---

Abstract

The effects of insulin on the in vitro growth characteristics of Trypanosoma granulosum were investigated. The insulin growth stimulus had a rapid onset with little or no lag phase but was relatively short lived, growth peaking typically on day 3 or 4 of culture. This contrasted with medium containing 10% foetal calf serum, normally used for in vitro culture of this trypanosome, which after a 3 day lag stimulated sustained log-phase growth. Insulin demonstrated a biphasic dose response with maximum growth typically occurring at 3–10μg/ml and a much reduced effect at higher concentrations. These findings suggest a possible role for host insulin and/or trypanosomal insulin-like molecules in regulating parasite growth in vivo.

---

Keywords: Insulin, Trypanosome, Parasite, Growth.

^*Corresponding author

---

1 Introduction

Many micro-organisms make specific metabolic or mitotic responses to mammalian insulin; some have additionally been shown to synthesize insulin-like molecules (Lenard, 1992; Plotkin and Viselli, 2000). These phenomena have been demonstrated in a number of unicellular eukaryotes with Tetrahymena spp. being particularly widely studied (Christensen, 1993; Christensen et al., 1996; Heygesi and Csaba, 1997). However, there has been relatively little work on trypanosomes in this respect.

Successful culture of most trypanosomes and especially bloodstream forms, normally requires the use of semi-defined media containing foetal calf serum (FCS). Insulin is a key component of most fully defined media devised for mammalian cell culture. These facts, together with the published work on microbial insulin-like molecules, suggested a study of the effect of insulin on the growth of trypanosomes. Trypanosoma granulosum, a fish parasite known to be a useful model for studies of trypanosomal growth and metabolism (Davies et al., 1995; Mastri et al., 2001) has been used in these experiments.

Fully defined media substituting bovine insulin for FCS will permit growth of fish trypanosomes, such as Trypanosoma danilewskyi (Wang and Belosevic, 1994) and T. granulosum (Davies et al., 1995). The studies presented here have further examined the mitogenic effects of insulin on T. granulosum establishing the dose response, the onset and duration of the effect, relative to FCS.

2 Materials and methods

Stocks of T. granulosum were cultured essentially as described previously (Davies et al., 1995). Stock cultures were grown to 1–2×10⁷/ml in modified SDM-79 medium with 10% FCS. The cells were then washed in modified SDM-79 medium without FCS. For each experiment they were then seeded in 25cm³sterile flasks at an initial cell density of 0.5–1.0×10⁶/ml. Trypanosomes were cultured in modified SDM-79 medium either alone or supplemented with 10% FCS and/or bovine insulin (Sigma–Aldrich Ltd, UK). Bovine insulin was used in this study to permit more direct comparisons with bovine serum commonly used in semi-defined media.

Cultures (total volume 10ml in each case) were incubated at 20°C for 3–7 days and cell numbers counted every 24h using an Improved Neubauer Haemocytometer (Scientific Laboratory Supplies, Nottingham, UK).

For the dose–response curve (experiment 2) cells were counted on day 3 only as experiment 1 and similar previous experiments had shown peak cell numbers in insulin treated cultures after 3–4 days.

For all experiments triplicate cultures were used with each culture counted twice at each timepoint. Results are mean counts plus or minus standard deviations. Data were assessed for statistical significance using unpaired Student's t tests.

3 Results

The first experiment (Fig. 1) compared the growth of the trypanosomes in SDM-79 supplemented with 10% FCS, 10μg/ml bovine insulin (a concentration previously shown to be effective at stimulating growth of T. granulosum) or both. Fig. 1 clearly demonstrates the growth stimulation with insulin. Other marked characteristics of insulin only treated-cultures were little or no lag phase and a short lived stimulus, with maximum cell numbers typically achieved on day 3 or 4 and a marked decline after this. In contrast, SDM-79 with 10% FCS typically generated sustained logarithmic growth preceded by a lag phase. SDM-79 containing both 10% FCS and 10μg/ml insulin generated a similar growth pattern to that containing FCS only, cell numbers were slightly higher when both supplements were used but the difference was not statistically significant.

---

Fig. 1

Growth of T. granulosum in modified SDM-79 media alone (■), and in modified SDM-79 media supplemented with either 10% FCS (♦), 10μg/ml insulin (□) or both (▵). Cell numbers are means of triplicate cultures each counted twice, error bars indicate standard deviations. Cell numbers were significantly different (P<0.05) between cultures treated with insulin only and controls at each timepoint except days 1 and 6. The difference between insulin only and FCS only treated cultures was significant on days 2, 3 and 5. Comparisons between FCS only and FCS plus insulin treated cultures were not significant.

---

The second experiment illustrated a dose–response curve for insulin. Fig. 2shows the cell numbers present on day 3 for cultures with medium containing a range of insulin concentrations. A biphasic effect was demonstrated, growth increases with insulin concentration to a peak at 3–10μg/ml, with a much reduced effect at higher concentrations. A final study (experiment 3) was performed as for experiment 1 but with additional insulin (to final concentration of 10μg/ml) added to the insulin-only medium on day 2 of culture. Compared with experiment 1 this prevented the fall in cell numbers on day 4 (with additional insulin 2.17±0.34×10⁶/ml, without additional insulin 1.11±0.21×10⁶/ml) and day 5 (with additional insulin 2.04±0.31×10⁶/ml, without additional insulin 0.94±0.15×10⁶/ml) but did not regenerate the growth rates seen on days 2 and 3, or those seen with FCS.

---

Fig. 2

Dose response of the mitogenic effect of insulin on T. granulosum. Growth on day 3 of culture, expressed as a percentage of control (zero insulin concentration) is plotted against log insulin concentration. Percentages are means of triplicate cultures each counted twice, error bars indicate standard deviations. Insulin treated cultures showed significant difference (P<0.05) from control for 1, 3, 10 and 30μg/ml of insulin.

---

4 Discussion

Insulin-like molecules are present in a wide range of organisms including micro-organisms; they have been categorized as insulin-like on the basis of chromatographic characteristics, immunoreactivity and biological activity (Lenard, 1992; LeRoith et al., 1980). Although genes coding for insulin-like peptides have been well documented in invertebrates, including insects, molluscs, crustacea and the nematode Caenorhabditis elegans an insulin-like gene has so far been described in only one micro-organism, the fungus Neurospora crassa. This was classified as a pseudogene (Muthukumar and Lenard, 1991). As well as synthesizing insulin-like molecules, many micro-organisms respond to mammalian insulin by an increased metabolic rate and cell division (Lenard, 1992; Plotkin and Viselli, 2000).

Trypanosomes have been poorly studied with regard to these phenomena. They warrant further investigation, as some species are important human or livestock pathogens. In contrast to Tetrahymena, they are examples of primitive unicellular eukaryotes and their bloodstream forms will be continuously exposed to host insulin.

The data presented here reveal that insulin and FCS have markedly different effects on the growth of T. granulosum. The growth response to insulin is rapid with little or no lag phase, but is relatively short-lived, peaking typically on day 3 or 4 of growth in culture. This contrasts with FCS, which following a 3 day lag promotes sustained log-phase growth. The growth response to insulin is specific, since equivalent concentrations of protein in the form of either bovine serum albumin or transferrin do not stimulate growth of T. granulosum (Coleman et al., 1997; Davies et al., 1995). The dose–response experiment demonstrates maximal growth at insulin concentrations similar to those shown for T. danilewskyi (Wang and Belosevic, 1994) and Tetrahymena (Heygesi and Csaba, 1997). It is also similar to insulin concentrations shown to prevent cell death and enhance replication in Tetrahymena seeded at low densities in chemically defined medium (Christensen et al., 1993). The reduced effect at higher concentrations and its relatively short duration also agrees with the findings of Heygesi and Csaba (1997) for Tetrahymena. It has been shown that under certain experimental conditions insulin concentrations below 1ng/ml can stimulate proliferation in Tetrahymena (Christensen et al., 1996). Our experiments could not demonstrate a mitotic effect for T. granulosum with insulin concentrations lower than 0.1μg/ml.

How does insulin affect growth? For the free living protozoan Tetrahymena it could be replacing or supplementing an insulin-like molecule produced by the organism itself as an autocrine/paracrine growth factor (Wheatley et al., 1993). This may happen with T. granulosum, but in the case of this parasitic protozoan, it is also possible that insulin added to cultures is substituting for host insulin. In growing cultures, the lag phase may be characterized and perhaps even in part regulated by low microbial ‘insulin’ concentrations; this could explain the relative lack of a lag phase with insulin stimulated cultures of T. granulosum. As demonstrated here, and reported by others, insulin concentrations approaching 100μg/ml (10⁻⁵M) are much less effective in promoting growth. This may also form part of a control mechanism. At very high cell densities local insulin concentrations could become sufficiently high so as to moderate further growth. Indirect support for this suggestion is provided by Koppelhus et al. (1994), who studied Tetrahymena chemoattractants and showed that for only a tenfold increase in concentration, insulin changed from being a strongly positive to a powerfully negative chemoattractant. This phenomenon was not apparent for any of the other peptides they studied.

FCS typically contains 0.05–0.2ng/ml insulin (the value for the batch used in our experiments was 0.12ng/ml); this is much lower than insulin concentrations used in defined media in our experiments and several other studies. Adding insulin at 10μg/ml to FCS produced little enhancement of growth promotion and did not eliminate the lag phase (Fig. 1). It is possible that FCS contains factor(s) which suppress the action of insulin (either host or parasite derived) on parasite growth, enabling the host to delay the onset of trypanosomal proliferation, hence the lag phase in all experiments where FCS was used. The short duration of the insulin effect is not overcome by the addition of further insulin on day 2 as this maintained cell numbers but did not restore growth (experiment 3). After generating peak cell numbers on day 3 insulin is either readily degraded or becomes limited in its mitogenic effects. Insulin action appears to be mainly restricted to the early stages of growth and therefore other component(s) of FCS, in addition to insulin, are probably required for sustained exponential growth.

In conclusion, insulin clearly stimulates the growth of T. granulosum in vitro and hence may play an important role in vivo. The characteristics of insulin stimulation compared to FCS suggest that the interaction between these mitogens is complex and should be furtherexplored.

Acknowledgments

We wish to acknowledge the technical assistance of Mrs Gurmeet Sappal.

References

Christensen ST. Insulin rescues the unicellular eukaryote Tetrahymena from dying in a complex, synthetic nutrient medium. Cell Biol Int 1993:17:833-7
Crossref   Medline   

Christensen ST, Quie, H, Kemp, K, Rasmussen, L. Insulin produces a biphasic response in Tetrahymena thermophila by stimulating cell survival and activating proliferation in two separate concentration intervals. Cell Biol Int 1996:20:437-44
Crossref   Medline   

Coleman K, Mackintosh, D, Davies, AJ. An investigation into the mitogenic components of foetal calf serum required for the culture of Trypanosoma granulosum. J Eukaryot Microbiol 1997:44:24A

Davies AJ, Mastri, C, Thorborn, DE, Mackintosh, D. Experiments with growth of the eel trypanosome, Trypanosoma granulosum Laveran & Mesnil, 1902, in semi-defined and defined media. J Fish Dis 1995:18:599-608
Crossref   

Heygesi H, Csaba, G. Time and concentration dependence of the growth promoting activity of insulin and histamine in tetrahymena. Application of the MTT-method for the determination of proliferation in a protozoan model. Cell Biol Int 1997:21:289-93
Crossref   Medline   

Koppelhus U, Hellung-Larsen, P, Leick, V. An improved quantitative assay for chemokinesis in Tetrahymena. Biol Bull 1994:187:8-15
Crossref   Medline   

Lenard J. Mammalian hormones in microbial cells. Trends Biochem Sci 1992:17:147-50
Crossref   Medline   

LeRoith D, Shiloach, J, Roth, J, Lesniak, MA. Evolutionary origins of vertebrate hormones: substances similar to mammalian insulins are native to unicellular eukaryotes. Proc Natl Acad Sci USA 1980:77:6184-8
Crossref   Medline   

Mastri C, Thorborn, D, Davies, AJ, Ariyanayagam, MR, Hunter, KJ. Polyamine and thiol metabolism in Trypanosoma granulosum: similarities with Trypanosoma cruzi. Biochem Biophys Res Commun 2001:282:1177-82
Crossref   Medline   

Muthukumar G, Lenard, J. A preproinsulin-like pseudogene from Neurospora crassa. Mol Cell Endocrinol 1991:82:275-83
Crossref   Medline   

Plotkin BJ, Viselli, SM. Effect of insulin on microbial growth. Curr Microbiol 2000:41:60-4
Crossref   Medline   

Wang R, Belosevic, M. Cultivation of Trypanosoma danilewskyi(Laveran & Mesnil, 1904) in serum-free medium and an assessment of the course of infection in goldfish, Carassius auratus (L.). J Fish Dis 1994:17:47-56
Crossref   

Wheatley DN, Christensen, ST, Schousboe, P, Rasmussen, L. Signalling in cell growth and death: adequate nutrition alone may not be sufficient for ciliates: a minireview. Cell Biol Int 1993:17:817-23
Crossref   Medline   

---

Received 30 May 2002; accepted 19 September 2002

doi:10.1016/S1065-6995(02)00251-2

---