Euwallacea fornicatus [Polyphagous Shot hole Borer (PSHB)], native to Southeast Asia (Li et al. 2016), is a recent invader in South Africa (Paap et al. 2018). It vectors a fungal mutualist, Fusarium euwallaceae, that enables beetle colony establishment in hosts. Fusarium euwallaceae colonizes xylem tissues surrounding PSHB galleries. PSHB feeds on the fungus that grow within galleries (Eskalen et al. 2013). Highly susceptible hosts support beetle reproduction, but F. euwallaceae may infect hosts without beetle colony establishment through indiscriminate host probing by PSHB (Eskalen et al. 2013). PSHB colonization results in Fusarium dieback, reported from numerous tree species in South Africa, the U.S.A., Israel, Europe and Australia (Eskalen et al. 2013; Mendel et al. 2017; IPPC, 2021; Lynch et al., 2021; Van Rooyen et al. 2021). Despite common use of the term ‘Fusarium dieback’, experiments demonstrating F. euwallaceae pathogenicity are scant and, in commercial crops, are restricted to Avocado (Jones and Paine 2017), Macadamia (Twiddy et al. 2021), and Almond (Moreno et al. 2018).

Apple production in South Africa is valued at ca. $700 million (Hortgro 2020) while the South African grapevine industry is worth nearly $3,5 billion (www.sawis.co.za). PSHB infestations on grapevine (Vitis vinifera) and apple (Malus domestica) was recently observed in gardens in the Western Cape province (Van Rooyen et al. 2021) (Figs. 1–2). Host lists from the U.S.A. include grapevine and Malus spp. as susceptible to F. euwallaceae (Eskalen et al. 2013), but Koch’s postulates have not been satisfied in any previous reports. This study determines the host status of apple and grapevine to F. euwallaceae.

Fig. 1
figure 1

Grapevine wound after attempted colonization by the Polyphagous shot hole borer

Full size image
Fig. 2
figure 2

Fusarium euwallaceae -stained wood surrounding a gallery of the Polyphagous shot hole borer in apple

Full size image

Isolates were obtained from grapevine (CMW-IA2, George, August 2019), apple (CMW-IA3, George, September 2018) and Cherry (CMW-IA4, George, August 2019) by exposing PSHB galleries in wood, excising pieces (3 mm3) of fungus-stained wood (Fig. 2) and placing these onto Potato Dextrose Agar plates (PDA, Biolab, South Africa). Plates were incubated for 7 d at 24 °C. Pure cultures were obtained by transferal of single hyphal tips to fresh PDA plates. Isolate identity was confirmed by DNA sequence comparisons of the elongation factor 1-α region to that of the ex-holotype strain of F. euwallaceae (GenBank JQ038007). Genomic DNA was extracted using a Sigma-Aldrich™ plant extraction kit, and the target region was amplified using the primer pair EF1 and EF2 and PCR thermocycling conditions following Paap et al. (2018). Sequences (GenBank OM963006—OM963008) were identical to the F. euwallaceae type. Reference isolates are kept at the Culture Collection of Innovation Africa, South Africa.

Apple cv. ‘Pink lady’ and grapevine (cvs. ‘Merlot’ and ‘Cabernet Sauvignon’) on commercial farms [Lourensford estate (-33°35′36″S, 18°55′53″E) and Nietvoorbij (33°54′05″S 18°52′34″E), respectively] were inoculated with the three F. euwallaceae isolates in early Autumn (March 2021). Inoculum was prepared by saturating sterile wooden toothpicks in half-strength PDA, placing these onto PDA plates and inoculating plates with F. euwallaceae. Plates were incubated for 21 d until toothpicks were covered with mycelia. F. euwallaceae -overgrown toothpicks were used as inoculum and sterile PDA-saturated toothpicks served as controls.

Inoculations in apple were carried out by drilling five 2.5 mm wide and 1 cm deep holes, 5 cm apart, in a spiral-like manner around a branch (ca. 2 cm diam., one branch per tree). Fungus-laden toothpicks were inserted into these holes, with all holes on a tree receiving the same isolate. The experiment was replicated 10 times (n = 40 trees). Grapevine was inoculated by inoculating a single hole in the trunk with the fungus or control (ca. 5 cm diam.), at 10 cm from the head. Forty individual grapevine trunks were inoculated per isolate and control for each cultivar (n = 160 vines per cultivar). Inoculation wounds were sealed with Parafilm™.

Apple branches were harvested 9 weeks post-inoculation and ten replicate grape vines per cultivar, chosen at random, were harvested after 1-, 3-, 6-, and 12-months post-inoculation. Branches and trunks were cut longitudinally through inoculation points to measure the length of the resulting lesion in the xylem tissues to assess the growth of F. euwallaceae (Fig. 3). Re-isolations were made to confirm F. euwallaceae as the causal organism.

Fig. 3
figure 3

Fusarium euwallaceae -stained wood around the inoculation point in apple (left bottom) and grapevine (right bottom). Controls shown at top. Scale bars 1 cm

Full size image

Lesion length data were analyzed using R software (version 3.6.3; https://www.rstudio.com). Data from both hosts were non-normally distributed (Shapiro–Wilk test, p > 0.05). Data for grapevine was compared between treatments using Kruskal–Wallis tests. Data for apple was analyzed using a generalized linear mixed-effects model (lme4 package, Bates et al. 2014). Data fitted a gamma distribution and tree individual was used as random variable. Significant main effects were separated using a Tukey test (multcomp package, Hothorn et al. 2008).

In grapevine, lesions around inoculation points after one-month penetrated ca. 2 mm into the xylem (Fig. 3). Fusarium euwallaceae could be recovered from three inoculated Cabernet Sauvignon and four Merlot plants, but not from controls. There was no significant difference in lesion length between the isolates and the controls [Merlot (H = 3.1698, df = 3, p = 0.3662), Cabernet Sauvignon (H = 3.181, df = 3, p = 0.3646)]. No further lesion development was evident after 3-, 6- and 12-months post inoculation. No isolates of F. euwallaceae could be recovered after three months. The model for apple lesion length data had an AIC value of 797 (deviance = 785, DF = 182). Treatment had a significant effect on lesion length (X2 = 50.34; p < 0.001). Isolates caused similar lesion lengths, but these were significantly longer than in controls (Fig. 4). F. euwallaceae was consistently reisolated from inoculated trees but not controls.

Fig. 4
figure 4

Lesion length of different isolates of F. euwallaceae inoculated into apple branches. Boxes indicate 25–75% data range, whiskers indicate 1.5 × the interquartile range

Full size image

We conclude that F. euwallaceae is not pathogenic to grapevine and does not persist internally longer than three months. F. euwallaceae is confirmed as pathogenic to apple. This, coupled with recent observations of PSHB breeding in a mature apple tree in a garden (FR, pers. observ.), indicates a risk of Fusarium dieback in apple. Currently PSHB is not known from commercial apple orchards, but these should be continuously monitored.