulcers in vivo,2o which suggests that the potentially negative effects of the drug are a function of dose, length of treatment, and interindividual responses, and that by controlling these fac- tors its positive effects can be maximized.

Finally, the HU-induced effects on the morphologic and functional properties of VECs, as reported in this study, may be caused or accompanied by alterations in cell mem- brane permeability, transformation of endothelial cells, or decreased number/density ofVEC adhesion molecules. Elu- cidation of such mechanisms may have significant impor-

ACKNOWLEDGMENT

The authors are indebted to Kathleen Rainey for maintaining the cultures of endothelial cells, and to the Word Processing Center of Wright State University School of Medicine for typesetting the manuscript.

REFERENCES I . Dover GJ, Humphries RK, Moore JG, Ley TJ, Young NS,

Charache S, Neinhuis AW: Hydroxyurea induction of hemoglobin F production in sickle cell disease: Relationship between cytotoxic- ity and F cell production. Blood 76:735, 1986

2 . Charache S, Dover GJ, Moyer MA, Moore JW: Hydroxyurea- induced augmentation of fetal hemoglobin production in patients with sickle cell anemia. Blood 69: 109, 1987

3. Dover GJ, Charache S: Increasing fetal hemoglobin produc- tion in sickle cell disease: Results of clinical trials. Prog Clin Biol Res251:455, 1987

4. Ballas SK, Dover GJ, Charache S: Effect of hydroxyurea on the rheological properties of sickle erythrocytes in vivo. Am J Hem- atol32: 104, 1989

5. Goldberg MA, Brugnara C, Dover GJ, Schapira L, Charache S, Bunn H F Treatment of sickle cell anemia with hydroxyurea and erythropoietin. N Engl J Med 323:366, 1990

6. Noguchi CT, Rodgers GP, Schechter AN: Intracellular poly- merization. Disease severity and therapeutic prediction. Ann NY Acad Sci 565:75,1989

7. Platt OS, Thorington BD, Brambilla DJ, Milner PF, Rosse WF, Vilchinsky E, Kinney TR: Pain in sickle cell disease. Rates and risk factors. N Engl J Med 325: 1 I , 199 1

8. Hurta RA, Wright JA: Amplification of the genes for both

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom

Page 8: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

560 ADRAGNA, FONSECA. AND LAUF

components of ribonucleotide reductase in hydroxyurea resistant mammalian cells. Biochem Biophys Res Commun 167:258, 1990

9. Mendoza SA, Schneider JA, Lopez-Rivas A, Sinnett-Smith JW, Rozengurt E: Early events elicited by bombesin and structurally related peptides in quiesoent Swiss 3T3 cells. 11. Changes in Na and Ca fluxes, Na/K pump activity, and intracellular pH. J Cell Biol102:2223,1986

10. Rubin DB, Drab EA, Ward WF, Bauer KD: Cell cycle pro- gression in irradiated endothelial cell cultured from bovine aorta. Radiat Res 116:364, 1988

1 1. DAnna JA, Church VL, Tobey RA: Changes in H 1 content, nucleosome repeat lengths and DNA elongation under conditions of hydroxyurea treatment that reportedly facilitate gene amplifica- tion. Biochim Biophys Acta 868:226, 1986

12. Hebbel RP, Yamada 0, Moldow CF, Jacob HS, White JG, Eaton JW: Abnormal adherence of sickle erythrocytes to cultured vascular endothelium. J Clin Invest 65: 154, 1980

13. Wautier JL, Paton RC, Wautier MP, Corvazier E, Caen J: In- creased adhesion of erythrocytes to endothelial cells in diabetes mellitus and its relation to vascular complications. N End J Med 305:237,198 1

14. Narla M, Evans E: Adherence of sickle erythrocytes to vas-

cular endothelial cells: Requirement for both cell membrane changes and plasma factors. Blood 64:282, 1984

15. Smith BD, LaCelle P L Erythrocyte-endothelial cell adher- ence in sickle cell disorders. Blood 68: 1050, 1986

16. Antonucci R, Walker R, Herion J, Omnger E: Enhancement ofsickle erythrocyte adherence to endothelium by autologous plate- lets. Am J Hematol34:44, 1990

17. Kourembanas S, Hannan RL, Faller DF: Oxygen tension regulates the expression of the platelet-derived growth factor-B chain gene in human endothelial cells. J Clin Invest 86:670. 1990

18. Kourembanas S, Marsden PA, McQuillan LP, Faller DV: Hypoxia induces endothelin gene expression and secretion in cul- tured human endothelium. J Clin Invest 88: 1054, l99 l

19. Adragna NC: Changes in cation composition of aortic endo- thelial cells in culture. Tissue Cell 20:3 13, 1988

20. Omnger EP, Fowler VG, Teague MJ, Johnson AE, Coffey B, Phillips G, Parker JC: Sickle cell leg ulcers: A possible role for hydroxyurea therapy. 17th Annual Sickle Cell Disease Program Conference, Nashville, TN, March 14-18, 1992

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom

Page 9: Hydroxyurea Affects Cell Morphology, Cation Transport, and - Blood

1994 83: 553-560  

NC Adragna, P Fonseca and PK Lauf cell adhesion in cultured vascular endothelial cellsHydroxyurea affects cell morphology, cation transport, and red blood 

http://www.bloodjournal.org/content/83/2/553.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

  Copyright 2011 by The American Society of Hematology; all rights reserved.Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American

For personal use only.on December 20, 2018. by guest www.bloodjournal.orgFrom