Abstract
The substitution of insects for laboratory animals in toxicity testing is likely to become a reality in the framework of prescreening. Haematotoxicological studies of newly developed chemicals, such as food components, drugs, etc. performed on insects can offer advantages in, for example, environmental toxicology. Reliable routine predictions should produce an increase in our knowledge of haemocyte physiology. Although the differences between human physiology and morphology and those of insects are great, the basic functions of insect haemocytes and mammalian leukocytes appear not to have changed during evolution. The use of insects in haematotoxicity assays represents a preclinical testing strategy which will lower costs, accelerate screening and offer ethical benefits.
The haematotoxic effects of xenobiotics are very dangerous for human health and can be partly tolerated only during the use of anticancer drugs. Changes in haematopoiesis following the administration of other chemicals cannot be tolerated. 1 The preclinical and clinical development of anticancer drugs differs from that of other pharmaceuticals because of the life-threatening nature of the disease; treatment with anticancer drugs at clinically efficacious doses usually induces serious side-effects. 2
The influences of toxic compounds are evaluated at four levels: (i) in vitro tests, (ii) studies on laboratory animals, (iii) clinical (drug, food) or hygienic (environmental) studies and (iv) medical practice or epidemiological retrospective validation. 1,3–8 A shift to the next (higher) level of testing introduces an increased risk to human health. The shift from in vitro tests to clinical studies omitting tests on laboratory animals presents very high risks including the death of a volunteer or patient, and the use of tests on laboratory mammals is therefore necessary, as we still do not have sufficient knowledge of cell physiology and cell interactions to make good interpolations for the whole organism.
At present, safety tests on laboratory animals seem to be indispensable although in vitro tests are very useful assays in certain defined events. Knowledge of mammalian genetics and physiology and the use of animals with defined genetic variations improve the quality of toxicological tests. 9 The use of laboratory animals (most frequently mice, rats, dogs, rabbits, guineapigs) for safety tests is restricted in all countries of the European Community, USA, Canada and in many other countries of the world, on economic, ethical and legal grounds.
Research on vertebrate (including human) cells can be carried out on cells in cultures, in vitro or on laboratory animals, in vivo, both so-called biomodels. In vitro studies as well as mathematical and physical models are usually called ‘alternative models’. 10–12 Many in vitro and ex vivo methods usable in toxicology have been developed or are under development to reduce or replace animal usage in toxicity tests. 13,14 For example, highly effective screening for the genotoxicity of hair dyes can be achieved using three assays, namely the bacterial gene mutation assay, the mammalian cell gene mutation assay (the mouse lymphoma tk assay is preferred) and the in vitro micronucleus assay; these need to be combined with the metabolic activation systems optimized for individual chemical types. 15
The available in vitro tests significantly reduce the number of laboratory animals used within the framework of screening and make this screening cheaper, although many characteristics remain undetectable by known in vitro assays. Laboratory animal production has decreased to half since the 70s due to cell-biological assays. 16 False-positive results of safety assays in both in vitro and in vivo evaluations are tolerated as a minor problem compared with the risk of an unevaluated compound.
Many characteristics, especially in the process of drug development, can be examined using cell, tissue or organ cultivation. In haematology, no suitable in vitro test has been developed to evaluate marrow precursor cytotoxicity, 2,17,18 in contrast to marrow stem cell evaluation where the CFU-GM, as well as cell types found in the bone-marrow stroma, can now be evaluated in vitro. Many other vertebrate cell properties can also be investigated only by using laboratory animals, since our knowledge of cell physiology is still insufficient to explain their role in the whole organism.
Tests on insects
An enlarged interpretation of alternatives in toxicology testing includes the replacement of vertebrate with invertebrate species, especially in testing ‘environmental’ chemicals. 19 Using invertebrate species in environmental toxicity evaluations offers some benefit because of their position in the natural food chain.
Results published by several authors 20 provide good evidence of the applicability of using invertebrate tests as prescreening methods in toxicity testing; the use of a battery of invertebrate toxicity tests can offer a good prediction of human acute toxicity. Invertebrates seem to be useful in genotoxicity analysis using the somatic mutation and recombination test. 21 The high costs of carcinogenicity 22 and other toxicological safety evaluations of chemicals by experiments on rodents, have encouraged interest in the use of invertebrate and lower vertebrate species.
Insects and other arthropods already seem to be important models in several scientific areas such as genetics, neuroscience, etc. (cf. reviews from papers 23–26 published earlier), which also serve to increase our knowledge of comparative physiology. Most of the physiological characteristics of mammals can be examined in invertebrates, with exceptions of structures and processes that have no equivalent in the latter. Nevertheless, some essential physiological reactions have been highly conserved during evolution and have an ancient origin in vertebrate and invertebrate animals and plants; for example, no immunoglobulins have been found in invertebrates but various defence molecules are present in haemolymph plasma and haemocytes. 27,28 The toxicological significance of pathological and biochemical characteristics in invertebrates depends on physiological specificity but most metabolic processes are very similar in vertebrates and invertebrates; invertebrates may even exhibit a high sensitivity to chemicals, especially pesticides, that have no apparent effects on vertebrates. 19
It has been shown that insect haemocytes could well reflect the different effects of xenobiotics on human subjects when testing food isoflavanoids such as genistein. 29 We assume that parallel tests on several selected insect species would offer good predictions of certain haematological and immunological reactions in the human body. To produce good predictions for human subjects, we need to obtain data on comparative haemokinetics, and the sensitivity of different insect species and then to carry out experiments on different species with selected xenobiotics, similar to the studies performed in the 20th century on laboratory mammals. 1,30
Haematotoxicity tests
Comparative pathological studies, carried out mainly during the 70s and 80s, revealed the differences between the reactions of the human organism and that of the laboratory animal following haematotoxic doses of the xenobiotics being evaluated. For example, changes in the total leukocyte count and the bone marrow cell count of rats following a single cytotoxic drug administration are the same as in a human subject but more rapid; the morphokinetics of blood and haematopoietic cell counts are very similar. An increase can be found in the circulating white blood cells in rats with marrow hypoplasia following repeated doses of a cytotoxic drug. This finding is very important, as marrow suppression endangers patient health. 1 But rats are mammals, not insects, and thus we must expand this area of comparative clinical pathology to select insect species.
Ethical limits are projected onto laws protecting both domestic and laboratory animals (vertebrates), these limits reflect the views of the lay public and, therefore, general political influences. 31 In contrast, a strengthened effort to protect human health against environmental hazard, demands assays predicting xenobiotic influences on human subjects rapidly, cheaply and without the use of vertebrates. We need to monitor all substances used in industry, agriculture, chemotherapy or food products.
There are approximately one million species of invertebrates whose use is not restricted in the European Union, USA, Canada, etc. Several invertebrates are useful models in genetics and physiology (see Nation, 32 etc.), but have still not been evaluated for their eventual use in haematoxicology. There are only a small number of published scientific papers 33,34 (and several others to date) on the special effects of evaluated pollutants on the haemocytes of wild animals without an evaluation of their use in safety xenobiotic tests for haematotoxicological screening with respect to hazards to human health. Moreover, Wrisberg et al. 35 and several other writers used haemocytes in the micronucleus test for the screening of mutagenicity.
The differences between human physiology and morphology and those of insects are great. But these differences seem to be smaller than the differences between bacteria and the human body, although genotoxicological screening on microorganisms has been verified by long experience and reliable results; e.g. the Ames Salmonella assay is the most widely used in vitro genotoxicity assay (cf. Purves et al. 36 ). Thus, the problem is not in the evolutionary distance between men and insects but in the definition of the limits of the use of insects as alternative biomodels.
Clearly, the physiology of the insect blood system is very different from that of humans. However, the haemocytes themselves are very similar to vertebrate blood cells in a number of aspects. In particular, the haemocytes of insects (see Table 1 for a survey) are similar to vertebrate blood cells with respect to their main function in the immunological reaction (see Hoffmann and Reichhart; 37 Iwanaga and Bok, 28 for example). Insects do not have special haematopoietic organs as do humans; insect haemocytes can proliferate in the body fluid called haemolymph. 38–40
Potential role of insect alternatives in preclinical haemocytotoxicity tests for xenobiotic safety
HL: haemolysis; HS: haemopoietic suppression; HR: haemopoietic recovery; LR: leukaemogenic reactions; CF: cell function alterations; BL: blood loss; CI: coagulation injuries
The effects on haemocytes can be similar to human blood cell reactions although there are large evolutionary differences between men and insects (Table 1), and increasingly, alternative biomodels for haematotoxicology among insect species are being discovered. Several similarities between blood cells of mammals and insect haemocytes have been documented. 41 There are several types of haemocyte morphology which slightly varies among species. 42–45 Prohaemocytes have a morphology close to human lymphocytes; prohaemocytes are also frequent in the differential count. Granulocytes are similar to human neutrophil granulocytes and can also be numerous. Plasmatocytes are morphologically close to human monocytes but are not numerous. Insect spherulocytes, oenocytoids, lamellocytes and crystal cells are like no human blood cells but they are as rare as other haemocyte types found only in a small number of invertebrate species.
Haemocytes participate in the humoral and cellular immune defence reactions against microbes and parasites 27,46,47 in a way similar to mammalian leukocytes. Haemocytes also participate in phagocytosis and encapsulation of foreign intruders in the haemolymph and in its coagulation. Mammalian haematopoietic production consists of erythroid, lymphoid and myeloid lineages. Erythroid cells are not needed in insects because of their different methods of transport of oxygen.
The mammalian mature cells in both the lymphoid and myeloid lines, i.e. leukocytes, play an important role in the immune response. The mature lymphoid cells, i.e. lymphocytes, possess an ability to achieve somatic gene rearrangement, which enables them to produce many immunoglobulins and receptors. We are not sure if there is an equivalent of lymphoid cells in insects, although, prohaemocytes have morphological characteristics close to lymphocytes. 43–45 On the other hand, in some species, e.g. Drosophila, prohaemocytes seem to be progenitors of other haemocytes and the fat body is rather a major immune-responsive tissue that originates from the mesoderm during histogenesis. 47–49
Insect plasmatocytes form the bulk of capsules around foreign bodies too large to be phagocytosed, and nodules around masses of bacteria or necrotic material. Plasmatocytes are involved in the production of antimicrobial peptides. Their role in phagocytosis is disputed, but Ribeiro and Brehélin 44 revised contradictory findings when they also observed low amounts of phagocytosing plasmatocytes. These authors suggest that different results in studies on plasmatocyte phagocytosis probably depend on the conditions of the experiments. Plasmatocytes seem to be conformable to mammalian blood monocytes.
The main function of insect granulocytes is phagocytosis; another important function is the exocytosis of opsonin-like substances. 50 The signalling pathways, that regulate the formation of phagosomes and ingestion, are activated by the invader and include focal adhesion kinase and mitogen-activated protein kinase pathways in both mammalian phagocytes and insect plasmatocytes or granulocytes. 51 In insects and mammals, activation of surface receptors induces intracellular signalling pathways leading to the cytoplasmic remodelling required for phagosome maturation and particle killing or dissolution. Evidently, the basic phagocytic signalling pathways in insect haemocytes (plasmatocytes and granulocytes) and mammalian leukocytes (monocytes and granulocytes) have remained unchanged during evolution. 49 Apparently, phagocytosis is an evolutionarily conserved process essential for the removal of invading pathogens and apoptotic bodies.
Current perspectives of insect haematotoxicology
The question arises of whether toxic effects on haemocytes are similar to or different from human haematology. The published data mentioned above show that we still do not know which insect species have the most similar reactions in their haemocytes to those in the human blood and haematopoietic system, which are the differences in morphokinetics between human blood cells and haemocytes following ‘standard’ toxic substance administration, and which stage of insect development is more sensitive.
Many published articles describing the haematological differences among insect taxons (zoological approach) do not usually reflect the characteristics which make possible the comparison of adverse reactions between insects and mammals (the toxicological approach). Therefore, future research in this field should use fundamental laboratory methods which are applied to both human and veterinary haematology, including haemocyte numbers, differential counts and routine cytochemistry.
The techniques of human haematology seem to be very useful in insect haematology. 41 The general principles of cell structure are, of course, part of the haemocyte morphology and function and our preliminary results indicate that we can use the same cytological methods as in human haematology. 49,51,52 Activation of the superoxide forming respiratory burst oxidase of insect haemocytes has a similar molecular mechanism as in human neutrophils and justifies the use of insects in place of mammals for modelling the innate immune response. 53 The basic phagocytic signalling pathways in insects and mammals appear to have remained unchanged during evolution. They are an important function in mammalian monocytes and granulocytes as well as in insect granulocytes and plasmatocytes.
In contrast to human leukocytes, a count of insect haemocytes cannot be estimated without special cell counters; routine blood cell counters are not applicable, but older, simpler, methods for haemocyte counts are available. Special immunocytochemical techniques and computer image analysis should be introduced in this area of comparative clinical pathology. An alternative biomodel among invertebrates could be the species used for insect physiology studies. 32 Analogous to the toxicological studies carried out on mammals (cf. the use of laboratory rats instead of smaller mice in long-term toxicological studies 1 ), the use of an insect species which is not entirely small-sized would be preferable.
It seems that the best use of safety xenobiotic assays applied to insects could be at the stage of toxicological screening before preclinical subchronic and chronic studies on laboratory mammals; these highly expensive tests will follow when no undesirable haematotoxicity in insects was detected. On the contrary, haematoxicological reactions in insects indicate the requirement for a pilot subchronic study on a small number of rodents of one species; if such a pilot study confirms previous findings on insects, the new xenobiotic would have to be rejected. Optionally, it could be chemically modified and subjected to the recommended sequence of tests with insects and rodents.
Although insects are very useful at present in detecting the undesirable effects of ‘environmental’ chemicals, their routine use for toxicological screening requires some additional standardization with respect to laboratory environmental conditions, age-related sensitivity, biorhythms and accuracy of haematological methods using various toxic ‘standards’.
‘Standard laboratory insects’ will offer a cheap biomodel (approximately 100-fold reduction in costs) which is faster (approximately 10 times) than recent bioassays on mammals, and moreover, will not be subject to the ethical limitations of studies on animals.