Locoweeds (mostly Astragalus and Oxytropis) are legumes that contain swainsonine, a toxic alkaloid produced by the vertically-transmitted, fungal endophyte Alternaria section Undifilum. These plants cause economic losses through livestock poisoning, yet there is little evidence that swainsonine plays a role in anti-herbivory. Similarly, while swainsonine synthesis relies on nitrogen from the plant, and many fungal endophytes are thought to be mutualists, the ecological role of this endophyte across the mutualism-commensalism-parasitism continuum is yet unknown. We examined the fitness of A. mollissimus and O. sericea plants with and without the endophyte by measuring growth and fecundity parameters, and potential transgenerational effects, in a multi-year, common garden experiment located in or beyond the northern range of locoweeds. We hypothesized that field-grown plants with the endophyte would possess a fitness advantage over plants without the endophyte under growing conditions different from their maternal source. Oxytropis sericea and Astragalus mollissimus var. mollissimus and var. thompsoniae plants germinated from seeds collected in New Mexico, symbiosed with the Alternaria section Undifilum fungal endophyte (E+) and with the fungal endophyte mechanically removed (E-) by removal of the seed coat, were established in an outdoor, common garden in southwest Montana. To evaluate the ability of this vertically-transmitted endophyte to mediate transgenerational effects in progeny, two successive generations of E+ and E- plants were also established in the garden. We measured overwinter survival, photosynthetic gas exchange, flower and seed production, and progeny seed germination on these pairs of plants over nine years, and compared final above and below-ground biomass. Both A. mollissimus varieties grew as annuals under common-garden field conditions, ca. 1500 km north of their native range, with A. mollissimus var. mollissimus plants producing more reproductive stems, flowers, and seed pods, and surviving longer than A. mollissimus var. thompsoniae, regardless of endophyte presence or swainsonine production. The more broadly distributed Oxytropis sericea plants grew as perennials with survival unaffected by endophyte status; however, E+ O. sericea plants produced slightly more reproductive stems, flowers/reproductive stems, and reproductive stem and log crown biomass. Maternal effects detected in the parental generation, Gen 0, were highly variable, but this variation disappeared in subsequent generations, with no endophyte transgenerational effects detected across most variables. Astragalus and Oxytropis gas exchange, fecundity, and seed germination also were unaffected by endophyte status when grown under environmental conditions differing from the collection sites. Contrary to our initial hypothesis of mutualism, the Undifilum endophyte did not appear to improve locoweed survival or fecundity, and the presence or absence of the endophyte had no detectable transgenerational effects. However, the endophyte did slightly increase the number of reproductive stems and flowers/stem and the weight of the crown and reproductive stem tissue in the perennial O. sericea, suggesting more work is needed to examine endophyte effects on carbohydrate biochemistry and pollination parameters. Our findings highlight the importance of evaluating putative commensal host-endophyte relationships under field conditions.

Three locoweed taxa were compared because of their different life histories and ability to produce swainsonine. Astragalus mollissimus varieties have short life histories for perennials, including sometimes growing as annuals or biennials (Soltani et al., 2021), while O. sericea typically lives much longer (Fox et al., 1998; Welsh, 2007). Astragalus mollissimus var. mollissimus (A.m.m.) is a high producer of swainsonine, whereas A. mollissimus var. thompsoniae (A.m.t.) produces little if any swainsonine, barely reaching PCR detection limits of <0.001% (Cook et al., 2016a; Ralphs et al., 2008; Vallotton et al., 2012); we included A.m.t. in this study to allow separate examination of any swainsonine effect. Oxytropis sericea produces intermediate levels of swainsonine (Vallotton et al., 2012). Seeds used to establish this study were collected from northern New Mexico as described in Delaney et al. (2011) for O. sericea and A.m.m., and Vallotton et al. (2012) for A.m.t.

To directly investigate endophyte effects on locoweeds, greenhouse-grown seedlings were transplanted into a common garden (15.2 m x 4.6 m) in a randomized nested design as pairs with endophyte (E+) and without endophyte (E-) within nine zones in the garden at the Montana State University Post Farm near Bozeman, MT (N 45° 40' 30.5484", W 111° 9' 14.2177"; elevation 1450 m), in an Amsterdam silty loam (Engel et al., 2007) with pH 8, 2.1% organic matter, and nitrate-nitrogen at ca. 38 pounds/acre (AgVise lab results from 6” soil samples). Astragalus mollissimus seedlings were planted in the common garden from 2011 to 2015 (var. mollissimus, 96 pairs total) and 2012 to 2015 (var. thompsoniae, 74 pairs total). Most O. sericea (New Mexico seed source; Generation 0) seedlings were planted in 2011 (18 pairs) and 2013 (24 pairs), with additional pairs planted in 2012 (2 pairs), 2014 (3 pairs), and 2015 (3 pairs) to replace deceased plants (50 pairs total). Most Oxytropis plants died, and all A.m.m. and A.m.t. plants died during the 2013-2014 winter, regardless of endophyte status, and pairs were replanted as feasible.

To evaluate any transgenerational effect of the endophyte in O. sericea, 50 E+/E- pairs each of two subsequent generations (Post Farm seed source; Generations 1 and 2) were also planted: First generation (Gen 1) plants were germinated from seeds collected from 10 E- and 10 E+, 1-year-old, parent plants (Gen 0) established in 2013; five seedlings from each pair were transplanted into the garden in fall 2015 (50 pairs). Second generation (Gen 2) plants were germinated in 2016 from seeds collected from 5 E- and 5 E+, one-year-old (Gen 1) plants established in 2015; ten seedlings from each of two of these five pairs were transplanted into the garden in 2017 (50 pairs).

We measured overwinter survival, photosynthetic gas exchange, flower and seed production, and progeny seed germination on these pairs of plants over nine years until their death, and compared final above and below-ground biomass for surviving plants in July 2020.