British Journal of Haematology. 1988, 70, 137-141

Inhibition of bone marrow myelopoiesis and erythropoiesis in vitro by anti-retroviral nucleoside derivatives

MARGARET JOHNSON, TERESA CAIAZZO, JEAN-MICHEL M O L I N A , ROBERT DONAHUE* A N D JEROME GROOPMAN Division of HematologylOncology, Department of Medicine, New England Deaconess Hospital, Harvard Medical School, Boston, Mass., and *Genetics Institute, Inc., 87 Cambridge Park Drive, Cambridge, Mass. 02140, U.S.A.

Received 9 November 1987; accepted for publication 9 May 1 9 8 8

_________

Summary. We have studied the in vitro effects of nucleoside deoxyadenosine (DDA) was less inhibitory in this culture inhibitors of reverse transcriptase on bone marrow myeloid system. Studies of candidate antiretroviral agents in bone and erythroid progenitor development. Both 3’-azido-3’- marrow culture may predict clinical toxicity as well as lead to deoxythymidine (AZT) and 2’,3’-dideoxycytidine (DDC) strategies for combination therapy which minimize impair- potently inhibited in vitro haematopoiesis, while 2’,3’-di- ment of haematopoiesis.

Several nucleoside derivatives have recently been identified as having potent in vitro inhibitory effects on the human immunodeficiency virus (HIV), the retrovirus that causes the acquired immunodeficiency syndrome (AIDS) and related conditions (ARC) (Mitsuya & Broder, 1987; Mitsuya et al. 1987: ‘farchoan & Broder, 1987). In vitro, 3’-azido-3’- deoxythymidine (AZT). 2’,3’-dideoxycytidine (DDC) and 2‘,3’-dicleoxyadenosine (DDA) inhibit the activity of the viral reverse transcriptase and lead to chain termination during the replication of the retrovirus. AZT, in initial studies and in a subsequent randomized double-blind placebo-controlled clinical study, has been shown to have clinical benefit with reduction in morbidity and mortality in patients with AIDS andARC(Fischletal,1987; YarchoanetaL 1986).Themajor dose limiting toxicity of AZT to date has been haematological, with development of significant anaemia and leucopenia (Richman et al. 198 7). DDC has just entered clinical trials and its toxicity profile is not yet fully defined, but leucopenia and thrombocytopenia have been noted. Because such nucleoside derivatives are likely to have a role in the therapy of HIV- related {disease, we studied the effects of AZT or DDC, and more recently DDA. on in vitro myelopoiesis and erythropoie- sis. Such studies may be useful in predicting haematologic toxicity in vivo and designing treatment regimens with antiviral drugs of non-overlapping side effects.

Correspondence: Dr Jerome E. Groopman, Division of Hematology/ Oncology. New England Deaconess Hospital. 1 10 Francis Street, Boston, MA 02215, U.S.A.

MATERIALS AND METHODS

Healthy subjects known to be HIV seronegative underwent bone marrow aspiration using standard techniques. Simi- larly, HIV seropositive individuals with AIDS or ARC who were not receiving myelosuppressive therapy and did not have infectious or neoplastic disorders involving the bone marrow were selected for bone marrow aspiration. The study was approved by the Institutional Review Board of the New England Deaconess Hospital and informed consent was obtained from all participants. Briefly, 1-3 ml of bone marrow was aspirated into preservative-free heparin (Sigma, St Louis, Mo.) and the light density non-adherent population obtained as previously described (Donahue et al, 1987). Nonadherent cells were cultured at a concentration of 5 x lo4 cells/ml in a semisolid matrix employing Iscove’s Modified Dulbecco’s Medium (IMDM) with 30% fetal calf serum (MA Bioproducts. Walkersville, Md.). bovine serum albumin (Sigma fraction V, St Louis, Mo.), beta-mercapto- ethanol, and methylcellulose (Dow Chemical, Cincinatti, Ohio). Recombinant human granulocyte-macrophage col- ony-stimulating factor (rGM-CSF) and recombinant erythro- poietin (rEP0) were generous gifts of Dr Steven Clark of the Genetics Institute (Cambridge, Mass.) and were added to culture at optimal concentrations to support granulocyte- macrophage colony growth (CFU-GM) and early erythroid progenitor growth (BFU-E) as previously described (Donahue et al, 1987). The rGM-CSF expressed by CHO cells was purified as described (Donahue et al, 198 7) and diluted to give a protein concentration of 50 pg ml-l. This was then diluted 1 : lo3 with IMDM to give a final concentration of 1.8 nM. Two units of highly purified recombinant erythropoietin

137

138 Margaret Johnson et a1

- 10 50 I00

-"I NORMAL BONE MARROW WITHOUT EPO 1 El AZT

DDC

h I T T I

NORMAL BONE MARROW WITH €PO

AZT DDC

01 10 50 100

0 l.!L

01 *

10

Gz1 AZT ODC

- - 50 100

LAZT3 or CDDCI, p M

Fig 1. Effects of AZT or DDC on normal bone marrow myeloid and erythroid progenitors. CFU-GM and BFU-E were cultured in methylcellulose in the presence of 1.8 nM rGM-CSF with or without 2 U rEPO and scored on days 12-14. (a) CFU-GM grown without rEPO: (b) CFU-GM with rEPO: (c) BFU-E. Each bar represents the mean colony count of duplicate plates, plus and minus the standard deviation.

were added dropwise to the appropriate cultures on day 3 after the initiation of incubation. CFU-GM and BFU-E were counted on days 12-14. AZT and DDA were generous gifts of Drs S. Broder and H. Mitsuya, National Cancer Institute (Bethesda, Md), and DDC, a generous gift of Drs S. Broder and

H. Mitsuya. National Cancer Institute (Bethesda, Md), and Dr W. Soo, Hoffman-La Roche (Nutley, N.J.). The lyophilized AZT, DDC or DDA were reconstituted with IMDM to make a 1 mM stock, and stored at - 2OoC until needed for culture. On day of addition to culture, the stock solutions were thawed

Anti-retroviral Nucleosides and Haemopoiesis 139

ARC BONE MARROW w i rHour EPO '"r : iE 401 I I

Y 5 -

1OC

OL 01 10 50 L 10 0

50LzGF MARROW

IL

01 10

5 0 ~ R ~ ~ ~ w im EPO

n AZT W DDC

13 AZT DDC

IL-

5 0 100

[AZTJ or fDDC/, ,uM

Fig 2. EB'ects of AZT or DDC on myeloid and erythroid progenitor development from an HIV infected donor with AIDS-related complex (ARC). Methods and data expressed as in Fig 1 . (a) CFU-GM grown without rEPO: (b) CFU-GM with rEP0 (c) BFU-E.

140 Margaret Johnson et a2 Table I. Inhibition of rGM-CSF dependent myloid and erythroid colony formation by AZT or DDC

3’-Azido-3’-deoxythymidine (AZT)

CFU-GM CFU-GM Concentration (PM) with EPO without EPO BFU-E

2’,3’-Dideoxycytidine (DDC)

CFU-GM CFU-GM with EPO without EPO BFU-E

0.1 40%f 2 1%

1 .o 62%f 15%

5.0 75%f 15%

10 76%f22%

(n= 10)

( n = l l )

( n = l l )

( n = 1 1 )

42%f25% (n= 12)

56%f20% ( n = l l )

78%f 15% ( n = l l )

79%f17% (n=12)

80x5 15% ( n = 8 )

99%f2% ( n = 8 )

99%f 3% ( n = 8 )

l O O % f O % (n = 8 )

44%f21%

83%f 13% (n=9)

96%f 5% (n=9)

l O O % f O % (n=9)

(n=9) 40% 12%

92%f9% ( n = 8 )

98%f3% (n=7)

l O O % f O % (n=8)

( n = 8 ) 84%+25%

(n=6)

l O O % f O % (n=6)

l O O % f O % ( n = 7 ) l O O % f O % (n=7)

Results represent the mean per cent inhibition (fstandard deviation) for n bone marrow donors.

Table 11. Mean per cent inhibition (fstandard deviation)

DDA 2‘3‘-dideoxyadenosine concentration ( p ~ )

0.01 0.05 0.1 1.0 5.0

10.0 20.0 30.0 40.0 50.0 75.0

100.0

CFU-GM with EPO

CFU-GM without EPO

2 2 f 2 7 1 6 f 2 9 2 1 f 1 8 3 4 f 3 0

9 f 9 20 f 20 3 5 f 1 5 25 f12 32f22 45 f 18 4 9 f 3 1 70 f21

1 7 f 1 6 1 5 f 2 1 3 9 f 1 0

9 f 1 5 2 3 f 2 0 2 7 f 1 0 4 6 f 7 3 6 f 2 5 4 7 f 1 4 4 7 f 1 4 5 3 f 7 70 f17

BFU-E

21 f 31 1 2 f 2 2 27 f14 44 f42 32 f 38 25 f 32 63 + 36 38f28 4 2 f 4 1 58 f24 6 1 f 3 0 9 0 f 8

Results represent the mean per cent inhibition (fstandard deviation) for n = 4 bone marrow donors.

and diluted with IMDM so that the final concentration in culture ranged from 0.1 to 10 PM. Control cultures were treated with IMDM alone.

RESULTS

AZT, DDC and DDA were studied for their effects on erythroid and myeloid bone marrow progenitor development over a concentration range which had previously been determined to have anti-HIV activity in vitro (Mitsuya & Broder, 1987: Mitsuya et al, 1987; Yarchoan & Broder, 1987). Representa- tive experiments are shown in Figs 1 and 2. AZT or DDC suppressed BFU-E and CFU-GM development even at mini- mum concentrations inhibitory to HIV. 1 PM or 0.1 PM respectively. Progenitors derived from HIV-infected indi- viduals with AIDS or ARC appeared to be equally sensitive to the suppressive effects of these nucleoside derivatives com-

pared to progenitors derived from healthy HIV-negative subjects (Fig 2) . Table I presents the mean percent inhibition by AZT or DDC of CFU-GM and BFU-E derived from HIV infected or uninfected bone marrow donors. Interestingly, DDA demonstrated much less inhibition of CFU-GM (Table 11) at concentrations of 0.01-10 PM with 10 PM predicted as the optimal anti-HIV level in vitro (Yarchoan et al. 1986). At concentrations of 50, 75 and 100 PM DDA there was 45- 70% inhibition of myeloid progenitor development.

DISCUSSION

The nucleoside derivatives AZT and DDC have entered clinical use for the therapy of HIV-related disorders. A placebo-controlled multicentre study of AZT demonstrated reduction of subsequent opportunistic infections and prolon- gation of life in selected patients with AIDS or ARC (Fischl et al. 1987). DDC has more recently entered clinical trials, and limited data are available regarding its toxicity profile and clinical effects. Both leucopenia and thrombocytopenia have been noted in some patients treated at doses below which plasma levels could be measured (S. Broder. personal commu- nication). DDA has not yet been studied in man. Bone marrow suppression has been the dose-limiting side effect of AZT (Richman et al. 1987). Discontinuation of therapy with AZT is often necessary due to leucopenia and/or persistent anaemia requiring multiple transfusions. In vitro studies of AZT, DDC and DDA to date have focused on the antiviral and cytotoxic effects of these nucleoside derivatives in lymphoid and monocytoid cell lines susceptible to HIV (Mitsuya & Broder, 1987: Mitsuya et al, 1987: Yarchoan & Broder, 1987). These studies demonstrated a wide therapeutic index, with minimal or no cellular toxicity observed at 10-fold or greater the optimal antiviral concentrations. Our studies clearly demonstrate significant inhibition of bone marrow myeloid and erythroid progenitor development by AZT or DDC at concentrations which are minimal for antiviral activity in vitro. This was seen in bone marrow cultures derived from either HIV infected or uninfected donors. DDA did not manifest such profound suppression of haematopoie-

Anti-retroviral Nucleosides and Haemopoiesis 14 1 controlled trial. New England Journal of Medicine, 317, (4). 185- 191.

Mitsuya, H. & Broder, S. (1987) Strategies for antiviral therapy in AIDS. Nature. 325, 773-778.

Mitsuya. H.. Jarrett, R.F., Matsukara. M., Dimarzo-Veronese. F., Devico, A.L., Sarngadharan, M.G., Johns, D.G., Reitz, M.S. & Broder, S. (1987) Long-term inhibition of human T-lymphotropic virus type III/lymphadenopathy- associated virus (human immu- nodeficiency virus) DNA synthesis and RNA expression in T cells protected by 2’,3’dideoxynucleosides in vitro. Proceedings of the National Academy of Sciences of the United States of America, 84, 2033-2037.

Richman. D.D., Fischl. M.A.. Grieco, M.H.. Gottlieb, M.S.. Volberding, P.. Laskin, O.L.. Leedom, J.M., Groopman, J.E., Mildvan, D., Hirsch, M.S., Jackson, G.G.. Durack, D.T.. Nusinoff-Lehrman, S., AZT Collaborative Working Group (1 987) The toxicity of 3’-acido-3’- deoxythymidine (Azidothymidine) in the treatment of patients with AIDS and AIDS-related complex: a double-blind, placebo- controlled trial. New England Journal of Medicine, 317, (4). 192- 197.

Yarchoan, R. & Broder. S. (1987) Development of antiretroviral therapy for the acquired immunodeficiency syndrome and related disorders: a progress report. New England Journal of Medicine, 316, 55 7-564.

Yarchoan. R.. Weinhold. K.J.. Lyerly. H.K.. Gelmann, E.. Blum. R.M., Shearer, G.M.. Mitsuya, H.. Collins, J.M.. Myers, C.E., Klecker. R.W., Markham. P.D., Durack, D.T., Lehrman. S.N.. Barry, D.W., Fischl, M.A., Gallo. R.C., Bolognesi, D.P. & Broder, S. (1986) Administration of 3’-azido-3’-deoxythymidine, an inhibitor of HTLV-III/LAV replication, to patients with AIDS or AIDS-related complex. Lancet. i, 575-580.

sis in vitro at its antiviral concentration of approximately 10 PM comlpared to AZT at 1 PM or DDC at 0.1 PM. Our study therefore suggests that the toxicity of AZT or DDC may not overlap with DDA. and that simultaneous or sequential administration of these nucleoside derivatives should be exploredl. Candidate anti-HIV agents should be regularly studied for their effects on bone marrow progenitors in order to anticipate potential toxicity and develop regimens that minimize impairment of haematopoiesis.

ACKNOWLEDGMENT This research was supported in part by grants from NM, HL33774, AI/HL24475, the New England Deaconess Hospi- tal Cancer and Immunodeficiency Research Fund, the Fonda- tion Merieux and the Fondation pour la Recherche Medicale.

REFERENCES Donahue, R.E., Johnson, M.M.,Zon, L.I.. Clark, S.C. &Groopman. J.E.

(1 98 7 t Suppression of in vitro haematopoiesis following human immunodeficiency virus infection. Nature, 326, 200-203.

Fischl, M.A.. Richman. D.D., Greico, M.H., Gottlieb. M.S., Volberding. P.A., Laskin. O.L., Leedom, J.M., Groopman, J.E., Mildvan, D.. Schooley. R.T.. Jackson, G.G., Durack, D.T., King, D., AZT Collab(orative Working Group (1987) The efficacy of 3’-azido- 3’deoxythymidine (azidothymidine) in the treatment of patients with AIDS and AIDS-related complex: a double-blind. placebo-