There are 3 classes of PI3Ks and of these, class 1 is the most extensively studied due to having a major role in cancer development (Yang et al., 2019) ( Table 1 ). This class is therefore the target of most therapeutic inhibitors in clinical development. Although not as well characterized as class 1, class 2 PI3Ks also have been shown to have role in cancer development (Gulluni et al., 2017; Gulluni et al., 2019; He et al., 2021). Specifically, this class has been shown to be important for migration of prostate cancer cells (Mavrommati et al., 2016). Class 3 PI3Ks are involved in membrane trafficking, endosome-lysosome maturation, and autophagosome formation (Jean and Kiger, 2014). Class 3 has been shown to play an important role in autophagy in the liver and heart (Jaber et al., 2012). However, this class has not yet been shown to have any role in disease or cancer (Jean and Kiger, 2014). Although PI3Ks are ubiquitous throughout the body, with select isotypes found in various compartment that have unique roles (Leahy et al., 2012).

There are 4 class 1 PI3Ks isotypes that are named after their catalytic subunit proteins ( Table 1 ). Class 1A consists of isotypes p110α, p110β, and p110δ, while class 1B consists of one isotype, p110γ (Yang et al., 2019). Isotypes p110α and p110β are constitutively expressed throughout the body and have many cellular functions (Leahy et al., 2012; Yang et al., 2019). Conversely, PI3K isotypes p110δ and p110γ are only found in immune cells and they can activate or repress downstream pathways (Fruman et al., 2017). Interestingly, p110γ are expressed in lymphocytes and are responsible for chemotaxis, making this specific PI3K isotype an excellent target for adjuvant immunotherapy to limit intracellular survival of pathogens (Leahy et al., 2012).

Class 2 PI3Ks have 3 isotypes: PI3K-C2α, PI3K-C2β, and PI3K-C2γ (Islam et al., 2020) ( Table 1 ). Of these isotypes, PI3K-C2α, PI3K-C2β are found throughout the body, while PI3K-C2γ has been shown to be isolated to liver, pancreas, and reproductive organs (Islam et al., 2020). Isotype Class 2 PI3Ks are responsible for glucose uptake by the liver and have a role in blood pressure regulation (Koch et al., 2021). Studies with PI3K-C2β knock out mice revealed that because of its role in glucose uptake, this enzyme has potential as a target for diabetes treatment (Koch et al., 2021).

There are 4 class 1 PI3Ks isotypes that are named after their catalytic subunit proteins ( Table 1 ). Class 1A consists of isotypes p110α, p110β, and p110δ, while class 1B consists of one isotype, p110γ (Yang et al., 2019). Isotypes p110α and p110β are constitutively expressed throughout the body and have many cellular functions (Leahy et al., 2012; Yang et al., 2019). Conversely, PI3K isotypes p110δ and p110γ are only found in immune cells and they can activate or repress downstream pathways (Fruman et al., 2017). Interestingly, p110γ are expressed in lymphocytes and are responsible for chemotaxis, making this specific PI3K isotype an excellent target for adjuvant immunotherapy to limit intracellular survival of pathogens (Leahy et al., 2012).

Class 2 PI3Ks have 3 isotypes: PI3K-C2α, PI3K-C2β, and PI3K-C2γ (Islam et al., 2020) ( Table 1 ). Of these isotypes, PI3K-C2α, PI3K-C2β are found throughout the body, while PI3K-C2γ has been shown to be isolated to liver, pancreas, and reproductive organs (Islam et al., 2020). Isotype Class 2 PI3Ks are responsible for glucose uptake by the liver and have a role in blood pressure regulation (Koch et al., 2021). Studies with PI3K-C2β knock out mice revealed that because of its role in glucose uptake, this enzyme has potential as a target for diabetes treatment (Koch et al., 2021).

Class 3 has just a single isotype that is named after the catalytic subunit Vps34, with a corresponding regulatory subunit named Vsp15 (Jean and Kiger, 2014) ( Table 1 ). Although insulin does not affect the activity of class 3 PI3Ks, it also has promise as a target for diabetes treatment because of its role in the feedback loop of glucose homeostasis (Nemazanyy et al., 2015). Interestingly, this therapy would target the regulatory subunit, not the usual therapeutic catalytic subunit because knocking down the regulatory subunit Vsp15, not the catalytic subunit Vsp34 has been shown to increase insulin sensitivity (Nemazanyy et al., 2015).

Considering the ubiquitous nature of PI3Ks throughout the body, when designing therapeutics for inhibition of PI3K classes, the specific isotype and its functions must be thoroughly investigated. For infectious disease adjuvants, focusing on classes and isotypes with functions in lymphocytes would result in a more specific therapeutic effect with less unwanted side effects. Therefore, designing specific inhibitors for class 1 isotypes p110δ and p110γ would be ideal for this purpose. Furthermore, it would be advantageous to repurpose PI3K inhibitors as adjuvants for bacterial infections with the extensive amount of research and development into inhibitors of class 1 PI3Ks as cancer therapeutics (Leahy et al., 2012).

PI3Ks are ubiquitous throughout the entire body and have been shown to be important for many physiological processes (Fruman et al., 2017). Research in the late 1980’s revealed that PtdIns-3,4,5-P3, the product of PI3K activation is central for malignant cancerous growth (Whitman et al., 1988; Auger et al., 1989). The PI3K/AKT/mTOR pathway is dysregulated in almost all cancer types leading to uncontrolled growth (Hoxhaj and Manning, 2020). Specifically, activation of the PI3K pathway in cancer cells is responsible for proliferation, invasion, metastasis, and angiogenesis (Rascio et al., 2021). Since the PI3K pathway is a driver of uncontrolled growth and spread of a variety of cancers, there are currently academic and clinical efforts in place to develop PI3K, AKT, and mTOR inhibitors as cancer therapeutics (Yang et al., 2019; Castel et al., 2021).

One important role for PI3K in the liver is regulation of glucose uptake and glycogen storage (Jean and Kiger, 2014; Koch et al., 2021). This makes PI3Ks possible targets for to treat the symptoms of diabetes (Maffei et al., 2018). Similarly, patients with a mutation in PI3K resulting in SHORT syndrome have difficulties with glucose homeostasis mimicking type 1 diabetes (Fruman et al., 2017). However, unlike diabetes, this defect does not affect the production of insulin but the ability of insulin to activate PI3K (Fruman et al., 2017). Its role in glucose regulation makes inhibiting PI3K for cancer therapy problematic because inhibition causes a release of glucose that in turn initiates the release of insulin. The consequence of this insulin release is re-activation of PI3K in tumor cells and ineffective chemotherapy (Fruman et al., 2017). However, only certain classes and isotypes have a role in glucose metabolism and their roles are unique (Maffei et al., 2018). For example, class 2 PI3Ks are activated by insulin (Koch et al., 2021), class 3 PI3Ks are involved in glucose feedback loop (Nemazanyy et al., 2015), and class 1A PI3Ks are the isotypes responsible for glucose regulation in the liver (Maffei et al., 2018). Furthermore, class 1A isotypes p110α and p110β are the major regulators of glucose in the liver, while class 1B p110γ regulates glucose in immune cells (Maffei et al., 2018).

Lastly, relevant to this review, PI3K has a major role in the innate and adaptive immune system. For professional phagocytes such as macrophages and neutrophils, PI3K plays a major role in both phagocytosis and killing of intracellular bacteria within phagosomes (Gillooly et al., 2001). Class 1A isotype p110δ and class 1B isotype p110γ and are the isotypes used specifically in immune cells (Okkenhaug, 2013). In addition to each isotype having a unique role, PI3Ks can activate or repress downstream genes depending on the cell type and associated receptor (Fruman et al., 2017). For example, class 1 Pi3Ks are involved in the phagosome cup formation, while class 3 PI3K is involved in the maturation (Koyasu, 2003). Specifically, class 1B isotype p110γ allows for neutrophil migration by producing PtdIns-3,4,5-P3 at the leading edge of the cell (Jean and Kiger, 2014). In addition, class 1A isotype p110δ, class 1B isotype p110γ, and class 3 PI3K Vps34 are responsible for bacterial clearance by immune cells (Thi and Reiner, 2012).

The extensive list of roles for PI3Ks make them attractive targets for cancer therapy, immune therapy, and diabetes treatment. When looking for therapeutic alternatives to complement our classical antibiotic therapy for drug resistant bacterial infections, targeting the PI3K specific isotypes is an intriguing possibility. Combination therapy with antibiotics and isotype selective PI3K inhibitors has the potential of increasing therapeutic outcome of drug resistant bacterial infections without interfering with PI3Ks in other cell types leading to unwanted side effects.

In 1957 wortmannin, the first PI3K inhibitor was discovered after isolation from the fungal species Penicillium wortmannin ( Cleary and Shapiro, 2010 ) ( Figure 2A ). Similarly, Eli Lilly developed LY294002 as a reversible broad inhibitor of PI3k (Cleary and Shapiro, 2010) ( Figure 2B ). The development of LY294002 was based on optimizing the naturally occurring flavonoid quercetin ( Figure 2C ), that can inhibit a broad range of host kinases (Abdul-Ghani et al., 2006; Imai et al., 2012). These inhibitors have been used extensively in laboratories studying the cellular functions of PI3K but due to their poor solubility and physiochemical characteristics these inhibitors have not been used therapeutically (Cleary and Shapiro, 2010). After the discovery of PI3K inhibitors, efforts to improve their pharmacological characteristics resulted in several new inhibitors being developed (Cleary and Shapiro, 2010). Many studies have worked to improve the physiochemical characteristics of wortmannin and LY294002 through analog design resulting in more stable forms of wortmannin and more soluble analogs of LY294002 (Cleary and Shapiro, 2010).

In addition to the broad inhibitors of PI3K, therapeutics have been developed with the ability to bind both PI3K and the very similar catalytic subunit of mTOR (Cleary and Shapiro, 2010). Although they have not yet acquired FDA approval, there are dual PI3K and mTOR inhibitors in phase 3 of clinical development increasing the effectiveness of PI3K/AKT/mTOR pathway inhibition (Cleary and Shapiro, 2010; Yang et al., 2019). These have shown promise over therapeutics that solely inhibit mTOR, which has a negative feedback loop that activates PI3K and AKT when inhibited. However, it is uncertain at this point whether dual inhibition of both PI3K and mTOR is superior to the inhibition of PI3K alone (Cleary and Shapiro, 2010). In addition, multiple trials for dual inhibitors have been terminated due to low tolerability and adverse side effects (Yang et al., 2019). If the tolerability can be improved, perhaps dual PI3K/mTOR inhibitors could be beneficial for cancer therapy because studies have shown PI3K to be involved in chemotherapy resistance (Cleary and Shapiro, 2010; Garber, 2014).

In recent years, isotype selective inhibitors have been more frequently pursued for development because they have been shown to have fewer side effects than broad-spectrum inhibitors (Cleary and Shapiro, 2010). For example, the PI3K isotype in the liver that is important for sending signals from the insulin receptor is p110α. Broad-spectrum PI3K inhibition can cause hyperglycemia by releasing extra glucose that in turn causes a large release of insulin (Fruman et al., 2017). Therefore, avoiding inhibition of PI3K isotype p110α can eliminate the hyperglycemic side effects that accompany broad PI3K inhibition (Fruman et al., 2017). In addition, it would be optimal to target specific isotypes when eliminating intracellular bacterial survival because p110γ and p110δ are the main PI3K isotypes for lymphocyte signaling (Cleary and Shapiro, 2010). This reveals great promise for avoiding unwanted side effects when using specific inhibitors to eliminate intracellular survival of bacterial pathogens.

Currently, there are isotype selective inhibitors that have been approved by the FDA ( Table 2 ) and others are in stage 3 clinical trials (Garber, 2014). Specifically, in 2014 Gilead had the first PI3K isotype selective inhibitor Zydelig™ (Idelalisib) approved by the FDA for treatment of non-Hodgkin’s lymphoma, chronic lymphocytic leukemia, and follicular lymphoma (Garber, 2014). Zydelig™ is the only approved inhibitor specifically targeting the p110δ isotype of PI3K (Garber, 2014; Yang et al., 2019). To target PI3K p110α, Novartis acquired approval for Alpelisib™ (BYL719), for treatment of breast cancer (Lee et al., 2022). In addition to these very specific inhibitors, Copiktra^® (Duvelisib) and Aliqopa™ (Copanlisib) that have been approved shown to inhibit both p110γ and p110δ isotypes (Garber, 2014; Flinn et al., 2019; Yang et al., 2019). With these current approvals and more in various stages of clinical trials, there are many options to repurpose for adjuvant therapeutics.

Development of p110γ selective inhibitors has been slow because long term cancer treatment with these therapeutics can dampen the immune response to bacterial infections (Yang et al., 2019). However, acute inhibition of PI3K does not have the same effects on the immune system as long-term inhibition (Adefemi et al., 2020). Although T cell activation is inhibited, short term acute PI3K inhibition enhances the myeloid immune response to infections resulting in better infection control (Adefemi et al., 2020). Interestingly, it has been shown that chronic inflammation in elderly patients causes aberrant migration of neutrophils and PI3K inhibitors can help improve chemotaxis accuracy (Sapey et al., 2014). Aberrant neutrophil migration has also been observed in patients with severe sepsis (Patel et al., 2018). This exciting potential therapeutic application for PI3K inhibitors needs to be explored more thoroughly because the devastating mortality rates associated with sepsis (Vincent et al., 2002; Fleischmann-Struzek et al., 2020). Overall, selective inhibitors are optimal to repurpose as adjuvants for bacterial infection treatment because the PI3K isotypes used by the immune system are not ubiquitous throughout the body and the treatment length for bacterial infections is shorter than cancer therapy.

Intracellular bacteria can evade host immune clearance, reside, and replicate within macrophages and epithelial cells (Xue et al., 2010).. This ability to replicate and survive within the host is paramount to their success as a pathogen (Allwood et al., 2011; Mitchell et al., 2016). These pathogens are very difficult to treat because they can hide within the mammalian cells to evade antibiotic treatments (Kamaruzzaman et al., 2017; Tucker et al., 2021).

Francisella tularensis is potential bioterrorism agent that causes fatal pneumoniae and no vaccine is available (Maurin, 2015). F. tularensis has an intracellular life cycle where it escapes from the phagosome and replicates within the macrophage cytosol (Celli and Zahrt, 2013; Ledvina et al., 2018) ( Figure 3A ). F. tularensis uses OpiA, its own bacterial PI3K to promote bacterial escape into the cytosol and this kinase has been shown to be resistant to the fungal PI3K inhibitor wortmannin (Ledvina et al., 2018). Furthermore, it has been suggested that additional intracellular bacteria, may harbor their own bacterial PI3K to facilitate phagosomal escape because OpiA family proteins (OFP) have been found in the genomes of intracellular replicating Legionella, Vibrio, and Rickettsia genera (Ledvina et al., 2018). Another potential bioterrorism agent, Bacillus anthracis replicates in the cytosol of host cells and manipulates the host PI3Ks to allow for intracellular replication (Xue et al., 2010). B. anthracis spores invade lung epithelial cells to replicate then escape to spread the infection (Xue et al., 2010) ( Figure 3B ). Xue et al. showed that treatment of A549 cells with the PI3K inhibitors LY294002 or wortmannin resulted in a drastic reduction in B. anthracis spore internalization, revealing the importance of PI3K activation by this species (Xue et al., 2010).

Burkholderia pseudomallei is a U. S. Tier 1 select agent for which there is no vaccine and high mortality rates are associated with the infections (Adefemi et al., 2020). This species is a facultative intracellular pathogen that can invade and replicate in both professional phagocytes and non-phagocytic cells ( Figure 3C ) (Adefemi et al., 2020). Specifically, B. pseudomallei can invade both epithelial cells and macrophages by using PI3K to hijack the host cell actin, ultimately creating multinucleated giant cells leading to apoptosis and escape (Hii et al., 2008; Pflughoeft et al., 2019). Interestingly, Ganesan et al. found that treatment with chloroquine in combination with doxycycline resulted in greater murine survival from B. pseudomallei infections by inhibiting an enzyme downstream to PI3K, glycogen synthase kinase-3β (Ganesan et al., 2020). This study shows the therapeutic benefit of repurposing host targeting drugs as adjuvants with a common antibiotic for treatment of bacterial infections (Ganesan et al., 2020).

The obligate intracellular Chlamydia species ability to manipulate host cell kinases is necessary for its success and survival as a pathogen (Sah and Lutter, 2020). More specifically, inhibition of PI3Ks by Chlamydia species allows for cell invasion, suppression of the host apoptosis and the acquisition of nutrients necessary for survival (Sah and Lutter, 2020) Once inside the host cell Chlamydia trachomatis differentiates from elementary bodies to reticulate bodies within inclusions that will eventually be either extruded from the cell or released by bacterial lysis after the cells mature and differentiate back into elementary bodies ( Figure 3D ) (Adefemi et al., 2020). In C. trachomatis, an effector protein TepP can recruit and activate PI3K on membranes to initiate bacterial invasion (Carpenter et al., 2017). Interestingly, the infectious elementary bodies of C. trachomatis that spread the infection have the highest amount of TepP proteins (Carpenter et al., 2017).

Salmonella species are facultative intracellular pathogens that have the ability to use multiple mechanisms to invade and replicate within host cells (Boumart et al., 2014) Salmonella typhimurium manipulates PI3K to pass through the gut epithelial lining, escape through the basolateral wall, and invade macrophages to replicate within the Salmonella containing vacuole (SCV) ( Figure 3E ) (Hurley et al., 2014). These species invade non-phagocytic cells by two mechanisms: the Zipper mechanism, shared by Listeria monocytogenes and the Trigger mechanism, shared by Shigella flexneri ( Boumart et al., 2014 ). S. typhimurium manipulation of PI3K allows it to highjack macrophages, using them to disseminate the infection (Garcia-Gil et al., 2018). The S. typhimurium effector protein SopB activates PI3K pathway in B cells to facilitate survival (Roppenser et al., 2013; Garcia-Gil et al., 2018), by not allowing them to form the NLRCR4 inflammasome and fight the infection (Garcia-Gil et al., 2018).

Mycobacterium tuberculosis not only survives within macrophages but replicates within the phagosome (Liu et al., 2016). M. tuberculosis uses a recombinant leucine-responsive regulatory protein (rLpr) to increase activation of PI3K via the toll-like receptor 2 (TLR-2) (Liu et al., 2016). In addition, the marker for phagolysosome fusion Rab7 is targeted by M. tuberculosis, not allowing the phagosome to mature to the late stage (Nguyen and Yates, 2021). Following replication within the phagosome, the phagosomal membrane is permeabilized allowing the bacteria to escape to the cytosol, triggering necrosis of the macrophage and allowing escape of M. tuberculosis to infect other macrophages ( Figure 3F ) (Behar et al., 2010).

Some pathogens are not all traditionally considered when investigating bacterial survival within host cells (Sendi and Proctor, 2009; Silva, 2012). However, these pathogens can manipulate the immune system to their advantage to allow survival and spread of the infection (Silva, 2012; Silva and Pestana, 2013). Therefore, when considering adjuvant PI3K therapy these pathogens should also be considered potential targets.

Staphylococcus aureus is a dangerous pathogen with community acquired infections easily spread due to their ability to invade healthy individuals (Boyle-Vavra and Daum, 2007; DeLeo et al., 2009; Bukharie, 2010; Yoong and Pier, 2010; Turner et al., 2019). Skin infections caused by S. aureus can quickly invade and disseminate leading to systemic infection with sepsis (Edwards et al., 2010). Although traditionally thought of as an extracellular pathogen, S. aureus has been shown to invade non-phagocytic mammalian cells (Nakagawa et al., 2017). For example, keratinocyte invasion by S. aureus induces lysis continued invasion of the dermis layer. When engulfed by macrophages and neutrophils partial evasion of phagolysosome killing leads to escape and dissemination of infection (Oviedo-Boyso et al., 2011; Hommes and Surewaard, 2022) ( Figure 4A ). Interestingly, S. aureus internalization decreases in bovine epithelial cells when treated with PI3K inhibitors (Oviedo-Boyso et al., 2011). Escherichia coli K1 has been shown to invade endothelial cells to cross the blood brain barrier and cause bacterial meningitis (Reddy et al., 2000). E. coli has been shown to manipulate the human brain microvascular endothelial cells and force micropinocytosis (Loh et al., 2017) ( Figure 4B ). Attachment to the epithelial cell lining of the blood brain barrier by E. coli activates several signaling pathways including the PI3K pathway (Loh et al., 2017).

A close relative to E. coli, K. pneumoniae is historically know an extracellular pathogen, unable to survive for extended periods in intracellular compartments like its close relative E. coli (Belon and Blanc-Potard, 2016). However, studies have now shown that K. pneumoniae is able to survive for up to 48 hours within the vacuole once engulfed by macrophages (Oelschlaeger and Tall, 1997). This vacuole, named the Klebsiella containing vacuole (KCV) does not fuse with lysosomes and therefore deviates from the canonical endocytic pathway used by macrophages to clear engulfed pathogens (Bengoechea and Sa Pessoa, 2019) ( Figure 4C ). This lack of maturation of the phagosome also results in less activation of the adaptive immune system because less antigens are presented on the surface of macrophages (Cano et al., 2015). Although it is unclear the factor that allows K. pneumoniae to persist within the phagosome, the capsule does not appear to play a major role because both hypervirulent and classical K. pneumoniae can manipulate the PI3K pathway to stop phagosome maturation by inhibiting fusion with the lysosome compartment ( Figure 4C ) (Cano et al., 2015). These classical intracellular pathogens have been shown to display less intracellular survival in the presence of AKT inhibitors targeting the enzyme immediately downstream to PI3K in the PI3K/AKT/Rab14 axis (Bengoechea and Sa Pessoa, 2019). In addition, it has recently been revealed that K. pneumoniae manipulates PI3K through the mammalian protein SARM1 (sterile α and HEAT armadillo motif-containing protein) revealing a potential target upstream for future therapeutic development (Feriotti et al., 2022). With the many bacterial species that manipulate PI3K to invade the host and survive intracellularly, repurposing PI3K inhibitors to release them from their protective niches would be beneficial to allow the host immune system and antibiotics to be more effective at treating these infections.

Studies have shown that acute PI3K treatment can improve the early-stage progression of infections (Adefemi et al., 2020). Many genera of bacteria manipulate PI3K to avoid phagolysosome fusion and PI3K inhibitors allow for efficient fusion of the lysosome and bacterial clearance ( Figure 5A ). This has been used to show that PI3K/AKT inhibition can eliminate intracellular S. typhimurium and M. tuberculosis (Bengoechea and Sa Pessoa, 2019). Interestingly, inhibition of PI3K in M. tuberculosis by isotype specific inhibitors has the potential to decrease the characteristic late-stage infection IL-17A induced pathology by interfering with Th17 differentiation (Leisching, 2019). Shapira et al. performed high-content screening of kinase inhibitors and found inhibition of PI3K controls autophagy and apoptosis decreasing intracellular survival (Shapira et al., 2020). This effect has also been seen with the facultative intracellular K. pneumoniae that manipulates the PI3K pathway to avoid phagolysosome fusion of late-stage endosomes (Cano et al., 2015). Treatment with a PI3K inhibitor revealed a decrease in K. pneumoniae intracellular survival within macrophages (Cano et al., 2015; Bengoechea and Sa Pessoa, 2019). Inhibiting bacterial pathogen survival in host immune cells has great potential for treatment of chronic infections.

PI3K inhibition also has the potential to decrease bacterial invasion and survival within non-phagocytic cells like epithelial cells. Testing PI3K inhibitors when infecting intestinal cells with S. typhimurium, Huang et al. revealed that this species uses PI3K activation to decrease inflammation. This decreased inflammation allows S. typhimurium to survive and PI3K inhibition led to decreased bacterial survival within intestinal epithelial cells (Huang et al., 2005). Like S. typhimurium, PI3K manipulation is also important for epithelial cell invasion by Helicobacter pylori and Listeria monocytogenes revealing the potential of PI3K inhibitors for a variety of intracellular pathogens ( Figure 5B ) (Booth et al., 2003). Furthermore, the penetration of the blood brain barrier by E. coli leading to meningitis can be stopped by using PI3K inhibitors (Loh et al., 2017). These studies reveal the potential of PI3K therapeutics for non-phagocytic cells revealing a broader application for adjuvants for a variety of infections and bacterial species.

PI3K inhibitors have also shown promise in treating pathogens that do not have a true intracellular stage (Whitman et al., 1988; Cano et al., 2015; Yang et al., 2019). With these infections, inflammation and toxic damage can be mitigated by using PI3K inhibitors. For example, the natural PI3K inhibitor deguelin reduces Staphylococcal Enterotoxin B induction of T-cell proliferation toxicity (Whitfeild SJ et al., 2017) ( Figure 5C ). However, more work is needed to understand all the potential effects of PI3K treatment for bacterial infections. For example, the treatment of an E. coli induced endotoxemic mice with PI3K inhibitors lead to a decrease in mouse survival because of increased LPS-induced inflammation (Schabbauer et al., 2004). This data reveal that PI3K inhibitors can also be repurposed to protect from the effects of bacterial toxins during infection but much research is needed to pursue these applications.