A comparison of the Antileukaemic Effects of Recombinant Human Tumour Necrosis Factor-α and its Muteins on Leukaemia L1210 and Leukaemia P388 in Mice

We investigated the influence of recombinant human tumour necrosis factor alpha (TNF-α) and its derivatives termed muteins III, V, VI—in which the first 3 to 7 amino acids of native TNF-α have been replaced—on the survival time of mice inoculated with leukaemia L1210 or leukaemia P338. TNF-α prolonged the survival of mice with leukaemia L1210 but did not have any therapeutic activity in leukaemia P388-bearing mice. Muteins-treated mice with leukaemia P388 lived longer than animals receiving TNF-α, while those inoculated with leukaemia L1210 did not show any significant prolongation of life compared with the TNF-α treated group. The results presented in this report indicate that the antileukaemic activity of TNF-α is governed at least in part by the nature of the N-terminal amino acids.


Introduction
The multifunctional cytokine tumour necrosis factor-{x (TNF-(x) plays a role in the regulation of many biological responses in vivo, and has been implicated in a wide range of pathological conditions, including the host response to leukaemia growth. [1][2][3][4][5] As for cytokines in general, the first event in triggering a cellular response is a specific high affinity interaction with membrane receptor molecules initiating a cascade of signal transfer reactions inside the cell. However, although two cell surface receptors for TNF-et have been identified, the amino acid residues necessary for the biological activity of TNF-0t have not been characterized. Experiments performed with derivatives of TNF-( termed muteins, in which the first 3 to 7 amino acids of native TNF-(x have been replaced, indicate that the receptor-binding domain of TNF-(x may be located near the N-terminus of the molecule. 6,7 In the present study we compare the antitumour effects of TNF-(x and its N-terminal region muteins in murine leukaemia model.

Materials and Methods
Animals: All the mice used in this experiment were produced at the Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, WrocIaw. At the time of initiation of the experiment they were young adults, [8][9][10][11][12] weeks old. They were given standard laboratory food and water ad libitum. In this study, female mice of DBA/2 strain were used.
Leukaemias: Leukaemia L1210 and P388 were obtained from the Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, Wrocl'aw, and were maintained by serial passage in the ascitic fluid of DBA/2 mice. Leukaemic cells from the fluid were resuspended in 0.9% sodium chloride such that 106 leukaemia L1210 or P388 cells were injected i.p. into DBA/2 recipients.
Cytokines: TNF-( and its muteins' syntheses and biochemical analyses were performed in the Department of Bioorganic Chemistry, L6d., Poland. Production of TNF muteins were supported by the Committee for Scientific Research Grant No. 662889203 to M.K. and B.S. Recombinant human TNF-(, had a specific activity of 2 x 107 U/mg. Muteins III, V, VI were constructed using synthetic oligonucleotides to introduce changes in the DNA sequence, encoding the 7 N-terminal amino acids of native TNF-(. The DNA was expressed in Escherichia coli, and the resulting muteins were purified by ion exchange chromatography. Amino acid sequences were analysed by automated Edman degradation using an Applied Biosystems ABI 477A protein sequencer. The N-terminal amino acids sequences of TNF-0t and its muteins are shown in Table   1. Endotoxin contamination amounted to approximately 1.9 ng endotoxin per mg protein, as estimated using a commercially available assay (Sigma Chemical Co., St Louis, MI, USA). Recombinant cytokines were reconstituted using sterile phosphate-buffered saline (PBS), and premixed at a concentration such that all doses were injected in 0.2 ml. VaI-Arg-Ser-Ser-Ser-Arg-Thr-Arg-His-Arg-His-VaI-Arg-Ser-Ser-Ile-Val-Ile-Met-Arg-Ile-Arg-Met-Met-methionine; Val-valine; Arg-arginine; Ser-serine; Thr-threonine; Lys-lysine; His-histidine; Ile-isoleucine. Treatment: Animals received tumour challenges on day 0, and all treatments were initiated i.p. on the following day. Cytokines were administered at doses of 250 and 400 t.tg/kg as daily injections for 8 days or at a dose of 400 l.tg/kg given 2, 4, 6 and 8 days after the inoculation of leukaemic cells. The control group of mice received i.p. injections of PBS.
Antileukaemic assay: Animals were observed daily for survival for a minimum of 60 days. The median survival time (MST) was assessed as follows: MST (x + y)/2, where 'x' denotes the earliest day when the number of dead animals > n/2, 'y' denotes the earliest day when the number of dead animals > (n/2)+ 1, and 'n' denotes the number of animals in that group. Therapy efficacy against leukaemia was assessed as a percentage of median survival time of the treated group (T) to that of the control group Statistical analysis: Statistical analysis was performed using Student's t-test. Results were considered significant when the p value was < 0.05.

Results
Different treatment regimes with TNF-0t or its muteins III, V, VI against the two types of murine leukaemias were studied to compare their antitumour activities. As shown in Table 2, TNF-0t prolonged survival of leukaemia L1210-bearing mice when given daily at a dose of 2501.tg/kg (0.01 > p > 0.001), but shortened it when given at a dose of 400 I.tg/kg (p < 0.001). Sequential application of TNFat a dose of 400 t.tg/kg did not significantly change the lifetime of mice with leukaemia L1210, but was better tolerated than the same dose applied in a daily treatment regime (0.01 > p > 0.001) (Table 3). Conversely, TNFmuteins did not show any therapeutic activity against leukaemia L1210 when used daily at a dose of 250 I.tg/kg, but had a comparable therapeutic effect, except mutein III to native TNF-0t, when used at a dose of 400 l.tg/kg, either in daily or sequential injections (0.01 >p > 0.001) (Tables 2 and 3). There were no significant changes in lifetime, either for daily or for sequential application, between the groups of mice receiving different TNF-{x muteins (Tables 2 and 3). T/C, MST of the treated group x 100 (%) MST of the control group Statistical significance compared with the control group. "T/C, MST of the treated group x 100 (%) MST of the control group Statistical significance compared with the control group.
The survival time of leukaemia P388-bearing mice receiving TNFor its mutein V, either 250 btg/kg in daily or 400 l.tg/kg in sequential injections, was almost the same as that of the control group (Tables  4 and 5). Daily injections of 400 btg/kg of these cytokines shortened the lifetime of mice compared with animals from the control group (p < 0.001). The lifetime of mice treated with mutein III or VI, at daily doses of 250 or 400 I.tg/kg, did not differ significantly from that observed in the control group (Table 4). Sequential application of 400 btg/kg of these muteins significantly prolonged the survival time of leukaemia P388-bearing mice in comparison with TNF-0t, mutein V or control mice (0.01 >p>0.001). The longest survival time was seen in the group receiving mutein VI in the sequential treatment schedule (Table 5). T/C, MST of the treated group x 100 (%) MST of the control group *Statistical significance compared with the control group. T/C, MST of the treated group x 100 (%) MST of the control group *Statistical significance compared with the control group.

Discussion
TNFis a cytokine which holds strong promise for application in cancer therapy because of its marked antiproliferative effects against various tumour cell lines in both in vitro and in vivo conditions. [9][10][11][12][13][14][15] Originally defined as an antitumoural agent, it is now recognized as a mediator of inflammation and cellular immune responses. The first step in the induction of these activities is the binding of TNFto specific cell surface receptors. Two such receptors, termed p55 and p75, have recently been cloned, 16,17 although the TNFdomain responsible for receptors binding has not been precisely identified. Experiments performed with antibodies generated against amino acids 1-31 of TNF-ot support the concept that the receptor-binding domain of TNF-{x may be located near the N-terminus of the TNF-ot molecule. 7,18,19 In this work we have compared the antitumour effects of TNF-ot and its derivatives termed muteins III, V, VI, in which the first 3 to 7 amino acids of native TNF-ot have been replaced, against two kinds of murine leukaemias, using two different therapy regimens.
The results of our previous 4,5 and present studies show that submaximally and maximally tolerated doses of TNF-ot are those of maximum antitumour effectiveness. Daily administration of this cytokine is more effective than sequential application, although sequential administration of TNF-ot seems to be less toxic than this observed in daily protocol. Yet, neither the dose nor the schedule mentioned above had any therapeutic activity in leukaemia P388-bearing mice. Thus, not only dosage, but also duration and sequence of TNF administration, as well as the type of target neoplasia, seem to be of key value in the treatment strategy. An important thing in understanding the mechanisms of antitumour activity of this cytokine is to explain how selectivity is achieved with respect to the diverse TNF-o responses in different cell types and tissues. A clue to understanding this mechanism may come from tudies defining the amino acid requirements for the biological activity of TNF_0.6.7. [20][21][22] From the data of others it is known that an increase in basicity of the N-terminal segment of TNF-0t improves its cytotoxic activity. 7,z This observation is consistent with our finding that the basic mutein III has more expressed antileukaemic activity than the native molecule of TNF-c. It should be stressed that this mutein binds to both TNF-0t receptors, as indicated in in vitro studies on human epithelioid carcinoma cells (HeLa cells) or Burkitt lymphoma cells (Jijoye cells). This is in contrast with observations concerning mutein VI, which does not bind to any of TNF-0t receptors, either on HeLa or Jijoye cells, and exhibits extremely cytotoxic effect in vitro, as well as in our P388 murine leukaemia model. Interestingly, this mutein exhibits very subtle proinflammatory effect and fails to induce endothelial cell responses, including expression of intercellular adhesion molecule-1 (ICAM-1) and secretion of interleukin 6 (IL-6). 6,7 In addition, experimental mice exhibit much better tolerance of high doses of mutein VI compared with native TNF-, since daily injections of 400 lag/kg of mutein VI significantly prolonged the survival time of mice compared with TNFtreated group, whose lifetime was even shorter than that observed in control animals. These results suggest that the modification of N-terminal region of TNFaugments and/or alters mechanisms of antitumour action of the native molecule of TNF-0t and allows the minimizing of the side effects of therapy with this cytokine. Mutein V, which in vitro appears to exert its effect through an interaction with p75 TNFreceptor, caused only slight prolongation of life of leukaemia L1210-bearing mice, and no effect against leukaemia P388. This is in agreement with in vitro observations, indicating that cytotoxic as well as proinflammatory effects of mutein V represent intermediate level compared with native molecule of TNF-ot on the one hand, and mutein III or VI on the other. 6,7 In the past few years, a number of experimental observations were made that have provided insights into the cytotoxic mechanism of TNF-z action. The question of structure-function relationship to TNF<z activity seems to be an important clue in the understanding of the operation of the TNF4z-signalling system. The results presented in this report and the observations of others indicate that mechanisms of TNF4z cytotoxic ,action are governed at least in part by the nature of the N-terminal amino acids. Further studies will hopefully yield a cohesive model of how TNF4z shares its biological activities, and will allow more rational transposing of the positive preclinical data into effective clinical treatment regimens.