The fungal genus Venturia Sacc. (anamorph Fusicladium Bonord.) includes plant pathogens that cause substantial economic damage to fruit crops worldwide (Sivanesan, 1977; Schubert et al., 2003). Although, Venturia inaequalis (Cooke) G. Winter is considered a model species in plant pathology (Machardy, 1996) and it is the causal agent of apple scab, the most important apple disease worldwide, other Venturia spp. also cause scab on other fruit trees. In this review, the main characteristics of plant-pathogenic species of Venturia are discussed with special attention to V. inaequalis affecting apple (Malus spp.); V. pyrina and V. nashicola affecting European pear (Pyrus communis) and Asian pear (P. pyrifolia var. culta and P. ussuriensis), respectively; V. carpophila affecting peach (Prunus domestica) and almond (Prunus dulcis); Fusicladium oleagineum affecting olive (Olea europea); F. effusum affecting pecan (Carya illinoinensis); and F. eriobotryae affecting loquat (Eriobotrya japonica). Aspects of V. asperata affecting apple and V. cerasi affecting cherry (Prunus cerasus) are also discussed (Table 1).

Since the late nineteenth century, apple scab has been extensively investigated, and substantial information—covering all key aspects of the biology and genetics of the fungus and the epidemiology and control of the disease—has been published and reviewed by Machardy (1996) and Bowen et al. (2011). In contrast to the efforts devoted to investigating V. inaequalis, little work has been conducted on Venturia spp. affecting other fruit trees, as shown by the number of papers published for each species (Figure 1A). This difference in research effort and number of publications, however, does not directly reflect the importance of the host crops worldwide (Figure 1B). The difference might be explained by (i) minor investments in these non-apple crops, (ii) less specialized management directed at the non-apple crops, and (iii) the common use of the information developed for V. inaequalis for managing the other fruit scabs. Concerning the last point, researchers generally assume that infection of any scab fungus may occur under environmental conditions similar to those required by V. inaequalis. The Mills and Laplante's (1954) table, which is the most popular system for scheduling fungicides against apple scab, has been broadly recommended for management of pear scab (Sobreiro and Mexia, 2000; Mitcham and Elkins, 2007; Travis et al., 2012; Elkins et al., 2016), cherry scab (Schweizer, 1958), peach scab (Keitt, 1917; Pineau et al., 1991), and loquat scab (Ramos, 2008; GVA, 2014). However, there is no clear evidence that the environmental conditions conducive for infection are similar for all of these Venturia species. In fact, recent studies have revealed important differences concerning the environmental requirements for infection by F. eriobotryae and F. oleagineum vs. V. inaequalis (Viruega et al., 2011; González-Domínguez et al., 2013). In addition, substantial differences exist in the ecophysiologies and the life cycles of their hosts.

This review has two main objectives: (i) to identify the main gaps in our knowledge regarding the biology and epidemiology of Venturia spp. affecting fruit trees; and (ii) to identify similarities and differences between these Venturia spp. in order to improve disease management. To accomplish these objectives, the authors have thoroughly reviewed the studies regarding the phylogenetic relationships, host ranges, biologies, and epidemiologies of Venturia spp. Finally, the review discusses the implications of the similarities and differences in the fungi and the diseases for the management of the different scab diseases.

The genus Venturia belongs to the phylum Ascomycota, class Dothideomycetes (Schoch et al., 2009). Traditionally, this genus has been included in the family Venturiaceae, order Pleosporales, according to its “Pleospora-type centrum and bitunicate asci” (Sivanesan, 1977). However, recent molecular phylogenetic analyses of Dothideomycetes, using both nuclear and mitochondrial gene regions, have indicated that the family Venturiaceae forms a well-supported monophyletic group separate from the Pleosporales (Kodsueb et al., 2006; Kruys et al., 2006; Zhang et al., 2011). Thus, Zhang et al. (2011) recently reordered Venturiaceae into Venturiales ord. nov. (together with Sympoventuriaceae fam. nov.).

The genus Venturia Sacc. (1882) was first noted in 1844 by Notaris, who described, V. dianthi and V. rosae. Cesati & Notaris added new species in 1863 (Sivanesan, 1977). Saccardo reevaluated the genus in 1882, excluding both V. dianthi and V. rosae (Sivanesan, 1977). Sivanesan (1977) listed 52 species of Venturia, which comprised parasitic species with pseudothecia immersed in the host tissue, bitunicate asci, and olive-brown, septate ascospores (Sivanesan, 1977). Currently, 290 species are recognized in the Mycobank dabatase (http://www.mycobank.org/Biolomics.aspx?Table=Mycobankx, March/2017).

The anamorphs of Venturia spp. have been traditionally classified in three genera: Fusicladium Bonord., Pollaccia Baldacci & Cif., and Spilocaea Fr. Assignment to these genera depends on the morphology of the conidiogenous cells. These cells are sympodial in Fusicladium and percurrent in Pollaccia and Spilocaea (Hughes, 1953; Sivanesan, 1977). Recents works that used both morphological and molecular characters concluded that the anamorphic species of Venturia should not be separated into these three genera because (i) most species have both sympodial and percurrent conidiogenous cells, and (ii) molecular phylogenetic analysis clearly shows that Venturia and its anamorphs are monophyletic group (Schubert et al., 2003; Beck et al., 2005). Because most anamorphs of Venturia have been classified as Fusicladium, this name was proposed to designate the asexual stage of Venturia spp. (Braun et al., 2002), and used in the monograph written by Schubert et al. (2003). Very recently, the International Commission on the Taxonomy of Fungi has proposed the use of Venturia instead of Fusicladium for the species with only anamorph stage, following the guidelines of the “Amsterdam Declaration on Fungal Nomenclature” (May, 2017). In this paper we keep the dual nomenclature because it facilitates to distinguish the species with or without sexual phase.

In 2003, Partridge and Morgan-Jones (2003) proposed the new genus Fusicladosporium, including the anamorphs of pecan (F. effusum), peach (F. carpophillum), and maple scab (F. humile; teleomorph V. acerina). The authors considered that significant morphological differences (conidia formed in chains and the prominence of conidial scars on conidiophores) distinguished these anamorphs from those in the genera Cladosporium and Fusicladium. However, this new taxon seems unjustified because two older generic names for Venturia anamorphs with catenate conidia are available (Hormocladium Höhn. and Ramalia Bat) and because phylogenetic analysis of the ITS region demonstrates that the erection of Fusicladosporium results in a polyphyletic genus (Schubert et al., 2003; Beck et al., 2005). Thus, Fusicladosporium is currently considered a synonym of Fusicladium (Schubert et al., 2003; Crous et al., 2007; Scherm et al., 2008; Seyran et al., 2009; Lalancette et al., 2012).

As part of the current review, a multiple correspondence analysis (CA) was performed on epidemiological components of seven Venturia spp. CA is a multivariate statistical method that makes it possible to represent contingency tables in a pictorial form (Savary et al., 1995). CA is based on a raw data matrix, in which the rows are the objects and the columns are the variables. In this case, the objects are the seven Venturia spp.—V. carpophila, V. inaequalis, V. pyrina, V. nashicola, F. effusum, F. eriobotryae, and F. oleagineum—for which epidemiological information exists for seven qualitative variables and eight quantitative variables (Table 4).

In our case, qualitative variables (e.g., the known presence of the sexual stage in nature) were classified using Yes or No; quantitative variables (e.g., optimal temperature for conidial infection) were ranked as high, medium, or low based on the average ± standard deviation (SD) of each data set. For example, the maximum temperature for conidial infection (TM) of the seven Venturia spp. ranged from 25 to 37°C, with an average of 31°C and a SD of 4. Thus, Venturia spp. in which the maximum temperature for infection was TM ≤ 27°C (i.e., 31–4) were classified as low, those with 27 > TM < 35°C were classified as medium, and those with TM ≥ 35°C were classified as high (Table 4).

The data matrix of Table 4 was used to perform CA with two dimensions (D1 and D2) using the multiple correspondence analysis procedure of SPSS (ver. 23; SPSS Inc.). Three analyses were performed: (i) with all components (CA1); (ii) with all components except those concerning the distribution in different climate types (CA2); and (iii) with only those components concerning the asexual stage (CA3). These analyses accounted for 80.0, 88.5, and 98.2% of data variance, respectively (Table 5).

Overall, these analyses separated the Venturia spp. into two main groups when the seven species were plotted on the D1-D2 space (Figure 6). The first group contained V. inaequalis, V. pyrina, V. nashicola, and V. carpophila, and the second group contained F. oleagineum and F. eriobotryae, with F. effusum having an intermediate position depending on the CA analysis.

In CA1 (Figure 6A), grouping of the species was mainly based on the presence/absence of the sexual stage, infection of deciduous vs. non-deciduous trees, and the thermal and moisture requirements for ascosporic infection. These variables had the most influence (i.e., had high discrimination values, DVs) in D1, which accounted for 53.3% of the variance in the dataset (Table 5); the DV was >0.829 for these variables (Table 6). In CA1, V. nashicola was far from the other species in D2 (Figure 6A), mainly because of its different distribution among climate types (presence/absence of the species in tropical and subtropical climates had DVs = 0.514 and 0.637, respectively; Table 6); the presence/absence of the species in tropical and subtropical climates together with wind dispersal of conidia were the most influential variables in D2 (Table 6). This result may be biased by the distribution of V. nashicola being restricted to China, Japan, South Korea, and Taiwan, a restriction that may be caused more by quarantine measures by other countries than by differences in environmental requirements. This inference is supported by CA2, in which the pathogen distribution among climates was not considered. In CA2 (Figure 6B), the species were grouped mainly based on the presence/absence of the sexual stage and environmental requirements for ascosporic infection in D1 (Table 6), which accounted for 63.6% of the variance (Table 5). In CA3 (Figure 6C), the grouping was mainly determined by moisture conditions for conidial germination, infection, and dispersal in D1 (Table 6), which accounted for 60.9% of the variance (Table 5). The conidia of V. inaequalis, V. pyrina, V. nashicola, and V. carphophila are similar in that they require only a few hours of wetness (6–12 h) to infect and are capable of being dispersed by wind if dislodged by rain or dew. Fusicladium oleagineum and F. eriobotryae require longer periods of wetness to infect (>18 h), have lower maximum temperatures for infection (25–28°C), and have conidia that germinate only in the presence of free water. In CA3, F. oleagineum and F. eriobotryae had similar scores in D1, but they were far apart in D2 (Figure 6C) because of different temperature requirements for conidial infection (Tables 4, 6). Both F. effusum and the Venturia group can infect with <12 h of wetness and have wind-dispersed conidia that do not require free water to germinate; however, the temperature requirements for conidial infection differ between F. effusum and the Venturia group.

The presence of two main groups of species probably reflects pathogen adaptation to host ecophysiology. This hypothesis is supported by the monophyly of the genus Venturia (Ishii and Yanase, 2000; Beck et al., 2005; Gladieux et al., 2010a; Bowen et al., 2011). F. eriobotryae and F. oleagineum are both pathogens of Mediterranean plants (loquat and olive, respectively). They are adapted to a warm and dry climate in which the low annual rainfall is distributed mainly in autumn and spring (Csa climate class; Graniti, 1990; Kottek et al., 2006). The absence of the teleomorph in nature may be related to the mild winter temperatures, which can be survived without a quiescent stage. In these fungi, dispersal of conidia occurs only during rain events, perhaps because conidia dispersed in water have a higher probability of germinating and causing infection. The requirement of free water for conidial germination and long periods of wetness for conidial infection may also be adaptations to a dry climate. These requirements would prevent the initiation of an infection cycle in the driest periods of the year. F. eriobotryae and F. oleagineum also have low mycelium growth rates and long latent periods. The evergreen habitus of their hosts means that these species do not require short infection cycles, because the trees are susceptible throughout the year and because the inoculum may survive on the tree during the season in which environmental conditions are not suitable for sporulation and infection.

Venturia spp. that attack deciduous trees, in contrast, require a sexual stage to survive the winter, when there is no host tissue to be infected and temperatures are low. Their conidia can germinate in the absence of free water, and infection requires fewer hours of wetness, especially in the case of V. inaequalis and V. nashicola. Together, these factors result in relatively short infection cycles, with a higher probability of occurrence. In this case, the deciduous habitus of the host makes the occurrence of infection obligatory, because the main inoculum source for the next season will be the fallen, scabbed leaves, even though overwintering in twig lesions and/or buds is possible.

This review has considered several important aspects of the phylogeny, host range, and life cycle of Venturia spp. affecting fruit trees. These species are responsible for some of the most important diseases of their hosts.

Venturia spp. affecting fruit trees are highly host-specific, as indicated by the general failure to obtain infection by cross-inoculation (Menon, 1956; Raabe and Gardner, 1972; Ishii and Yanase, 2000; Stehmann et al., 2001; Le Cam et al., 2002; Chevalier et al., 2004; Sánchez-Torres et al., 2007a, 2009; Abe et al., 2008). This conclusion partially contradicts some previous reviews of the genus Venturia (Sivanesan, 1977) and Fusicladium (Schubert et al., 2003), and has implications for the management of scab diseases in areas where different host species are grown, as is the case, for instance, in the Emilia-Romagna region of North Italy where apple, pear, cherries, and peaches are grown in close proximity. The inoculum produced in one crop cannot infect another crop, and management of a species of Venturia in one orchard has no effect on neighboring crops of other host species.

Although Venturia is one of the most famous and studied genera of plant pathogens, important gaps in understanding the life cycle still remain for some species. This is particularly unexpected for V. pyrina because of the worldwide importance of the crop and because the gaps involve key aspects of the life cycle, including pseudothecia formation, ascospore and conidia germination, and mycelial growth. The only work regarding dispersal of V. pyrina conidia was published in 1937 (Kienholz and Childs, 1937). In most cases, researchers have assumed that V. pyrina requirements are similar to those of V. inaequalis. This assumption, however, is not valid for ascosporic infection because ascospores of V. pyrina require more wet hours to infect (Figure 4). In the case of V. carpophila, specific experiments on the requirements for infection have never been performed, and this limits our ability to correctly manage the disease.

Venturia spp. can infect several parts of the host trees, but the main damage usually results from fruit infection, except for F. oleagineum affecting olive leaves, which can result in important economic losses. For this reason, fruit growers are generally risk-adverse to scab diseases and schedule a high number of fungicide applications to achieve a very high level of disease control. Machardy (1996) reported that USA apple growers schedule fungicide applications to achieve <1% of scabbed fruit at harvest. Similar thresholds are usually assumed for loquat, a high value fruit crop in Europe (E. Soler, personal communication). Current trends in disease management aim to avoid this high number of treatments, which involve risks to human health and the environment, and encourage the use of decision support systems (DSSs) (Rossi et al., 2012).

The Mills tables represent one of the first and better known DSSs and have been widely accepted by growers and advisors (Machardy, 1996). However, the use of these tables should be avoided for Venturia spp. other than V. inaequalis. For V. pyrina, F. oleagineum, F. effusum, and F. eriobotryae the Mills tables over-predict the number of scab infections because temperature and wetness requirements of these fungi are different from those of V. inaequalis (Figure 6; Gottwald, 1985; Villalta et al., 2000; Viruega et al., 2011; González-Domínguez et al., 2013). Moreover, the Mills tables can over-predict the number of infections even for apple scab (Machardy and Gadoury, 1989).

For some of the diseases considered in this review, epidemiological models have been developed to predict disease development (Table 7). Most of these models are simple and consider only one component of the pathogen life cycle, mainly ascospore maturation or infection. A main constraint of these models is that they have never been validated against independent data, i.e., model output (the prediction) has not been compared with a data set of real-world observations different from that used for model development (Rossi et al., 2010). Before these models are used in practical disease control, a robust validation with real data should be performed; the validation data should be obtained from different areas with different epidemiological conditions and for several years (Rossi et al., 2010).

In the case of V. inaequalis, mechanistic weather driven models have been developed for primary infections (Rossi et al., 2007) and for the whole life cycle (Xu et al., 1995). A mechanistic approach has also been used to develop epidemiological models for V. nashicola and F. eriobotryae, and these models consider most of the components of the life cycle (Li et al., 2007; González-Domínguez et al., 2014a). Several advantages have been previously reported for mechanistic vs. empirical models (Caffi et al., 2007; De Wolf and Isard, 2007; Rossi et al., 2010). Mechanistic models attempt to capture the full complexity of the pathogen life cycle and are generally considered to have greater explanatory ability than purely empirical models (De Wolf and Isard, 2007). A main advantage of the mechanistic models is that they can easily incorporate information from previous experiments regarding pathogen biology and epidemiology. Thus, this review should help researchers develop mechanistic models for those scab diseases that currently lack such models. Until such mechanistic models are available, however, the Magarey curves, developed in this review and which identify the requirements for infection, could be used as a starting point to predict infection risk, especially for V. pyrina and F. effusum.

All the authors contributed to the writing of the manuscript. EG and VR performed the data analyses.