Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Differences in Tolerance to Host Cactus Alkaloids in Drosophila koepferae and D. buzzatii

  • Ignacio M. Soto ,

    Affiliations Instituto de Ecología, Genética y Evolución de Buenos Aires, Consejo Nacional de Investigaciones Científicas y Tecnológicas, Buenos Aires, Argentina, Departamento de Ecología, Genética y Evolución, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina

  • Valeria P. Carreira,

    Affiliations Instituto de Ecología, Genética y Evolución de Buenos Aires, Consejo Nacional de Investigaciones Científicas y Tecnológicas, Buenos Aires, Argentina, Departamento de Ecología, Genética y Evolución, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina

  • Cristian Corio,

    Affiliations Instituto de Ecología, Genética y Evolución de Buenos Aires, Consejo Nacional de Investigaciones Científicas y Tecnológicas, Buenos Aires, Argentina, Departamento de Ecología, Genética y Evolución, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina

  • Julián Padró,

    Affiliations Instituto de Ecología, Genética y Evolución de Buenos Aires, Consejo Nacional de Investigaciones Científicas y Tecnológicas, Buenos Aires, Argentina, Departamento de Ecología, Genética y Evolución, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina

  • Eduardo M. Soto,

    Affiliations Instituto de Ecología, Genética y Evolución de Buenos Aires, Consejo Nacional de Investigaciones Científicas y Tecnológicas, Buenos Aires, Argentina, Departamento de Ecología, Genética y Evolución, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina

  • Esteban Hasson

    Affiliations Instituto de Ecología, Genética y Evolución de Buenos Aires, Consejo Nacional de Investigaciones Científicas y Tecnológicas, Buenos Aires, Argentina, Departamento de Ecología, Genética y Evolución, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina


The evolution of cactophily in the genus Drosophila was a major ecological transition involving over a hundred species in the Americas that acquired the capacity to cope with a variety of toxic metabolites evolved as feeding deterrents in Cactaceae. D. buzzatii and D. koepferae are sibling cactophilic species in the D. repleta group. The former is mainly associated with the relatively toxic-free habitat offered by prickly pears (Opuntia sulphurea) and the latter has evolved the ability to use columnar cacti of the genera Trichocereus and Cereus that contain an array of alkaloid secondary compounds. We assessed the effects of cactus alkaloids on fitness-related traits and evaluated the ability of D. buzzatii and D. koepferae to exploit an artificial novel toxic host. Larvae of both species were raised in laboratory culture media to which we added increasing doses of an alkaloid fraction extracted from the columnar cactus T. terschekii. In addition, we evaluated performance on an artificial novel host by rearing larvae in a seminatural medium that combined the nutritional quality of O. sulphurea plus amounts of alkaloids found in fresh T. terschekii. Performance scores in each rearing treatment were calculated using an index that took into account viability, developmental time, and adult body size. Only D. buzzatii suffered the effects of increasing doses of alkaloids and the artificial host impaired viability in D. koepferae, but did not affect performance in D. buzzatii. These results provide the first direct evidence that alkaloids are key determinants of host plant use in these species. However, the results regarding the artificial novel host suggest that the effects of alkaloids on performance are not straightforward as D. koepferae was heavily affected. We discuss these results in the light of patterns of host plan evolution in the Drosophila repleta group.


Phytophagous insects are excellent model systems to investigate the genetic and ecological basis of adaptation and interspecific divergence, since their host plants constitute the most immediate environmental factor affecting early life cycle stages [1]. In this regard, the role of ecology in speciation has been systematically evaluated [2], [3], [4] and a recent metanalysis involving groups of angiosperms, fishes, frogs, birds, pigeons, butterflies and fruit flies revealed a link between ecological divergence and reproductive isolation [5], [6]. Changes in habitat/diet were shown to be positively associated with reproductive isolation in insects [5], [7] supporting the notion that ecological consequences of host plant shifts may be responsible for the remarkable diversity of phytophagous groups.

Host specificity in phytophagous insects especially in monophagous species, is thought to be based on chemical and/or nutritional characteristics of the plant [1], [8]. Thus, shifts to new host plants often involve challenges to exploit a new food source, face chemically diverse environments (including potentially toxic compounds), new mating environments, parasitoids, bacteria and fungi [9], [10], [11]. Hence, host plant shifts may accelerate divergence in features associated with performance in new hosts [12], [13], [14], [15], [16], and sensory systems, like those involved in smell and taste [17], [18]. Similarly, changes in morphology associated with host plant shifts are well documented in insects [17], [19], [20], [21]. However, many other factors may influence host choice and the evolution of specialists versus generalists. For instance, chemical properties along with other features such as temporal and spatial availability are relevant factors that influence the suitability of hosts. Studies of host plant chemistry have resulted in a general understanding of insect-plant relationships (reviewed in [1]). Some host plants have evolved metabolic pathways responsible for an extraordinary variety of secondary metabolites that reduce damage by herbivores [1]. As a matter of fact, the chemical particularities of plants are thought as relevant factors that shape the ensemble of insects that use a plant as feeding or breeding site [8].

Most species in the genus Drosophila are saprophytophagous and breed on necrotic plant tissues and feed upon the microorganisms associated to the decaying process [22]. The ecology of Drosophila breeding sites has been an issue of interest for evolutionary biologists because of the prominent role that several members of the genus played in genetics and evolution [23]. Among such groups the repleta group (subgenus Drosophila) radiated in the New World due to the ability of flies to utilize decaying cacti as breeding substrates [24]. This capacity allowed some species subgroups to invade and diversify in the deserts of the Americas, areas that are rather inhabitable for other Drosophila [24], [25].

The radiation of the Cactaceae has been accompanied by the acquisition of a broad battery of secondary metabolites (allelochemicals) as alkaloids, medium chain fatty acids, sterol-diols, and triterpene glycosides. The latter serve as feeding deterrents and isoquinoline alkaloids obstruct neurotransmission (reviewed in [26]). Moreover, it has been argued that toxic compounds affect fitness related traits [28], [29] and determine patterns of host plant use in cactophilic Drosophila that inhabit the Sonoran Desert [27]. Moreover, in some cases the effects are so dramatic that “mistakes” in the choice of a breeding substrate might result in the death of the insect. As an example, D. pachea is restricted to senita cactus (Lophocereus schotii) due to a strict nutritional requirement for Δ7 sterols (only found in senita) to complete development (reviewed in [26], [30], [31]).

The D. buzzatii cluster is an ensemble of at least seven species in different stages of divergence and varying degrees of host specialisation [32]. The sibling species D. buzzatii and D. koepferae are sympatric in most of the distribution range of the latter in the areas arid lands of northwestern Argentina and southern Bolivia [33], [34]. The former breeds primarily on the decaying cladodes of several species of the genus Opuntia (prickly pears) and secondarily on columnar cacti of the genera Cereus and Trichocereus, whereas the reverse is true for D. koepferae [35]. Since divergence, D. buzzatii and D. koepferae have evolved differences in life-history and morphological traits expressed in both primary and secondary natural hosts [36], [37], [38], [39], and also a remarkable oviposition preference on their respective primary natural hosts [15], [40]. These observations have been interpreted as adaptations that allow flies to efficiently exploit resources that differ markedly in spatial and temporal predictability as well as in chemical properties [34], [35], [36], [41]. Opuntia species contain acids that are typical of succulents (eucomic, phorbic and piscidic), lipids, terpenes, free sugars and phenolic compounds [42] while cacti in the genus Trichocereus contains alkaloids such as candicine and trichocereine in T. candicans and T. terschekii respectively [43]. These chemical particularities have led to the suggestion that alternative cactus hosts may represent different chemical environments for the larvae that develop in the decaying plant tissues and for adult flies that feed on them [35], [36], [44]. In a previous study, we showed that alkaloids extracted from T. terschekii decrease viability and adult body size in D. buzzatii [45]. However, we did not test the hypotheses that alkaloids exerted a differential effect on performance in the columnar dweller D. koepferae vs. D. buzzatii.

Here, we carry out the first comparative assessment of the effect of cactus alkaloids on life history traits and fitness in D. buzzatii and D. koepferae and the evaluation of the intrinsic capabilities of each species to exploit a novel toxic host.

Materials and Methods

Collection of flies, cacti and stocks maintenance

Fly stocks used in this study derived from flies collected in San Agustín del Valle Fértil (30° 31′13′ ´ S, 67° 34′05′ ´ W; Province of San Juan, Argentina), where D. buzzatii and D. koepferae coexist [41]. Flies were collected by net sweeping over fermented banana baits. Collection permits for both flies and cacti tissues were issued to IMS and JP by the Conservation Management and Protected Areas (Secretariat of Environment and Sustainable Development. Province of San Juan, File N° 1300-0236-13).

In the laboratory, wild flies were separated by sex and females were allowed to oviposit in vials in order to establish isofemale lines. As females of both species are morphologically indistinguishable isofemale lines were identified to species by inspecting genitalia of F1 males [46]. The offspring of 30 isofemale lines were mixed in equal numbers to establish two outbred stocks, one for each species, that were maintained in standard laboratory instant medium (Carolina Supplies) under a 12:12h light/dark photoperiod at 25±1°C.

Fresh tissues and fermenting cladodes of O. sulphurea and stems of T. terschekii were also returned to the laboratory from the same locality [40], [41]. Pieces of fresh cacti were stored at −20°C and cactus necroses of each species were maintained at 4° C in 25×25×15 plastic containers with sterilized cotton caps where fresh cactus was added every month (during the three months of the experiment).

Extraction and isolation of alkaloids from T. terschekii

Fresh tissues of T. terschekii were ground and blended with EtOH (1 l/ 1 kg tissue) and then filtered. The organic extract was concentrated on a rotatory evaporator to an aqueous suspension and acidified with 500 ml of 10 % HCl. The aqueous acidic fraction was partitioned between CH2Cl2 (extracted three times with 500 mL) and water to yield a dichloromethane fraction and a water soluble fraction (see [45]). The former was evaporated in a rotatory evaporator yielding a non-basic fraction containing acid lipid soluble compounds (e.g. terpenoids, fatty acids, sterols, aromatic and other compounds). This fraction, hereafter referred to as the “non-alkaloid fraction; NA”, was included as a separate treatment in the experiments described below to investigate its possible biological effects, since, a priori, we did not know which fraction contained potential toxic compounds other than alkaloids responsible for the differential effects that T. terschekii has on D. buzzatii and D. koepferae. The organic fraction was dried to yield a crude alkaloid fraction, hereafter referred to as the “Alkaloid fraction; A”. The identification of the active components in the alkaloid fraction was accomplished via mass spectrometry. We confirmed the presence of three compounds: trichocereine, N-dimethylmescaline, a phenylethylamine alkaloid typical of this species, mescaline and the analogue α-methylmescaline [45]. The natural concentration of alkaloids in fresh T. terschekii estimated from the collected material was 0.33 mg/g of wet fresh weight and 4.50 mg/g in the dry sample (0.3% w/w). Both the alkaloid (A) and the non-alkaloid (NA) fraction were solubilized in dimethyl sulfoxide (DMSO 100 µgr/ml) and used to design the artificial diets used in the bioassays. The same extraction protocol was applied to O. sulphurea samples but no traces of alkaloids were detected.

Experimental design

In order to obtain first-instar larvae, two egg-collecting chambers for each species were set up with a Petri dish containing egg laying medium (2% agar + commercial yeast). One hundred pairs of sexually mature flies of the same stock were released into each chamber. Petri dishes were removed 12 h later, inspected for the presence of eggs and incubated for another 24 hours to allow larval hatching. For each treatment, groups of 30 first instar larvae were randomly sampled from the plates and seeded in vials with the corresponding rearing medium (five vials or replicates per treatment). All vials contained 0.8 g of standard Drosophila instant medium (Carolina Biological Supply Company, Burlington, NC) hydrated with 4 mL of a solution containing sodium methylparaben (Nipagin, 0.02 g/v) as fungicide and the corresponding amount of the alkaloid or non-alkaloid fraction depending on the experiment (see below).

For the assessment of the effects of alkaloids on each species' performance, we raised batches of larvae in vials containing laboratory culture medium and increasing doses of the alkaloid fraction (A-treatments). The first set of vials contained standard laboratory medium plus the alkaloid fraction to a final alkaloid concentration that was close to its concentration in cactus tissues (A1X treatment). The other two sets of vials contained standard lab medium and alkaloid concentrations that were 50% (A1.5X) and 100% (A2X) higher than in the A1X treatment. The rationale of including treatments varying in the concentration of the alkaloid fraction was to uncover natural variation that flies may encounter in nature. Actually, alkaloid concentration may vary depending on cactus age and other ecological variables as soil properties and elevation [47], [48], [49]. In addition, water evaporation may contribute to increase alkaloids concentration in the rotting pocket during the decaying process [50].

We also evaluated the possible effect of the non-alkaloid fraction obtained during the extraction and isolation of alkaloids because some columnar cacti contain other secondary compounds such as triterpene glycosides, sterol diols and rare fatty acids [27] that may affect flies. For instance, cis-vaccenic acid, a rare isomer of oleic acid is abundant in T. terschekii [51] and previously reported as a pheromone precursor in insects [52] [53].

Thus, the non-alkaloid fraction might also account, apart from alkaloids, for the differential effects that rotting cacti have on performance. In these experiments, we prepared three sets of vials with increasing doses of the non-alkaloid fraction. One set contain the same concentration of the NA fraction measured in fresh tissues (NA1X treatment) and the other two contained 1.5 (NA1.5X) and 2 (NA2X) times the amount added to the first set.

Finally, we evaluated if differences in performance between flies raised in media prepared with T. terschekii and O. sulphurea were due solely to the presence of alkaloids or, alternatively, to an interaction with the overall quality of the plant tissue. To test this hypothesis, we created an artificial host prepared with fermented tissues of O. sulphurea plus an amount of the alkaloid fraction that matched the alkaloid concentration in fresh tissues of T. terschekii, an artificial “novel host”. Then, we raised batches of 30 larvae of D. buzzatii and D. koepferae in this artificially created diet and, as controls, in vials containing semi-natural media prepared with T. terschekii or O. sulphurea rotting tissues.

Traits scored

Performance in each rearing condition was measured by means of the study of larval to adult viability (LV), developmental time (DT) and wing size (WS) as a proxy for body size. LV was estimated as the proportion of emerged adults relative to the number of larvae seeded in each vial/replicate of each treatment. DT was estimated as the elapsed time in hours from the time of transfer of first instar larvae to vials until adult emergence. For the measurement of this trait, emerged flies were collected and sexed every four hours. The right wings of adult males were removed and mounted on slides and images of wings were obtained with a digital camera mounted on a microscope. Ten landmarks were digitized using TpsDig [54] at the intersection of veins or at the intersection of veins with the margins of the wing following Soto et al. [39]. As a measure of WS, we calculated the centroid size of each individual configuration of landmarks, using the square root of the sum of the squared distances of each landmark to the centroid of the configuration, [55].

Statistical analyses

We combined LV, DT and WS into a single index that gives a proxy of overall host performance. We calculated a relative performance index (RPI; modified from [56], [57] for each vial using:

This equation is very straightforward since LV and WS are directly related and DT is inversely related to the efficiency in the use of a rearing substrate [58]. RPI Index combines the effect of viability, development time and size. However, development time and size could themselves be correlated in Drosophila (with opposite effects on the inclusive fitness) [59] but also see [60]. Thus, these two parameters may not be independent and inclusion of both in the calculation of RPI may provide a biased estimate of performance. Therefore we evaluated the degree of independence in our data through the estimation of Pearson correlations among traits for each species in every experimental condition.

Responses to increasing concentrations of alkaloid and non-alkaloid fractions were evaluated by means of regression analyses of performance on the dose in A and NA treatments for each species. In the regressions the value of the performance index calculated for each replicate was the dependent variable. Additionally we performed a test of Homogeneity of Slopes in order to evaluate differences in response to alkaloid and non-alkaloid treatments within each species. In these analyses, the value of the performance index and the concentration of the respective fraction were considered as the dependent and independent variables, respectively, and fraction, alkaloid vs. non-alkaloid, was included as a categorical independent factor.

Performance variation among flies raised in media prepared with T. terschekii, O. sulphurea and O. sulphurea plus alkaloids was evaluated by means of a two-way ANOVA with species and host as independent fixed variables. We also calculated coefficients of variation (CV) for each treatment as the ratio between the standard deviation and mean performance using the means calculated for each replicate as input data. As this coefficient measures the dispersion of data points (i.e. means of replicates) around the mean value corresponding to each treatment, it was used to compare the degree of variation among treatments even if their means were different. As there is no variance for each CV, confidence intervals were constructed using bootstrap estimates of the coefficient [61].

We also explored whether there was a correlation between performance and CV in each species by calculating Pearson product moment correlations. All data were inspected for normality and RPI was not normally distributed with a moderately positive skewness. Thus, in order to fulfill normality and homoscedasticity assumptions, we applied the square root transformation to the data (following [62]) before analyses. All statistical analyses were performed using GLM implemented in the STATISTICA 6.0 software package [63] except for bootstrapping that was performed using PoopTols [64].


Mean values for all traits measured in each experiment and treatment as well as the respective performance score and CV are reported in Table 1. Only D. buzzatii was significantly affected by the presence of alkaloids in the rearing medium. The regression of performance on alkaloid dose was significant in D. buzzatii but not in D. koepferae (Table 2, Figure 1). Increasing doses of the alkaloid fraction affected performance in D. buzzatii by decreasing LV and extending DT, but did not affect wing size (Table 2). However, alkaloids concentration did not affect any of the life history traits in D. koepferae (Table 1 and Table 2). Both species failed to show any response to the presence of the non-alkaloid fraction (Table 2). Though a trend of increasing viability at higher concentrations of the non-alkaloid fraction was observed in both species in Figure 1, regressions were not significant for D. buzzatii (p  =  0.46) or D. koepferae (p  =  0.41) (Figure 1). Heterogeneity of the regressions slopes of performance for alkaloid and non-alkaloid concentrations were significant in D. buzzatii (F1,36  =  4.77, p  =  0.035) but not in D. koepferae (F1,36  =  0.24, p  =  0.623). Regarding possible correlations among traits conforming the RPI index, a significant negative correlation was detected between viability and developmental time only in the alkaloid-increasing treatments in both species (r  =  −0.58 and r =  −0.53 for D. buzzatii and D. koepferae respectively). However, development time and size, regardless the expectation, did not show a significant correlation in any treatment for any species supporting it inclusion in the RPI index as different fitness proxies.

Figure 1. Performance as a function of concentration.

Mean relative performance (and 95 % confidence intervals) as a function of concentration for a) D. buzzatii and b) D. koepferae reared in medium with the alkaloid fraction (black symbols) or the non-alkaloid fraction (white symbols) extracted from the columnar cactus T. terschekii. Linear trends are shown.

Performance in cactus media

Analysis of variance of host dependent performance revealed significant differences between species, where D. buzzatii showed, on average, greater scores than its sibling (F1, 30 =  38.24, p< 0.01) between hosts (F2, 30 =  36.93, p< 0.01) and a significant host x species interaction (F2, 30 =  23.83, p< 0.01). Both species showed comparable performances in both natural hosts, but radically different responses when raised on the novel host. Performance differences between D. buzzatii reared in the Opuntia + alkaloid medium and T. terschekii medium were not significant. To the contrary, D. koepferae reared in the novel host exhibited significantly reduced performance as compared to flies raised in O. sulphurea and T. terschekii (Figure 2).

Figure 2. Performance in host cacti.

Mean relative performance (and 95 % confidence intervals) for D. buzzatii (circles) and D. koepferae (triangles) reared in their natural hosts (O. sulphurea and T. terschekii) and an artificial novel host made with Opuntia tissue added with alkaloids extracted from the columnar T. terschekii.

Coefficients of variation

There was a significant negative correlation between mean performance and CV in both species (r  =  −0.83 for D. buzzatii and r  =  −0.75 in D. koepferae, both p< 0.05; Figure 3) when all treatments were considered jointly. Hence, treatments in which flies had inferior performance also displayed greater variance among replicates (Figure 4). For alkaloid treatments, heterogeneity among samples increased with concentration in D. buzzatii but decreased in D. koepferae (Figure 4a). For treatments with the non-alkaloid fraction, higher concentrations were associated with lower among sample variance in both species (although this was more evident in D. koepferae; Figure 4b). The analysis of performance in the two types of cactus hosts showed that, along with the reduced performance of D. koepferae in the O. sulphurea + alkaloids medium, we detected a pronounced and concomitant increment of the CV (Figure 4c).

Figure 3. Coefficient of Variation and Performance correlation.

Association between the coefficient of Variation (CV) and the mean relative performance considering all treatments in D. buzzatii (circles) and D. koepferae (triangles). Linear trends are shown.

Figure 4. Treatments Coefficients of Variance.

Coefficients of variance (CVs) for a) increasing concentrations of the alkaloid fraction extracted from T. terschekii, b) increasing concentrations of the non-alkaloid fraction extracted from T. terschekii and c) the natural hosts and the novel artificial host for D. buzzatii (grey bars) and D. koepferae (black bars). Confidence intervals were estimated via bootstrap (see text).


Dose-dependent effects of alkaloid fractions extracted from fresh tissues of T. terschekii had detrimental effects on performance in D. buzzatii but not in its sibling D. koepferae. The alkaloid fraction added to the rearing medium decreased viability and extended developmental time but did not affect wing size. However, D. koepferae did not show any significant response to increasing concentrations of alkaloids extracted from its natural host in any of these fitness components suggesting a well-developed tolerance or some similar kind of specialization. Egg-to-adult viability and mean development time are fitness parameters sensitive to environmental conditions and are known to be affected by alkaloids in several species of desert Drosophila [65], [66].

Conversely, increasing concentrations of the non-alkaloid fraction did not affect performance in D. buzzatii or in D. koepferae thus limiting the detrimental effects of the rearing cacti to the alkaloids-enriched portion. The analyses of the coefficients of variation of different treatments showed the other side of the same phenomenon. CVs were greater in treatments with high alkaloid concentration in D. buzzatii while CVs in D. koepferae were lower as both alkaloid or non-alkaloid fractions got more concentrated in the rearing medium. Higher performance scores were negatively associated with higher CVs indicating that a more efficient exploitation of the rearing media was related to small variation among replicates, an indication of low environmental stress [67]. These results support our predictions based on previous field and laboratory studies that alkaloids may be more harmful to D. buzzatii than to D. koepferae [45]. D. koepferae and D. buzzatii are differentially attracted to T. terschekii and O. sulphurea respectively [34], [36], [40].. Field studies have shown that D. buzzatii primarily uses rotting cladodes of prickly pears while D. koepferae uses columnar cacti [35] [40] and that D. buzzatii females lay more eggs on prickly pears whereas D. koepferae prefer to oviposit on columnar cacti [15].. Laboratory experiments also provided evidence of the pervasive and differential effect of cactus hosts on relative performance of D. buzzatii and D. koepferae. Both species are more viable, develop faster and showed increased mating success when reared in their respective primary host [34], [68], [69]. Moreover, adult flies of both species reared on the secondary host plant exhibit increased levels of wing fluctuating asymmetry [39], long considered a measure of developmental instability.

Previous studies have attributed the effects of cactus hosts to differences in chemical composition between hosts, i.e. both toxicological and nutritional properties of the environments offered by different host plants to growing larvae [35], [36], [39]. We confirmed that the presence of alkaloids is one key factor mediating the differential performance in D. buzzatii and D. koepferae. Although the metabolic pathways affected remain unclear, it is known that some alkaloids block steroid metabolism or assimilation of phytosterols [70], [71] and that alkaloid ingestion during larval life may negatively affect viability during metamorphosis [45], [72].

Before discussing the implications of our results, we would like to address the viability decline and extension of developmental time expressed by D. koepferae in experimental media (control, A1−A3 and NA1-NA3) when compared with overall performance in cactus media (Table 1). Similar trends were reported in previous studies [34], [41] suggesting that the nutrients required by D. koepferae are missing in Drosophila instant medium used to prepare the experimental media (potato flakes, commercial yeast, agar, glucose), at variance with our observations in D. buzzatii. In fact, these results are in line with our proposal that D. koepferae is a specialist and that D. buzzatii is a more generalist species. Nonetheless, these observations do not affect our conclusion of differential effects of alkaloids and the artificial cactus host (see below) on D. buzzatii and D. koepferae.

The evolution of cactophily suggests that acquisition of the capacity to degrade an array of toxic compounds present in rotting cacti could be considered an ecological apomorphy of the Neotropical lineage comprising the repleta, nannoptera and mesophragmatica species groups [32]. This ability allowed some subgroups to invade and diversify in cactus deserts, areas generally unfavourable for other Drosophila [24], [25]. Within the repleta group, the D. buzzatti cluster, which includes D. buzzatii, D. koepferae and at least five other Neotropical species, evolved from an ancestral Opuntia-feeder species that eventually began to exploit columnar cacti [32]. Therefore, D. buzzatii may represent the plesiomorphic state of host use compared to D. koepferae and all other extant species, that form the so called “serido sibling set”, which specialized in the exploitation of different columnar cacti [32], [33], [73]. However, this may not be true since ancestral character state reconstruction of host plant use in the repleta group indicated no phylogenetic structure [32]. In addition, though host plant use in the fasciola species subgroup is poorly known, some members of this basal lineage within the D. repleta group [32], use epiphytic species of the genus Rhipsalis, which is closely related to the Cactoidea [74] and other arboreal cacti as well as fruits, flowers, and fungi [24].

Alkaloids may be a determinant of patterns of host plant use in the D. buzzatii cluster. D. koepferae has evolved the ability to use a wide array of columnar cacti in the genera Cereus, Trichocereus, and Neoraimondia [75] which produce alkaloids, whereas D. buzzatii is more specialized on the relatively homogeneous habitat offered by prickly pears [76]. The chemical differences between cactus types may condition the direction of host shifts; a host shift would be easier for D. koepferae than for D. buzzatii since it may imply a shift from a more toxicological environment, as T. terschekii, to a less demanding one, as O. sulphurea [76].

Host shifts are fundamental components of diversification in the evolution of plant-herbivore interactions. To assess the potential of host shifts that mediate speciation it is crucial to unveil the mechanisms involved in the efficient exploitation of novel resources by specialists [19], [77], [78], especially in the critical initial phase of a recently assembled new plant– herbivore interaction. Unfortunately, host plant specialists shifting to new hosts are rarely directly observed in nature [79], [80]. Here we tested this hypothesis by creating an artificial “novel host”, an ecological opportunity in the form of a cactus, nutritionally equivalent to the prickly pear O. sulphurea but with the alkaloid content and concentration of the columnar T. terschekii. Thus, both species were exposed to a nutrient-rich medium to which D. buzzatii is well adapted with the addition of a toxic compound to which D. koepferae is more familiar. Paradoxically, and despite the observation that D. buzzatii was most affected by alkaloids, it performed better in the alkaloid containing artificial host. Surprisingly, D. koepferae suffered a dramatic reduction in performance, especially in terms of viability, and exhibited a substantial increase in the CV of performance in comparison with the other treatments or its sibling. We predicted that this host shift should have been a toxicological challenge similar to its primary host but in an Opuntia-like nutritional environment. These results suggest that alkaloid tolerance of D. koepferae may be dependent on other components of the nutritional environment.

What are these nutritional differences between cacti? For instance, the profile of fatty acids is substantially different between hosts [76], [51]. Besides, Opuntia species contain larger amounts of free sugars [42] than columnar cacti that have a complex chemistry that includes the presence of toxic alkaloids and other potentially toxic compounds like atypical fatty acids and triterpenes [10], [42], [51], [81]. Fermenting tissues of O. sulphurea and T. terschekii also differ in the yeast community associated to the decaying process in nature (Mongiardino Koch personal communication). Thus, one possible explanation may be that the presence of alkaloids affected a key nutritional component (a particular yeast species) rendering rotting O. sulphurea a nutritionally deficient medium for D. koepferae but not for D. buzzatii.

The diversification of the cactophilic D. buzzatii species cluster has involved a history of specialization to columnar cacti and alkaloid tolerance from a more generalist ancestral stock resembling the extant D. buzzatii. It remains to be determined how many independent host shifts to columnar cacti there have been and to understand the physiological mechanisms involved in specialization to reveal the evolutionary history of these flies.


Authors would like to thank Nicolás Mongiardino Koch and Pedro Fontanarrosa for insightful comments and discussions during the drafting of the manuscript and Sergio Szajnman for invaluable technical assistance. The corrections and insightful comments of three anonymous reviewers are also greatly appreciated.

Author Contributions

Conceived and designed the experiments: CC VPC EH IMS. Performed the experiments: CC IMS JP VPC EMS. Analyzed the data: IMS VPC. Contributed reagents/materials/analysis tools: EMS JP. Wrote the paper: IMS VPC EMS JP CC EH.


  1. 1. Schoonhoven LM, Loon JJA, Dicke M (2005) Insect-Plant Biology. Oxford University Press. pp. 421.
  2. 2. Schluter D (2001) Ecology and the origin of species. Trends Ecol Evol 16: 372–380.
  3. 3. Dres M, Mallet J (2002) Host races in plant-feeding insects and their importance in sympatric speciation. . Phil. Trans. R. Soc. Lond. B 357: 471–492.
  4. 4. Dieckmann U, Doebeli M, Metz JAJ (2004) Adaptive speciation. Cambridge University Press, Cambridge. Pp. 488.
  5. 5. Funk DJ, Nosil P, Etges B (2006) Ecological divergence exhibits consistently positive associations with reproductive isolation across disparate taxa. Proc Natl Acad Sci USA 103: 3209–3213.
  6. 6. Funk DJ, Nosil P (2008) Comparative analyses and the study of ecological speciation in herbivorous insects. In: Tilmon K editor. Specialization, Speciation, and Radiation: The Evolutionary Biology of Herbivorous Insects. University of California Press. pp. 117–135.
  7. 7. Rova E, Björklund M (2011) Can preference for oviposition sites initiate reproductive isolation in Callosobruchus maculatus? PLoS ONE 6(1): e14628
  8. 8. Agrawal AA (2011) Current trends in evolutionary ecology of plant defences. Funct Ecol 25: 420–432.
  9. 9. Kircher HW (1982) Chemical composition of cacti and its relationship to Sonoran desert Drosophila. In: Barker JSF, Starmer WT editors. Ecological Genetics and Evolution. Academic Press. Sydney, Australia. pp. 143−158.
  10. 10. Fogleman JC, Abril JR (1990) Ecological and evolutionary importance of host plant chemistry. In: Barker JSF, Starmer WT, MacIntyre RJ editors. Ecological and evolutionary genetics of Drosophila. Plenum Press, New York. pp. 121–143.
  11. 11. Via S (1990) Ecological genetics and host adaptation in herbivorous insects: The experimental study of evolution in natural and agricultural systems. Annu Rev Entomol 35: 421–446.
  12. 12. Etges WJ (1990) Direction of life history evolution in Drosophila mojavensis. In: Barker JSF, Starmer WT, MacIntyre RJ editors. Ecological and evolutionary genetics of Drosophila. Plenum Press, New York. pp. 121–143.
  13. 13. Mitter C, Futuyma DJ (1983) An evolutionary-genetic view of host-plant utilization by insects. In: Denno RF, Mcclure MS editors. Variable plants and herbivores in natural and managed systems. Academic Press, New York. pp. 427–459.
  14. 14. Jaenike J (1990) Host specialization by phytofagous insects. . Annu Rev Ecol Syst. 21: 243–273.
  15. 15. Fanara JJ, Hasson E (2001) Oviposition acceptance and fecundity schedule in the cactophilic sibling species Drosophila buzzatii and D. koepferae on their natural hosts. Evolution 55: 2615–2619.
  16. 16. Jaureguy LM, Etges WJ (2007) Assessing patterns of senescence in Drosophila mojavensis reared on different host cacti. . Evol Ecol Res. 9: 91–107.
  17. 17. Dambroski HR, Linn C, Berlocher S, Forbes AA, Roelofs W, et al. (2005) The genetic basis for fruit odor discrimination in Rhagoletis flies and its significance for sympatric host shifts. Evolution 59: 1953–1964.
  18. 18. McBride CS (2007) Rapid evolution of smell and taste receptor genes during host specialization in Drosophila sechelia. . Proc Natl Acad Sci USA. 104: 4996–5001.
  19. 19. Hawthorne DJ, Via S (2001) Genetic linkage of ecological specialization and reproductive isolation in pea aphids. Nature 412: 904–907.
  20. 20. Jones CD (1998) The genetic basis of Drosophila sechellia's resistance to a host plant toxin. Genetics 149: 1899–1908.
  21. 21. Jones CD (2004) Genetics of egg production in Drosophila sechellia. Heredity 92: 235–241.
  22. 22. Markow TA, O’Grady P (2008) Reproductive ecology of Drosophila. . Funct Ecol. 22: 747–759.
  23. 23. Powell JR (1997) Progress and prospects in evolutionary biology: the Drosophila model. Oxford Univ. Press, New York. Pp. 576.
  24. 24. Wasserman M (1982) Evolution of the repleta group. In: Ashburner M, Carson HL, Thompson JN editors. The Genetics and Biology of Drosophila. Academic Press, London & New York. pp. 61−139.
  25. 25. Throckmorton LH (1982) The virilis species group. In Ashburner M, Thompson JN, Carson HL editors. The Genetics and Biology of Drosophila. Academic Press. Pp. 227−296.
  26. 26. Fogleman JC, Danielson PB (2001) Chemical Interactions in the Cactus-Microorganism-Drosophila Model System of the Sonoran Desert. . Am Zool. 41: 877–889.
  27. 27. Fogleman JC, Heed WB (1989) Columnar cacti and desert Drosophila: the chemistry of host plant specificity. In: Interactions among plants and animals in the western deserts. Schmidt JO editor. University of New Mexico Press, Albuquerque. pp. 1− 24.
  28. 28. Heed WB, Mangan RL (1986) Community ecology of the Sonoran Desert Drosophila. In: Ashburner M, Carson HL, Thompson Jr JN editors. The Genetics and Biology of Drosophila. Academic Press, New York. pp. 311−345.
  29. 29. Etges WJ, Veenstra CL, Jackson LL (2006) Premating isolation is determined by larval rearing substrates in cactophilic Drosophila mojavensis. VII. Effects of larval dietary fatty acids on adult epicuticular hydrocarbons. J Chem Ecol. 32: 2629–2646.
  30. 30. Heed WB, Kircher HW (1965) Unique sterol in the ecology and nutrition of Drosophila pachea. Science 149: 758–761.
  31. 31. Lang M, Murat S, Clark AG, Gouppil G, Blais C, et al. (2012) Mutations in the neverland Gene Turned Drosophila pachea into an Obligate Specialist Species. Science 337: 1658–1661.
  32. 32. Oliveira DCSG, Almeida FC, O'Grady PM, Armella MA, DeSalle R, et al. (2012) Monophyly, divergence times, and evolution of host plant use inferred from a revised phylogeny of the Drosophila repleta species group. . Mol Phylogen Evol. 64: 533–544
  33. 33. Manfrin MH, Sene FM (2006) Cactophilic Drosophila in South America: a model for evolutionary studies. Genetica 126: 57–75.
  34. 34. Hasson E, Soto IM, Carreira VP, Corio C, Soto EM, et al.. (2009) Host plants, fitness and developmental instability in a guild of cactophilic species of the genus Drosophila. In: Santos EB editor. Ecotoxicology Research Developments. Nova Science Publisher, Inc., Hauppauge, Nueva York. pp. 89−109.
  35. 35. Hasson E, Naveira H, Fontdevila A (1992) The breeding sites of the Argentinian species of the Drosophila mulleri complex (subgenus Drosophila repleta group). Rev Chil Hist Nat. 65: 319–326.
  36. 36. Fanara JJ, Fontdevila A, Hasson E (1999) Oviposition preference and life history traits in cactophilic Drosophila koepferae and D. buzzatii in association with their natural hosts. . Evol Ecol. 13: 173–190.
  37. 37. Carreira VP, Soto IM, Hasson E, Fanara JJ (2006) Patterns of variation in wing morphology in the cactophilic Drosophila buzzatii and its sibling D. koepferae. J. . Evol Biol. 19: 1275–1282.
  38. 38. Soto IM, Carreira VP, Fanara JJ, Hasson E (2007) Evolution of male genitalia: environmental and genetic factors affect genital morphology in two Drosophila sibling species and their hybrids. . BMC Evol Biol. 7: 77.
  39. 39. Soto IM, Carreira VP, Soto EM, Hasson E (2008) Wing morphology and fluctuating asymmetry depend on the host plant in cactophilic Drosophila. . J Evol Biol. 21: 598–609.
  40. 40. Soto EM, Goenaga J, Hurtado J, Hasson E (2012) Oviposition and performance in natural hosts in cactophilic Drosophila. Evol Ecol 26: 975–990.
  41. 41. Fanara JJ, Folguera G, Iriarte P, Mensch J, Hasson E (2006) Genotype by environment interactions in viability and developmental time in populations of cactophilic Drosophila. . J Evol Biol. 19: 900–908.
  42. 42. Stintzing FC, Carle R (2005) Cactus stems (Opuntia spp.): a review on their chemistry, technology, and uses. Molecular nutrition & food research 49: 175–194.
  43. 43. Reti L, Castrillón JA (1951) Cactus Alkaloids. I. Trichocereus terscheckii (Parmentier) Britton and Rose. . J Am Chem Soc. 73: 1767–1769.
  44. 44. Fernández Iriarte PF, Hasson E (2000) The role of the use of different host plants in the maintenance of the inversion polymorphism in the cactophilic Drosophila buzzatii. . Evolution. 54: 1295–1302.
  45. 45. Corio C, Soto IM, Carreira VP, Padró J, Betti MIL, et al. (2013) An alkaloid fraction extracted from the cactus Trichocereus terschekii affects fitness components in the cactophilic fly Drosophila buzzatii. . Biol J Linn Soc Lond. 109 (2): 342–353.
  46. 46. Vilela CA (1983) A revision of the Drosophila species group (Diptera-Drosophilidae). . Rev Brasil Entomol. 27: 1–114.
  47. 47. Wink M (1988) Plant breeding importance of plant secondary metabolites for protection against pathogens and herbivores. Theor Appl Genetics 75: 225–233.
  48. 48. Wink M (2003) Evolution of secondary metabolites from an ecological and molecular phylogenetic perspective. Phytochemistry 64: 3–19.
  49. 49. Ogunbodede O, McCombs D, Trout K, Daley P, Terry M (2010) New mescaline concentrations from 14 taxa/cultivars of Echinopsis spp. (Cactaceae) (“SanPedro”) and their relevance to shamanic practice. . J Ethnopharmacol. 131: 356–362.
  50. 50. Meyer JM, Fogleman JC (1987) Significance of saguaro cactus alkaloids in ecology of Drosophila mettleri, a soil-breeding, cactophilic drosophilid. J Chem Ecol. 13: 2069–2081.
  51. 51. Padró J, Soto IM (2013) Exploration of the nutritional profile of Trichocereus terscheckii (Parmentier) Britton & Rose stems. . J. Prof. Assoc. Cactus Develop. 15: 1–12.
  52. 52. Pennanec'h M, Bricard L, Kunesch G, Jallon JM (1997) Incorporation of fatty acids into cuticular hydrocarbons of male and female Drosophila melanogaster.. J Insect Physiol 43: 1111–1116.
  53. 53. Ray AM, Millar JG, McElfresh JS, Swift IP, Barbour JD, et al. (2009) Male-produced aggregation pheromone of the cerambycid beetle Rosalia funebris. . J Chem Ecol. 35: 96–103.
  54. 54. Rohlf JF (2003) Morphometrics. TpsDig versión 1.38 Available: Accessed 2013 Dec. 12. Department of Ecology and Evolution, State University of New York.
  55. 55. Dryden IL, Mardia KV (1998) Statistical Shape Analysis, John Wiley, Chichester. Pp. 347.
  56. 56. Ruiz A, Heed WB (1988) Host-plant specificity in the cactophilic Drosophila mulleri species complex. . J Anim Ecol. 57 237–249.
  57. 57. Krebs RA, Barker JSF (1991) Coexistence of ecologically similar colonizing species. 1. Intraspecific and interspecific competition in Drosophila aldrichi and D. buzzatii. . Aust J Zool. 39: 579–593.
  58. 58. Cortese M, Norry FM, Piccinali R, Hasson E (2002) Direct and correlated responses to selection on wing length and developmental time in Drosophila buzzatii. Evolution 56: 2541– 2547.
  59. 59. Roff DA (2000) Trade-offs between growth and reproduction: an analysis of the quantitative genetic evidence. J. Evol. Biol. 13: 434–445.
  60. 60. Sgro CM, Hoffmann AA (2004) Genetic correlations, tradeoffs and environmental variation. Heredity 93: 241–248.
  61. 61. Efron B, Tibshirani R J (1993) An introduction to the bootstrap. Chapman and Hall, New York. Pp. 436.
  62. 62. Tabachnick BG, Fidell LS (2007) Using Multivariate Statistics (5th ed.). New York: Allyn and Bacon. Pp. 1008.
  63. 63. StatSoft (2001) Inc. STATISTICA (data analysis software system), version 6.0. Available:
  64. 64. Hood GM (2008) PopTools, version 3.0.3. Available: Accessed 2014 Jan 10.
  65. 65. Kircher HW, Heed WB, Russell JS, Grove J (1967) Senita cactus alkaloids: their significance to Sonoran Desert Drosophila ecology. . J Insect Physiol. 13: 1869–1874.
  66. 66. Fogleman JC, Hackbarth K R, Heed WB (1981) Behavioral differentiation between two species of cactophilic Drosophila. . III. Oviposition site preference. Am Nat. 118: 541–548.
  67. 67. Woods RE, Sgro CM, Hercus MJ, Hoffmann AA (1999) The association between fluctuating asymmetry, trait variability, trait heritability, and stress: a multiply replicated experiment on combined stresses in Drosophila melanogaster. Evolution 53: 493–505.
  68. 68. Soto EM, Soto IM, Carreira VP, Fanara JJ, Hasson E (2008) Host-related life history traits in interspecific hybrids of cactophilic Drosophila. Entomol Exp Appl. 126: 18–27.
  69. 69. Hurtado J, Soto EM, Orellana ML, Hasson E (2012) Mating success depends on rearing substrate in cactophilic Drosophila. Evol Ecol. 26: 733–743.
  70. 70. Schreiber K (1958) Uber einige Inhaltsstoffe der Solanaceen und ihre Bedeutung fur die Kartoffelkaferresistenz. . Ent Exp Appl. 1: 28–37.
  71. 71. Harley K L, Thorsteinson AJ (1967) The influence of plant chemicals on the feeding behavior, development and survival of the two-striped grasshopper. Melanoplus bivittatus (Say). Acrididae: Orthoptera. . Can J Zool. 45: 315319.
  72. 72. Narberhaus I, Zintgraf V, Dobler S (2005) Pyrrolizidine alkaloids on three trophic levels – evidence for toxic and deterrent effects on phytophages and predators. Chemoecology. 15: 121–125.
  73. 73. Morales-Hojas R, Vieira J (2012) Phylogenetic patterns of geographical and ecological diversification in the subgenus Drosophila. PLoS One 7: e49552.
  74. 74. Nyffeler R (2002) Phylogenetic relationships in the cactus family (Cactaceae) based on evidence from trnK/ matK and trnL-trnFsequences. . Am J Bot. 89: 312–326.
  75. 75. Fontdevila A, Pla C, Hasson E (1988) Drosophila koepferae: a new member of the Drosophila serido (Diptera-Drosophilidae) superspecies taxon. . Ann Entomol Soc Am. 81: 380–385.
  76. 76. Carreira VP, Padró J, Mongiardino Koch N, Fontanarrosa P, Alonso JI, et al. (2014) Nutritional composition of Opuntia sulphurea (G. Don in Loudon) cladodes. Haseltonia 19: 38–45.
  77. 77. Bush GL (1969) Sympatric host race formation and speciation in frugivorous flies of the genus Rhagoletis (Diptera: Tephritidae). Evolution 23: 237–251.
  78. 78. R’Kha S, Capy P, David JR (1991) Host-plant specialization in the Drosophila melanogaster species complex: a physiological, behavioral, and genetical analysis. Proc Natl Acad Sci USA 88: 1835–1839.
  79. 79. Bush GL (1975) Sympatric speciation in phytophagous parasitic insects. In: Price PW editor. Evolutionary strategies of parasitic insects and mites. Plenum Press, New York. pp. 187–206.
  80. 80. Thomas CD, Ng D, Singer MC, Mallet JLB, Parmesan C, et al. (1987) Incorporation of a European weed into the diet of a North American herbivore. Evolution 41: 892–901.
  81. 81. Starmer WT, Lachance M, Phaff HJ, Heed WB (1990) The biogeography of yeast associated with decaying cactus tissue in North America, the Caribean, and Northern Venezuela. . Evol Biol. 24: 115–190.