Functional and structural comparison of cytokines in different species

https://doi.org/10.1016/S0165-2427(99)00115-4Get rights and content

Abstract

As the number of recombinant cytokines increases, so does our knowledge of their structure and function in different species. The biological cross-reactivity of cytokines from one species on cells from a different species has been reported on in the literature but this information is scattered over many publications and some of it has not yet been published. Comparing sequence information combined with three-dimensional and receptor-binding information (i.e. biological cross-reactivity) in different species provides insight into the underlying rules governing cross-reactivity and conservation. It was observed that there is quite a strict threshold of 60% amino acid identity above which cytokines tend to cross-react. Below this threshold few cytokines cross-react on cells from a different species.

When comparing frequencies of reported species cross-reactivities between cytokines belonging to different cytokine – folding families it is obvious that not all cytokines within these folding families are equally cross-reactive. The underlying reason for these differences may lay in the ability of certain folding families to accumulate more mutations and still produce a protein, which is able to fold in the desired tridimensional structure. For example, cytokines belonging to the short 4-α-helix bundle can accumulate mutations in the 4-α-helixes and large loops connecting the 4-α-helixes and are the least cross-reactive. In contrast, cytokines belonging to the β-sheet based folds (β-trefoil and β-sandwich) are the most cross-reactive and also the most conserved cytokines amongst the different species studied.

Introduction

As the number of recombinant cytokines increases, so does our knowledge of their function in different species. At this stage the number of cloned cytokines in different species varies widely. The number of cloned cytokines at present are: 36 cytokines for human and rodents, 23 for bovine, 20 for sheep, 19 for pigs, 11 for cats and horses, 7 for birds, 5 for dogs and 4 for fish. This is obviously dependent on the adopted definition of ‘cytokine’ and the number of variants of the same cytokine one wants to include (the same cytokine in different species of the same group were considered as 1 cytokine for this purpose). Most of these cytokines were cloned using PCR – based techniques exploiting existing sequence homology with cytokines cloned from different species, especially mouse and human. Although cytokines were identified in species other than those listed above very few were cloned and little information is available about them.

Comparing sequence information combined with tri-dimensional and biological cross-reactivity information in different species provides insight into the rules that govern biological cross-reactivity of a cytokine on cells from a different species. The biological activity, depending on the cytokines considered, can be measured by a variety of effects they have on cells including: proliferation/survival, protection from viral infection, apoptosis, secretion of other cytokines, differentiation and induction of cell surface markers. However, each of these effects is the result of a successful interaction between the cytokine and the receptor on the cell surface. Therefore, whatever the readout chosen to measure biological activity, this measure represents binding of the cytokine from one species to the receptor of another species. Choosing this readout instead of, for example, cross-reactivity of monoclonal antibodies, therefore limits the study to one or a few tri-dimensional regions of the protein which has receptor binding properties. Moreover, this region is most likely exposed to selective pressure associated with the need to bind to the receptor on the surface of the cell. As a correlate for cross-reactivity, amino acid identity as opposed to DNA homology was chosen, as this reflects differences in tri-dimensional structure and therefore avoids the pitfalls of considering silent mutations in the analysis. The present study may, therefore, help estimate the likelihood of a particular cytokine from one species to cross-react with cells from another species based on amino acid sequence information.

Section snippets

Cross-reactivity patterns of IFN-γ closely relates to the evolutionary distance between the IFN-γ sequences in these species

Interferon-γ (IFN-γ) is one of the cytokines that has been cloned in a large variety of species. Moreover, because the bioassay, based on its ability to clear virus-infected cells, is relatively straightforward it is available in many different species. Therefore, this cytokine is ideal to investigate biological cross-reactivity and sequence homology. Based on the Kimura protein distance algorithm (Kimura, 1981) the IFN-γ molecule of different species was compared and found to form three groups

Cross-reactivity of cytokines derived from different species in function of homology at amino acid level

To expand the findings described above a collection of 47 cytokines known to possess the ability to cross-react with cells from another species and 22 cytokines known not to cross-react with cells from another species were investigated. In order to qualify for inclusion in the study the cytokines of both species needed to have been sequenced and their sequence made available. Those cytokines included interleukin (IL)-1α, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12, IL-13,

Cross-reactivity of cytokines belonging to different cytokine fold families

Cytokines can be classified in a number of cytokine – fold families according to their tridimensional structure or expected tri-dimensional structure. The cross-reactivity patterns of cytokines were correlated to the different cytokine – folding families to which these cytokines belong. In doing this, it was assumed that the homologue cytokines from different species would belong to the same cytokine fold family. This assumption was made on the basis that it would be very difficult to envisage

Concluding remarks

The data presented in this paper indicate that it is more likely that a particular cytokine would cross-react on cells from another species if the homologue cytokine in the second species has at least 60% amino acid identity with the cytokine of the first species. While it is not surprising that, statistically, cytokines would tend to cross-react more often as their % amino acid identity increases it was not anticipated that the cut-off would be as marked across the wide variety of species

References (4)

  • M.H. Hecht et al.

    Science

    (1990)
  • M. Kimura

    Proc. Natl. Acad. Sci. USA

    (1981)
There are more references available in the full text version of this article.

Cited by (0)

View full text