4. Discussion
Identifying Bt toxin receptors in insects has been challenging. Firstly,
it is due to the complexity of Bt toxicology in insects. The action of a
Bt toxin often requires binding to multiple insect proteins that
unequally contributed to Bt toxicity. Besides, receptors to a Bt toxin
in an insect species often are irrelevant to the toxicity of the same Bt
toxin in other insect species. Secondly, there are no suitable methods
that are efficient enough, regarding sensitivity and convenience, to
detect Bt-receptor interaction in a wide range of insect species. So
far, genetics-based methods like QTL analysis have demonstrated the
strength in the identification of Heliothis virescens cadherin
genes of and Bombxy mori ABBC2 genes in the resistance of those
insects to Bt Cry1A toxins (Atsumi et al. 2012, Gahan et al. 2001, Gahan
et al. 2010). Such approaches involve years of experiments in
hybridizing of insects with different Bt resistance traits, which are
often labour-intensity and time-consuming. Alternatively,
affinity-abased methods, like widely used ligand dot (Wei et al. 2016,
Qiu et al. 2017), require previous knowledge of Bt toxin binding
partners often lacked for a novel Bt toxin and unclear to numbers of
insect species. Furthermore, these methods are unlikely to identify low
abundance membrane proteins, that potentially contribute to Bt toxin
function (Candas et al. 2003, Krishnamoorthy et al 2007). Here, we
presented the on-membrane capture, a reliable technique provided a new
solution to identify Bt toxin receptors in insects. This technique
allows us to rapidly and unbiasedly analyze the targets of a given Bt
toxin in any insects without a presumption of the targets. Given that
application of Bt toxins is consistently threatened by evolved
resistance of insects, the on-membrane capture can evaluate the
potential targets of in-service and new designed Bt toxins that will
greatly facilitate assessing the risk of insect evolution resistance and
choosing appropriate sets of toxins for delaying resistance.
Taking advantage of the on-membrane capture, we identified numbers of
proteins in the DBM midgut as binding sites of Bt toxins. Among these
proteins, some well-characterized proteins like cadherin and APN2 are
captured by all three Bt toxins regardless of toxicity, which is
consistent with the previous conclusion that cadherin and APN2 not
associated with DBM resistance to Bt toxins (Nakanishi et al. 2002,
Chang et al. 2012, Guo et al. 2015). Moreover, previously undescribed
binding proteins of Bt toxins identified here shed new light on the
mechanisms of Bt toxicity in DBM. Four Bt toxin-binding sites have been
proposed to explain the action of Bt toxins in DBM (Ferré and van Rie
2002). Site 1 binds Cry1Aa only, site 2 binds Cry1Ab, Cry1Ac, and Cry1F
(Ferré and van Rie 2002), site 3 specifically binds Cry1B, and site 4
binds Cry1C. Modification in these binding sites leads to DBM become
cross-resistance to multiple Bt toxins (Ferré and van Rie 2002). For
instance, modification of site 2 abolishes the binding to Cry1Aa,
Cry1Ab, Cry1Ac and causes DBM cross-resistance to these Bt toxins (Ferré
and van Rie 2002). We found DBM arylphorin subunit alpha proteins are
likely belonging to site 2. Arylphorin is a mitogenic agent in the
midgut stem cells of Lepidopteran insects (Blackburn et al. 2004,
Micchelli and Perrimon 2006). Increased expression of arylphorin inSpodoptera exigua was found to be correlated with B.
thuringiensis resistance (Hernandez-Martinez et al. 2010). Likelihood
of arylphorin subunit alpha proteins involved in DBM resistance to
Cry1Ab and Cry1Ac toxins needs further investigations.
We also found that 13 previously unknown proteins specifically bind to
Cry1Bd. DBM in the field have developed high resistance to Cry1Ac and
become cross-resistance to Cry1Ab, and it remains highly susceptible to
Cry1Bd. These 13 proteins are likely to be candidates to study the
mechanisms of Cry1Bd toxin in DBM. Among them, GSSs are enzymes used by
DBM to protect itself against the accumulation of toxic compounds from
Brassicaceae (Ratzka et al 2002). When these plants are damaged by
herbivory, a myrosinase processes glucosinolates into compounds that are
toxic to the insect (Halkier and Gershenzon 2006). DBM counters this
process by using GSSs to convert glucosinolates into non-toxic compounds
(Ratzka et al 2002). Here, we found that GSS1 and GSS2 are receptors
interacted specifically with Cry1Bd, suggesting the detoxification
system of DBM is targeted by Cry1Bd. DBM GSSs are predicted
extracellular proteins as they have predicted N-terminal secretory
signal peptides. GSSs are likely to be selectively included in lipid
rafts where the interaction with Cry1Bd happens, similar to many
pore-forming toxins (Bravo et al. 2007).
Insect enzymes participated in detoxification of plant secondary
compounds are frequently associated with insecticide resistance (Wang et
al. 2018). GSSs may have roles in DBM develop resistance to Cry1A,
because GSSs do not interact with Cry1Ac or Cry1Ab andGSS -silenced larvae showed increased resistance against Cry1Ac
and Cry1Ab. GSSs have been found in higher levels in a Cry1Ac-resistant
DBM strain NO-QA (McNall 2014). But GSSs of a resistant strain (NO-QA)
and a susceptible strain (Geneva 88) showed no interaction with Cry1Ac,
implying that GSSs might be indirectly involved in the action of Cry1A
toxins. In eukaryotes, sulfatases are extensively glycosylated before
being transported to their destinations (Hanson et al. 2004). GSSs may
be likely involved in Cry1A toxicity through their terminal GalNAc
residues. Indeed, GalNAc has been shown to bind with the
carbohydrate-binding sites of domain III of Cry1Ac (Derbyshire et al.
2001). Alternatively, GSSs participate in Cry1A toxicity via a
signalling pathway. Sulfatases have been attributed to pivotal roles in
Wnt (Dhoot et al. 2001) and pheromone signalling (Ragsdale et al. 2013).
The MAPK signalling pathway has found to manipulate the expression of
multiple receptors relating DBM resistance to Cry1Ac (Guo et al. 2015).
It will be important for future studies to investigate whether MAPK
signalling pathways are involved in regulating the functions of GSSs,
and thus whether they influence the development of DBM resistance.