John Leonard Cantrell

Preparation and antitumor activity of nontoxic lipid A

SummaryHighly refined, disaggregated endotoxic glycolipids (B5) from heptose-less (Re) mutant Salmonella typhimurium quantitatively converted to nontoxic (lethality for chick embryos) and nonpyrogenic (fever in rabbits) lipid A by treatment with boiling 0.1 N HCl (B5-HC1). Nontoxic B5-HCl, like toxic B5, caused regression of line-10 tumors and elimination of lymph node metastasis in 27 of 32 (84%) syngeneic strain 2 guinea pigs at a dosage of 150 μg. At this dosage, toxic B5 led to a cure in 54 of 67 (81%) tumor-bearing animals. All cured animals rejected a second line-10 tumor cell transplant. This activity depended on combining the toxic or nontoxic endotoxins with mycobacterial trehalose mycolate (P3) and an essentially nontoxic peptide-containing side-fraction (ACP) recovered during the isolation of B5. In contrast to toxic B5 or endotoxins in general, nontoxic B5-HCl did not cause endotoxic shock when combined with adjuvant dipeptide (MDP) and injected IV into guinea pigs. Chemical analysis showed that the phosphate content of nontoxic B5-HCl was about one-half that observed in toxic B5 or in toxic KDO-free lipid A, which was obtained by treating toxic B5 with sodium acetate at pH 4.5 at 100° C (B5-pH 4.5). The molar ratio of glucosamine: phosphorus: fatty acids was 2:1:4 for nontoxic B5-HCl and was 2:2:4 for toxic B5-pH 4.5. These results demonstrate that endotoxic extracts could be selectively detoxified while retaining antitumor properties. Thus, nontoxic B5-HCl may be a potential candidate for immunotherapy of human cancer.

A new immunomodulator with potential clinical applications: Monophosphoryl lipid A, a detoxified endotoxin

I m m u n o s t i m u l a t o r y Ef fec t s of E n d o t o x i n Endotoxin is an extraordinary bacterial product because of its ability to stimulate a multitude of biologic responses in animals and in humans (3). It is isolated from the outer layer of the outer membrane of Gram-negative bacteria, (e.g., Escherichia coli), which are part of the natural intestinal flora in mammals. Endotoxin is a macromolecule composed of three regions differing in their chemical and biologic properties. The O-specific polysaccharide carries the main serologic specificity of bacteria; and is linked to a core polysaccharide common to groups of Gram-negative bacteria. This core is linked through tile 2-keto-3-deoxyoctonates to a lipid component termed lipid A, which is responsible for most of the biologic activities of endotoxin. Endotoxin is known to have numerous beneficial effects in experimental animals (3), including stimulation of lymphokine production, protection against X-radiation, enhancement of nonspecific resistance to infection, and induction of antitumor effects. In its native form, however, endotoxin cannot be used clinically because of extreme toxicity in humans, which is directly attributable to the lipid A portion of the molecule. Many investigators have explored the possibility of using various chemical methods to selectively reduce the toxicity of endotoxin, while retaining the beneficial biologic activities (8). In general, these efforts have been unsuccessful. Recently, however, a simple method of preparing nontoxic lipid A from endotoxin has been de= veloped, which has led to a better understanding of the structural relationship between lipid A and its biologic properties (4-7 , 10, 12). Detoxification of lipid A is achieved by acid hydrolysis, which removes one of the molecules two phosphate groups. The development and optimization of this approach was closely tied to the introduction of improved physical-chemical means of separating and purifying the individual components of toxic and nontoxic lipid A from complex mixtures of naturally: occurring structural homologs. The structures of the single components could then be determined by a variety of modem spectroscopic techniques; in turn, this allowed the various structural features of the toxic and nontoxic forms of lipid A to be correlated with the observed biologic and chemical properties. This review summarizes the results of these studies. In Thi s I ssue

Modulation of Humoral and Cell-Mediated Immune Responses by a Structurally Established Nontoxic Lipid A

In 1954 Westphal and his associates reported on the isolation of a moiety of bacterial endotoxin which they liberated by means of hydrolysis in dilute acetic or hydrochloric acid solutions (25). The water soluble phase of the hydrolysis reaction contained a haptenic polysaccharide which no longer retained the ability to stimulate physiological responses characteristic of the starting material (3,4). On the other hand, the hydrolysis products extractable with organic solvents did retain some of the endotoxic properties of the original substance, leading Westphal and coworkers to postulate that the “endotoxic” activities of LPS were attributable to a lipidic component, which they designated lipid A (5).

Immunotherapy of EL4 lymphoma with reovirus

SummaryEL4 lymphoma was grown as an ascitic tumor in the peritoneal cavity of C57Bl/6 mice. Animals with different tumor burdens (either 107 or 109 cells) were treated with a single intraperitoneal injection of BCNU using doses from 20–40 mg/kg. Response as measured by mean survival time and percent survival was dependent on tumor burden and dose of drug. The objective of chemotherapy was to increase the mean survival time, but not the percent survival, in order to evaluate the therapeutic effect of reovirus. Mice were given 108, 109, or 1010 Pfu of reovirus at various times with respect to chemotherapy. The number of mice cured after treatment with both BCNU and reovirus was significantly greater compared to mice treated with BCNU only. Mice cured with combination therapy developed tumor-specific immunity as measured by cytotoxic lymphocytes and serum, and resistance to a lethal tumor challenge.

Enhancement of Antitumor Resistance by Mycobacterial Products and Endotoxin

It has been more than 100 years since the suggestion was made that the control of cancer was mediated by immunologic methods. This hypothesis was based on the observation that tumors either partially or totally regressed in a few patients following an acute bacterial infection. In 1911, William Coley pioneered the study of mixed bacterial vaccines or their product known as “Coley’s toxin,” for treating cancers, and there is no doubt that these vaccines had some effect in many cases (Nauts et al., 1946). The effective ingredient appeared to be endotoxin, which caused hemorrhagic necrosis of the tumors. Thus, Gratia and Linz discovered in guinea pigs (1931) and Shwartzman and Michailovsky in mice (1932) that when animals with solid tumors are given single i.v. or i.p. inoculations with small doses of endotoxin, their tumors became hemorrhagic within 24 hr. This was originally done by analogy with the local Shwartzman reaction, with the idea that some hypothetical tumor virus might have prepared the site. It appeared that this type of tumor damage was mediated indirectly because little of the injected endotoxin would be likely to make contact with tumor cells, and would not exert any direct cytotoxicity in any case (Shapiro, 1940; Brailovsky et al., 1973).