The European Union (EU) is heavily dependent on imports of vegetable protein sources, only assuring a self-sufficiency of 79% for rapeseed, 42% for sunflower, and 5% for soya (1). Annual imports account for 17 million tons of crude protein (CP), over 76% of which of soya, the most prevailing protein source for feed and food (1). Although human consumption of vegetable proteins is raising in the EU, with the market for meat and dairy alternatives growing by 14 and 11% per year, respectively, the animal feed sector is by far the most important outlet (93% by volume) (1).

Grain legumes can effectively contribute to balance the European economic trade of plant-based protein sources for feed and food, but also play a key role in sustainable agricultural intensification (2). The cultivation of legumes has several benefits, including (i) fixation of atmospheric nitrogen (N) into the soil with improvement of soil fertility and reduction of chemical fertilizers (3); (ii) improvement of soil structure and health in crop rotation cycles and intercropping, allowing the replacement of traditional fallow and increasing the productivity of the next crop cycle (3, 4), (iii) promotion of biodiversity of rural landscape (5); (iv) contribution to mitigate greenhouse gases emissions and to tackle climate change (6, 7); and (v) improvement of farming profitability and efficiency (8).

White (Lupinus albus L.), yellow (Lupinus luteus L.), and blue or narrow-leafed (Lupinus angustifolius L.) lupins are native European legumes well adapted to acidic, sandy soils, a trait that differentiates them from other grain legumes (9). Lupins produce grains with high protein content (up to 44% dry matter, DM, basis) (10), with high lysine content but deficient in methionine and cysteine (11). Even though the world lupin production is increasing, in Europe the production area and yields are still modest (12, 13), reflecting a difficulty of European farmers to alter the cropping systems toward a transition to legume-supported farming (14).

Lupinus grains are harvest when dry, resulting in large amounts of biomass residues, including stalks, leaves, and pod shells, that traditionally remain in the fields to be burned or incorporated in the soil. Although burning is still used, post-harvest biomass incorporation is the most common practice as it improves soil fertility, through the increase of organic matter, N, phosphorus (P), and other nutrients (4, 15, 16), and the promotion of soil microbial composition and diversity (17). Post-harvest biomass residues may be further valorized as animal feed, especially in more extensive systems, either by direct grazing by small ruminants, or conservation for periods of fodder scarcity (18). Lupin biomass residues may be of greater relevance for the agricultural systems of the Mediterranean region, as the prolonged periods of severe and extreme drought aggravated by climate change and the increased incidence of large fires have drastically reduced the availability of pastures for ruminant animals.

Like other legumes, lupin straws have been reported to present higher CP and neutral detergent-soluble carbohydrates content than cereal straws, with overall greater DM digestibility but lower neutral detergent fiber (NDF) digestibility, due to the higher lignin content (19–21). Lupin straws and pod shells may present an additional challenge to be used as feed due to their potential alkaloid content. In bitter narrow-leafed lupin, quinolizidine alkaloids were shown to be produced in the aerial parts of the plant and then transferred to grains (22). The content of these secondary metabolites reduces in stems and leaves with plant maturation, increasing in pods from flowering to grain formation and then decreasing as they accumulate in grains (23). Although it may be anticipated that lupin straws and pod shells present low alkaloid content, no reports were found in the literature. Thus, to effectively assess the potential of lupin biomass residues as alternative feeds for ruminant animals, alkaloids content and profile must be characterized to predict the potential exposure of ruminants to these phytochemicals that may present toxic and teratogenic effects on ruminants (24).

In this context, the present study aimed to evaluate the potential of post-harvest biomass residues of three natives European Lupinus sp., white (L. albus ‘Estoril’), narrow-leafed (L. angustifolius ‘Tango’), and yellow (L. luteus ‘Cardiga’), as alternative feeds for ruminant animals. To achieve this aim, the production, nutritive value and detailed alkaloid characterization were assessed on straws and pod shells of the three lupin cultivars harvested in two locations (Mirandela and Vila Real). The biomass residues here assessed correspond to the first sowing date (mid-September) of the field experiment presented by Monteiro et al. (25), as it was the sowing with the highest production yield of Lupinus grain.

The world growing population and the increasing demand for food are pressing the agricultural systems to evolve toward more efficient and sustainable practices to effectively address food and nutrient security (35). To achieve this societal challenge, the European stakeholders have adopted the strategy of Circular Economy and Bioeconomy (36, 37), focused on preventing and reducing waste and increasing its value through their incorporation into new processes (38). The agricultural sector produces a large amount of biomass throughout the agri-food chain, from farm to fork, discarded as waste, with negative economic, social, and environmental impacts (39). Global agri-food waste was estimated to contain nutrients to support food security for 2000 million people (40), highlighting the importance of agri-food biomass to a circular and low-carbon economy (41–43). As the ultimate up-cyclers, ruminants have the ability to convert agri-food residues, by-products, and co-products, or inedible to humans into foods of high biological value, thus addressing food security and environmental sustainability (44, 45). In this context, locally produced agricultural residues’ biomass can be used as ruminant feeds, thus effectively contributing to valorizing agricultural wastes and providing underexplored low-cost feed resources while promoting the circular economy.

The interest in native European legume production for food and feed has increased over the last few years. In 2021, European lupin production accounted for 391,342 tons, corresponding to 28.7% of the world production yield (13). Although scarce data exist on the agro-residues generated during lupin seed harvest and post-harvesting processes, it has been estimated that 7 tons of biomass residues are produced per ton of seed (46). Unlike cereal straws, lupin biomass residues (e.g., stalks, stubbles, straws, pod shells, husks) are traditionally left in the fields and burned in open fires and/or incorporated in the soil to promote carbon and minerals content (47). Alternatively, the residues can be grazed by livestock, particularly small ruminants, thus adding value to these biomasses with no environmental burden. However, grazing presents hazards as lupins may be colonized by several fungi causing important yield-limiting diseases, such as Phomopsis blights caused by Diaporthe toxica (48). When lupin stalks, stubbles, straws, and grains are left in the fields, in particular after the first rain, Phomopsis blights may become a health issue to livestock as D. toxica mycotoxins may induce lupinosis, a liver degenerative disease that can also cause brain damage and death (49).

Under this rationale, we anticipate that the biomass residues obtained from lupin grain harvest may be valorized if collected, stored, and used as fodder for ruminants in periods of feed scarcity. On the other hand, as legumes are harvested some centimeters off the ground, the stalks and stubbles that remain in the field could be further incorporated into the soil or processed as soil amendment (46), along with the root biomass that contributes to promote the soil OM and nutrient content (50). This holistic approach addresses the quest for biomass for soil amendment and animal feeding, pointing out the need to sustainably balance biomass demands (4).

Straws and pod shells comprise the main post-harvest biomass residues from lupin crop production. In this study, the production, nutritive value, and alkaloid content were assessed separately for straw and pod shells. Regardless of location, the combination of straws and pods per variety accounted for 63.5–74.3% of total aerial biomass harvested (grains, straws, and pod shells), agreeing with the biomass residues of L. angustifolius and L. mutabilis Sweet, also known as Andean lupin, which accounted, respectively, for 71.0 and 87.5% of total harvest (46, 47, 51). The overall results here presented suggest a poor adaptation of the early cultivar L. angustifolius ‘Tango’ to the specific climatic conditions of both studied locations during the agronomic year of 2018/2019. White, narrow-leafed, and yellow lupins are widely distributed in the Mediterranean region, being particularly adapted to acidic and sandy soils, with poor fertility and low water-holding capacity (52). However, the agricultural year was atypically hot, with the average maximum and minimum temperatures higher and with lower precipitation than the historical data, except for two months (November and April) that were cooler and rainier than the observed between 1971 and 2010, which may have posed particular issues to the drought-escaping L. angustifolius (53).

European lupin biomass residues have long been regarded as of low value. The lack of interest on legume biomass residues is reflected on the scarce availability of nutritive value data in the literature. Despite species-specific differences, the proximate composition of the three lupin species biomass residues evaluated confirmed their high fiber, moderate CP, and low EE content. However, when compared to cereal straws, white, narrow-leafed, and yellow lupins straws present higher CP, soluble carbohydrates, and lignin, and lower NDF content (19, 54). The ash and ADF content of biomass residues here evaluated were found to vary between sowing locations, reinforcing the impact of edaphoclimatic conditions on lupins production and composition of biomass residues. In contrast, no differences in NDF, ADF, and ADL content were observed in dry stems of L. mutabilis Sweet produced in winter Mediterranean and summer North European crop conditions (55).

To leverage the use of lupin biomass residues as feed resources, it is of utmost importance to evaluate their digestibility. In vitro DMD and OMD were similar among lupin species (47.7–50.6%) and higher in pod shells (53.7%) than in straws (44.6%). To the best of our knowledge, no data is available in the literature on lupin pod shells’ digestibility. Regarding straws, L. albus ‘Estoril’ presented lower DMD than the one reported by López (56) for L. albus sp. (42.1 vs. 69.3%), while that of L. angustifolius ‘Tango’ was higher than the one determined by Mulholland et al. (51) in sheep fed L. angustifolius sp. (47.5 vs. 53.7%). Differences in the nutritive value of lupin straws may be due to different varieties or cultivars, stage of harvest, stem-to-leaf ratio, and edaphoclimatic conditions (56). The latter is highlighted by the greater DMD and OMD of biomass residues harvested in Mirandela location than in Vila Real. In legume residues, the stem-to-leaf ratio is of particular importance as losses due to defoliation are quite high resulting in lupin straw mostly made up of stems, thus reducing their nutritive value (57). Therefore, leaf loss should be reduced to the minimum if lupin biomass residues are to be used as feed, harvesting as soon as maturity is achieved and using machinery that efficiently preserves the leaf component. Nutrient digestibility and bioavailability may also be limited by the phytochemicals present in lupins, including polyphenols, phytosterols, tocopherols, triterpenes, and alkaloids (58).

Traditional Mediterranean livestock systems use ancestral practices based on sustainability and circular economy that promote the use of locally available resources and the balance between agricultural practices, livestock production, environment, and household economy in an integrated approach. By revisiting these practices, the biomass residues of lupin grain production may be a valuable feed resource for ruminants. Considering the combined production of straws and pod shells per lupin species, the ME and CP content (per ha) of biomass residues accounted for, respectively, 26,123 MJ and 216 kg in L. albus ‘Estoril,’ 11,117 MJ and 124 kg in L. angustifolius ‘Tango’ and 24,560 MJ and 227 kg in L. luteus ‘Cardiga.’ Assuming an average ME requirement of dry and mid-pregnancy sheep with 50 kg BW of 9 MJ day⁻¹ and that 85% of the requirements are provided by lupin biomass residues, the ME produced per ha here reported would be sufficient to maintain from 4 to 9 sheep for one year. Moreover, considering a diet with 10% CP (DM basis), the biomass residues could fulfill nearly half of sheep CP requirements.

To effectively assess the potential use of lupin biomass residues as animal feed, their alkaloid content was determined. Plants belonging to the genus Lupinus produce several alkaloids, including indoles, piperidines, and quinolizidines, which play important roles in defense mechanisms against predators and pathogens (59). Quinolizidines are the most abundant alkaloids in lupins, with over 170 bicyclic, tricyclic, and tetracyclic quinolizidines reported, being often referred to as lupin alkaloids (60, 61). Quinolizidine alkaloids are synthesized from lysine in the chloroplast, regulated by light and circadian rhythm, and then transported to leaves and other organs by the phloem, being accumulated in pods and then in seeds as they mature (22, 59). Alkaloid contents vary with species, cultivar and variety, environmental conditions, such as temperature, drought, soil conditions, location, and between years (59, 60, 62). Although these N secondary metabolites play important roles on plant defense mechanisms, they can also have toxic and teratogenic effects on ruminants. The teratogenic quinolizidine alkaloid anagyrine and the piperidine alkaloids ammodendrine and N-methylammodendrine have been described to cause crooked calf syndrome (63, 64), a disease induced by the consumption of the teratogenic lupin alkaloids during early pregnancy that causes maternal muscular weakness and ataxia as well as skeletal deformities in the newborn such as arthrogryposis, scoliosis, kyphosis, and cleft palate (65).

To the best of our knowledge, no study has yet assessed the alkaloids profile in white, narrow-leafed, and yellow lupin straws and pods. Although alkaloids content in stems, leaves, and pods decreases as they accumulate in seeds, L. albus ‘Estoril’ and L. luteus ‘Cardiga’ presented considerable amounts of alkaloids in straws (23.9 and 119 mg kg⁻¹ DM) and pod shells (20.5 and 298 mg kg⁻¹ DM). On the other hand, no alkaloids were detected in L. angustifolius ‘Tango’ biomasses. Lupinus albus ‘Estoril’ was characterized by the piperidine alkaloids smipine, followed by ammodendrine, and the tetracyclic quinolizidine lupanine. The bicyclic quinolizidine lupinine was the main alkaloid present in L. luteus ‘Cardiga,’ followed by the indole gramine, the tetracyclic quinolizidine sparteine, and the piperidine alkaloids ammodendrine and hydroxyammodendrine. Quinolizidine alkaloids of yellow lupin were at higher concentration in pod shells, which is in agreement with the late vigor of L. luteus (52), as a later seed maturation would be reflected in a longer transient accumulation of alkaloids in pods. Being an early vigor lupin species, no differences in alkaloid content were found between biomass residues in L. albus ‘Estoril.’ The impact of edaphoclimatic conditions on alkaloid content of straws and pod shells is also here unveiled. Indeed, lupin straw harvested in Mirandela presented higher piperidine alkaloids and the tetracyclic quinolizidine sparteine than straws harvest in Vila Real; no differences were found for pod shells alkaloids content. Conversely, indole alkaloids, represented by gramine alone, were detected at higher concentrations in pod shells harvested in Mirandela than in Vila Real, whereas straws presented lower and similar contents.

Few studies have evaluated the effects of lupin alkaloids on ruminants’ performance and feed intake and even fewer have assessed the maximum exposure level to these phytochemicals. Assuming Merino rams fed on lupin biomass residues ad libitum supplemented with a protein-rich concentrate (150 g day⁻¹), with a daily DM intake of 21.6 g kg⁻¹ BW, the maximum exposure to lupin alkaloids per day would be of 4.08 mg kg⁻¹ BW with L. luteus ‘Cardiga’ biomass harvested in Mirandela. The exposure would be reduced to nearly half (2.22 kg⁻¹ BW) if the same biomass was harvested in Vila Real. Much lower exposures to alkaloids would occur if sheep were fed L. albus ‘Estoril’ residues, achieving 0.64 mg kg⁻¹ BW per day for biomasses harvested in Mirandela and 0.23 mg kg⁻¹ BW per day for those harvested in Vila Real. No data was found in the literature on exposure limits in sheep or with alkaloids data that allowed the exposure estimation. However, in vitro, Aguiar and Wink (66) found that 1 mM sparteine or lupanine supplementation reduced gas production of hay after 24 h incubation with rumen inoculum collected from sheep fed a roughage-based diet, which corresponds to an exposure of 344 and 365 mg kg⁻¹ BW per day for sparteine and lupanine, respectively, assuming an average DM intake of biomass residues for L. albus ‘Estoril’ and L. luteus ‘Cardiga’ (i.e., 20.85 g kg⁻¹ DM per day) by 55 kg rams.

Estimations of the maximum exposure to lupin alkaloids were also calculated for beef cattle. For these animals, it was assumed a daily DM intake of 2% BW of a diet with 15% protein-rich concentrate. Maximum alkaloid exposure for beef cattle, per day, would be 3.94 mg kg⁻¹ BW for L. luteus ‘Cardiga’ biomasses harvested in Mirandela and 2.15 mg kg⁻¹ BW for those harvested in Vila Real. Daily exposures of 0.57 mg kg⁻¹ BW and 0.21 mg kg⁻¹ BW would be achieved if cattle were fed L. albus ‘Estoril’ biomass residues harvested in Mirandela and Vila Real, respectively. Data from three studies were used, one in heifers and two in dairy cows, to estimate the alkaloid exposure. Johnson et al. (67) found no effect on feed intake, weight gain, and feed efficiency of heifers fed a diet with L. albus ‘Tifwhite’-78 lupin seeds (193 g kg⁻¹, DM basis) for 70 days, thus suggesting that alkaloids intake at 4.5 mg kg⁻¹ BW per day were well tolerated by heifers (68). The effects of intact and heat-treated crushed L. albus seed supplemented at 150 and 300 g kg⁻¹ (DM basis) on feed intake and milk yield were assessed in dairy cows (69); lupanine and 13-hydroxylupanine being the only alkaloids determined. Voluntary feed intake and milk yield were reduced with high alkaloid supplementation and when cows were fed intact seeds compared to heat treated seeds, being the maximum exposure tolerated by dairy cows calculated by Schrenk et al. (68) as 5.2 mg lupanine and 13-hydroxylupanine kg⁻¹ BW day⁻¹. On the other hand, Engel et al. (70) reported no negative effect of feed intake, milk yield and milk fat and protein composition in dairy cows fed 1774 mg L. angustifolius quinolizidine alkaloids day⁻¹, while at 3548 mg day⁻¹ milk yield, but not feed intake and milk composition, was reduced. Assuming a BW of 650 kg, we may assume impaired milk yield at 5.45 mg quinolizidine alkaloids kg⁻¹ BW day⁻¹.

Considering all data available, no negative effects are expected on feed intake, rumen fermentation, and performance of sheep and cattle fed on lupin biomass residues, thus highlighting their potential as alternative forage. However, future studies should be conducted in vivo to validate the safety of lupin biomass residues as alternative feeds for ruminant animals as well as alkaloids maximum exposure. These studies are further supported by the detection of quinolizidine alkaloids in the milk of dairy cows fed as low as 2.73 mg lupin seed kg⁻¹ BW per day (assuming 650 kg BW) (70), highlighting the need to evaluate the potential transfer of alkaloids from dietary lupin biomass residues to dairy cows and sheep milk as it may pose toxicological concerns to consumers (70).

The present study addresses a poorly investigated topic, the use of undervalued lupin post-harvest biomass residues on ruminant feeding as a strategy to leverage the sustainability of production systems and ecological synergies between crop byproducts and animal management under a circular economy approach. Although species-specific traits were found, the native European lupin species, L. albus ‘Estoril,’ L. angustifolius ‘Tango,’ and L. luteus ‘Cardiga,’ biomass residues (straws and pod shells) have considerable production yield and higher nutritive value than cereal straws. Despite their alkaloid content, particularly of L. luteus ‘Cardiga,’ it is anticipated that lupin post-harvest biomass residues present no constraint for sheep and cattle even when consumed ad libitum. Overall results highlight the potential of Lupinus sp. biomass residues as alternative feeds for ruminant animals. As lupin production is estimated to continue to grow in Europe and worldwide, increasing the availability of post-harvest biomass residues and upcycling as feed for ruminants leverages its value and contribution to a more sustainable European farming system, being of particular importance in time of fodder shortage or scarcity in times of climate change.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

The rumen contents collection procedure was reviewed and approved by the Animal Ethics Committee of the School of Medicine and Biomedical Sciences, University of Porto, licensed by the Portuguese General Directorate for Food and Veterinary (permit #0421/000/000/2021), and performed by trained scientists (FELASA category C).

HT conceived and designed the field study. AM, CM, and CC conducted the field study and sample collection. AF and HT conceived and designed the experimental study. MM, IV, and CS conducted the analytical work. PC was involved in animal manipulation and sampling. MM, IV, and AF performed the statistical analysis. MM, IV, CS, AC, and AF analyzed the data. MM and AF wrote the first draft of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

This work received financial support from AgriFood XXI – Development and consolidation of research in the agrifood sector in Northern Portugal I&D&I project (NORTE-01-0145-FEDER-000041), co-financed by European Regional Development Fund through NORTE 2020 programme (Programa Operacional Regional do Norte 2014/2020), and from Portuguese Foundation for Science and Technology (FCT/MCTES) through projects UIDB/04033/2020, UIDB/50006/2020, and UIDP/50006/2020. MM and IV acknowledge FCT for funding through program DL 57/2016 – Norma transitória (Ref. SFRH/BPD/70176/2010 and SFRH/BPD/111181/2015, respectively), and CS acknowledge AgriFood XXI project for the contract.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.