The microsporidium Kneallhazia solenopsae (Knell, Allen & Hazard), formerly known as Thelohania solenopsae (Sokolova & Fuxa 2008), is a pathogen of Solenopsis fire ants. It was first reported in the red imported fire ant, Solenopsis invicta Buren, collected during South American surveys of fire ants in the early 1970s (Allen & Buren 1974) and initially described by Knell, Allen & Hazard (1977). Further details on its life cycle were subsequently provided by Sokolova & Fuxa (2008). Observations in the 1970s (Allen & Buren 1974; Allen & Knell 1980) and field studies in South America in the 1990s (Briano et al. 1995) led to the conclusion that K. solenopsae potentially could be utilized as a classical biological control agent. In 1996, K. solenopsae was detected in the U.S. from S. invicta populations in Florida and shortly thereafter identified in Mississippi and Texas (Williams et al. 1998; Williams et al. 2003).

The host range of K. solenopsae was reported to be specific to Solenopsis fire ants in the Saevissima species group with the following species being suitable hosts: S. invicta, S. richteri Forel, S. quinquecuspis Forel, S. saevissima Smith, S. macdonaghi Santschi, and S. interrupta Santschi (Oi & Valles 2009; references therein). Recently, Ascunce et al. (2010) reported that Solenopsis geminata (F.) and the S. geminata × S. xyloni McCook hybrid were capable of serving as hosts for K solenopsae. These ant species are in the geminata species group and are native to North America. Examination of archived S. invicta specimens in Texas documented the presence of K. solenopsae in S. invicta in the U.S. as early as 1984 (Snowden & Vinson 2006). Ascunce et al. (2010) hypothesized that K. solenopsae in North America may have transferred from S. geminata to S. invicta in this region, but indicated this scenario remains uncertain.

Post-release monitoring for non-target effects from biocontrol agents is an important aspect of assessing the success of biological control projects (Louda et al. 2003; Van Driesche & Murray 2004; Vazquez & Porter 2005). While K. solenopsae from South America was never released purposely in the U.S., it is still important to ascertain its host specificity on ants in the U.S. to indicate possible host transitions and concomitant effects on native and non-target species. In addition, there is interest to determine the potential of K. solenopsae to control other pest ant species. Our objective was to examine the potential of K. solenopsae to infect non-Solenopsis ant species by 1) examination of various non-S. invicta ant species collected from areas in which S. invicta is infected with K. solenopsae, and 2) inoculation of non-S. invicta ant colonies with K. solenopsae.

In Jul and Aug 2008, various species of adult worker ants were collected from nests and food lures at 3 sites in Gainesville, Florida. Kneallhazia solenopsae has been documented to be present at the Center for Medical, Agricultural and Veterinary Entomology (CMAVE) site since Apr 1996, and at another site, the former University of Florida (UF) poultry unit, since 2003 (D. H. O. unpublished data). Historical infection levels at the third site (near the UF Microbiology building) were unknown, but this site is within 0.7 km of CMAVE. The percent prevalence of K. solenopsae per site was based on the number of S. invicta samples that were infected out of the 14–28 nests or food lure samples per site. Infections were determined by phase-contrast microscopy examination of an aqueous extract of a macerated group of approximately 10–30 adult worker ants per sample for K. solenopsae spores (Williams et al. 1998). Detection of K. solenopsae in the other ant species was performed by polymerase chain reaction (PCR) (Valles et al. 2002) using pooled DNA samples of adult ants grouped by species and collection site with 2–25 ants per sample. All PCRs included positive controls to verify the proper function of the test.

For the colony inoculations, 0.5 – 1.0 g of brood from S. invicta colonies having a mean intracolony infection prevalence of 74% ± 22 SD was placed into laboratory colonies of 6 ant species. Intra-colony prevalence of K. solenopsae was based on microscopic examination of wet mounts of 10–20 individual adult workers per colony or 10 Giemsa stained slide mounts of individual 4th instar larvae or prepupae per colony. The following ant species were tested: the Argentine ant, Linepithema humile Mayr; the bicolor trailing ant, Monomorium floricola (Jerdon); the Florida carpenter ant, Camponotus floridanus (Buckley); the pavement ant, Tetramorium sp. E (formerly T. caespitum [Schlick-Steiner et al. 2006]); the southern fire ant, Solenopsis xyloni; and the tropical fire ant, Solenopsis geminata. In addition, S. invicta colonies were inoculated to serve as positive controls. This method of inoculation is the only known procedure for infecting fire ant colonies (Oi and Valles 2009).

Colonies were each examined for the presence of vegetative stages of K. solenopsae in Giemsastained slides of individual larvae and/or prepupae (n = 10) and the presence of spores in extracts of macerated groups of pupae (n ≈ 10) (Undeen 1997). These microscopy examinations were initiated at 6 or 8 wk after inoculation and continued every 2–4 wk until the studies were terminated usually after 19–24 wk. Infections were determined only from samples obtained after 8 wk (56 d) to avoid sampling brood used for inoculations. Eggs from infected and uninfected S. invicta colonies develop into adults within 29 to 47 d at daily minimum and maximum temperatures of 25.6–30.8 °C (Porter 1988; D. H. O. unpublished data). In addition, the volume of brood, numbers of workers and queens were visually estimated every 2–4 wk based on procedures used to assess laboratory colony status of fire ants and Pharaoh ants, Monomorium pharaonis (L.) (Banks & Lofgren 1991; Williams & Vail 1993). Laboratory inoculations were conducted from 1998–2002.

Kneallhazia solenopsae prevalence among S. invicta nests was 37.5% (6/16) at the microbiology site; 64.3% (18/28) at the poultry site, and 100% (14/14) at CMAVE. K. solenopsae was not detected in any of the 47 non-S. invicta samples, containing nine species, collected from the 3 field sites (Table 1). Although the presence of other ant species has been limited by the dominant S. invicta (King & Tschinkel 2008), the co-occurrence with infected fire ants provided an opportunity for the other ant species to acquire the pathogen. Indeed, K. solenopsae has been detected in parasitic Pseudacteon fire ant decapitating flies captured in similar habitats (Valles et al. 2009).

TABLE 1.

SPECIES AND NUMBERS OF NON-SOLENOPSIS INVICTA ANT SAMPLES TESTED FOR THE PRESENCE OF KNEALLHAZIA SOLENOPSAE COLLECTED AT THREE SITES IN GAINESVILLE, FLORIDA. K. SOLENOPSAE PREVALENCE AT THE SITES WERE AS FOLLOWS: CENTER FOR MEDICAL, AGRICULTURAL AND VETERINARY ENTOMOLOGY (CMAVE) 100%; POULTRY 56.6%; MICROBIOLOGY 37.5%.

Kneallhazia solenopsae was not detected in any of the inoculated (n = 23) or control (n = 20) colonies comprising 6 non-S. invicta species. In contrast, 50% (n = 14) of the inoculated S. invicta colonies became infected (Table 2) while all non-inoculated controls remained uninfected. Brood volume increased during the testing period among the non-S. invicta colonies (Table 2), as did worker counts; queens remained alive for the duration of the test. This indicated that these colonies were growing and not negatively impacted by the inoculations.

Kneallhazia solenopsae was not detected in the non-S. invicta ants collected 12 yr after its first detection at CMAVE. This is concordant with its host range reported from South America (Oi & Valles 2009) and its absence from non-S. invicta ants collected soon after its discovery in Florida (Williams et al. 1998). Infections also did not occur in non-S. invicta laboratory colonies that were inoculated directly. Thus, there is no evidence from this study of K. solenopsae transmission to non-Solenopsis ants. However, the recent detection of K. solenopsae in S. geminata, and the S. geminata × S. xyloni hybrid (Ascunce et al. 2010) illustrates the wisdom of periodic post-introduction monitoring.

SUMMARY

Post-entry host specificity testing was conducted on ants in Florida for the fire ant pathogen, Kneallhazia solenopsae. The pathogen was not detected in 47 samples that contained 9 non-Solenopsis invicta species and a total of 308 ants. Ants were collected from 3 field sites where K. solenopsae was present at the time of sampling. These sites were either at or within 0.7 km of an area where K. solenopsae was observed 12 yr earlier. Infections also were not detected in 23 laboratory colonies consisting of 6 nonS. invicta ant species that were inoculated with K. solenopsae, i.e., Linepithema humile Mayr, Monomorium floricola (Jerdon), Camponotus floridanus (Buckley), Tetramorium sp., Solenopsis xyloni McCook, and Solenopsis geminata (F.). Thus, transmission of K. solenopsae to non-S. invicta ants was not evident in our field and laboratory testing.

TABLE 2.

SPECIES AND NUMBERS OF NON-SOLENOPSIS INVICTA ANT COLONIES INOCULATED WITH BROOD FROM KNEALLHAZIA SOLENOPSAE-INFECTED S. INVICTA COLONIES AND TESTED FOR THE PRESENCE OF K. SOLENOPSAE. ASSOCIATED WITH EACH TESTED SPECIES, AND PRESENTED IN THE LAST COLUMN, ARE THE NUMBERS OF INOCULATED S. INVICTA COLONIES THAT BECAME INFECTED (POSITIVE CONTROLS).