Loading [MathJax]/jax/output/SVG/jax.js
Research article Topical Sections

In vitro activity of some flavonoid derivatives on human leukemic myeloid cells: evidence for aminopeptidase-N (CD13) inhibition, antiproliferative and cell death properties

  • Leukemia cells from patients with acute myeloid leukemia (AML) display high proliferative capacity and are resistant to death. Membrane-anchored aminopeptidase-N/CD13 is a potential drug target in AML. Clinical research efforts are currently focusing on targeted therapies that induce death in AML cells. We previously developed a non-cytotoxic APN/CD13 inhibitor based on flavone-8-acetic acid scaffold, the 2',3-dinitroflavone-8-acetic acid (1). In this context, among the variously substituted 113 compounds further synthesized and tested for evaluation of their effects on APN/CD13 activity, proliferation and survival in human AML U937 cells, eight flavonoid derivatives emerged: 2',3-dinitro-6-methoxy-flavone-8-acetic acid (2), four compounds (36) with the 3-chloro-2,3-dihydro-3-nitro-2-phenyl-4H-1-benzopyran-4-one structure, and three (79) with the 3-chloro-3,4-dihydro-4-hydroxy-3-nitro-2-phenyl-2H-1-benzopyran framework. Different structure-activity relationships were observed between APN/CD13 activity and growth/survival processes. We showed that compound 2, but not benzopyran derivatives 39, inhibited APN activity (although to a less degree than 1). Both 1 and 2 did not affect AML cell proliferation and survival, indicating that CD13’s APN activity is not required for these processes. In contrast, benzopyran compounds 39 inhibited in a concentration-dependent manner the growth of U937 cells by inducing death as evidenced by phosphatidylserine externalization. Cell death was associated with the presence of geminal nitro group and chlorine at the 3-position of the 2H-1-benzopyran scaffold. The presence of other substituents such as CH2COOH or CH2CH=CH2 groups at the 8-position, NO2 or I substituents at the 2'- or 3'-position, OCH3 or OCH2C6H5 groups at the 4'-position did not affect cell death. Importantly, the inhibitory effects evidenced with compounds 79 were not due to their potential decomposition into the corresponding (Z)-(2chloro-2-nitroethenyl)benzene and salicylaldehyde. Based on these preliminary data, the 3-chloro,3-nitro-2H-1-benzopyran derivatives could be classified as a new group of compounds with promising antitumor properties; this study therefore provides the opportunity to explore their potential efficiency in AML patients’ cells ex vivo.

    Citation: Sandrine Bouchet, Marion Piedfer, Santos Susin, Daniel Dauzonne, Brigitte Bauvois. In vitro activity of some flavonoid derivatives on human leukemic myeloid cells: evidence for aminopeptidase-N (CD13) inhibition, antiproliferative and cell death properties[J]. AIMS Molecular Science, 2016, 3(3): 368-385. doi: 10.3934/molsci.2016.3.368

    Related Papers:

    [1] Kaiqing Huang, Yizhi Chen, Miaomiao Ren . Additively orthodox semirings with special transversals. AIMS Mathematics, 2022, 7(3): 4153-4167. doi: 10.3934/math.2022230
    [2] Rukhshanda Anjum, Saad Ullah, Yu-Ming Chu, Mohammad Munir, Nasreen Kausar, Seifedine Kadry . Characterizations of ordered h-regular semirings by ordered h-ideals. AIMS Mathematics, 2020, 5(6): 5768-5790. doi: 10.3934/math.2020370
    [3] Huawei Huang, Xin Jiang, Changwen Peng, Geyang Pan . A new semiring and its cryptographic applications. AIMS Mathematics, 2024, 9(8): 20677-20691. doi: 10.3934/math.20241005
    [4] Waheed Ahmad Khan, Abdul Rehman, Abdelghani Taouti . Soft near-semirings. AIMS Mathematics, 2020, 5(6): 6464-6478. doi: 10.3934/math.2020417
    [5] Pakorn Palakawong na Ayutthaya, Bundit Pibaljommee . On n-ary ring congruences of n-ary semirings. AIMS Mathematics, 2022, 7(10): 18553-18564. doi: 10.3934/math.20221019
    [6] Saba Al-Kaseasbeh, Madeline Al Tahan, Bijan Davvaz, Mariam Hariri . Single valued neutrosophic (m,n)-ideals of ordered semirings. AIMS Mathematics, 2022, 7(1): 1211-1223. doi: 10.3934/math.2022071
    [7] Abdelghani Taouti, Waheed Ahmad Khan . Fuzzy subnear-semirings and fuzzy soft subnear-semirings. AIMS Mathematics, 2021, 6(3): 2268-2286. doi: 10.3934/math.2021137
    [8] B. Amutha, R. Perumal . Public key exchange protocols based on tropical lower circulant and anti circulant matrices. AIMS Mathematics, 2023, 8(7): 17307-17334. doi: 10.3934/math.2023885
    [9] Liaqat Ali, Yaqoub Ahmed Khan, A. A. Mousa, S. Abdel-Khalek, Ghulam Farid . Some differential identities of MA-semirings with involution. AIMS Mathematics, 2021, 6(3): 2304-2314. doi: 10.3934/math.2021139
    [10] Tariq Mahmood, Liaqat Ali, Muhammad Aslam, Ghulam Farid . On commutativity of quotient semirings through generalized derivations. AIMS Mathematics, 2023, 8(11): 25729-25739. doi: 10.3934/math.20231312
  • Leukemia cells from patients with acute myeloid leukemia (AML) display high proliferative capacity and are resistant to death. Membrane-anchored aminopeptidase-N/CD13 is a potential drug target in AML. Clinical research efforts are currently focusing on targeted therapies that induce death in AML cells. We previously developed a non-cytotoxic APN/CD13 inhibitor based on flavone-8-acetic acid scaffold, the 2',3-dinitroflavone-8-acetic acid (1). In this context, among the variously substituted 113 compounds further synthesized and tested for evaluation of their effects on APN/CD13 activity, proliferation and survival in human AML U937 cells, eight flavonoid derivatives emerged: 2',3-dinitro-6-methoxy-flavone-8-acetic acid (2), four compounds (36) with the 3-chloro-2,3-dihydro-3-nitro-2-phenyl-4H-1-benzopyran-4-one structure, and three (79) with the 3-chloro-3,4-dihydro-4-hydroxy-3-nitro-2-phenyl-2H-1-benzopyran framework. Different structure-activity relationships were observed between APN/CD13 activity and growth/survival processes. We showed that compound 2, but not benzopyran derivatives 39, inhibited APN activity (although to a less degree than 1). Both 1 and 2 did not affect AML cell proliferation and survival, indicating that CD13’s APN activity is not required for these processes. In contrast, benzopyran compounds 39 inhibited in a concentration-dependent manner the growth of U937 cells by inducing death as evidenced by phosphatidylserine externalization. Cell death was associated with the presence of geminal nitro group and chlorine at the 3-position of the 2H-1-benzopyran scaffold. The presence of other substituents such as CH2COOH or CH2CH=CH2 groups at the 8-position, NO2 or I substituents at the 2'- or 3'-position, OCH3 or OCH2C6H5 groups at the 4'-position did not affect cell death. Importantly, the inhibitory effects evidenced with compounds 79 were not due to their potential decomposition into the corresponding (Z)-(2chloro-2-nitroethenyl)benzene and salicylaldehyde. Based on these preliminary data, the 3-chloro,3-nitro-2H-1-benzopyran derivatives could be classified as a new group of compounds with promising antitumor properties; this study therefore provides the opportunity to explore their potential efficiency in AML patients’ cells ex vivo.


    The concept of semiring was firstly introduced by Dedekind in 1894, it had been studied by various researchers using techniques coming from semigroup theory or ring theory. The algebraic theories of semirings were widely applied in automata theory, optimization theory, parallel computation systems and the mathematical modeling of quantum physics, etc. [7]

    A semiring (S,+,) is an algebra with two binary operations + and such that the additive reduct (S,+) and the multiplicative reduct (S,) are semigroup connected by ring-like distributive laws, that is,

    a(b+c)=ab+acand(b+c)a=ba+ca, a,b,cS.

    In recent several decades, many authors extended the concepts and results of semigroups to semirings which is one of the development power of semiring theory. For instance, many researchers investigate idempotent semirings in which both additive reduct and multiplicative reduct are bands, which play a role in semirings just as the role of bands in semigroups in many aspects [2,4,17,18,19], etc. Semirings whose additive reduct is a band are also studied by many authors [3,11,25], etc. Karvellas introduced additively inverse semirings whose additive reduct is an inverse semigroup [10]. Zeleznikow studied regular semirings in which both additive and multiplicative semigroups are regular [27]. He also introduced the orthodox semirings firstly [28]. Grillet gave the structure theorem of semirings with a completely simple additive semigroup [6].

    Meanwhile, semirings are generalizations of distributive lattices, b-lattices, rings, skew-rings. Sen, Maity, and Shum extended the concept of Clifford semigroup to semiring by defining a class of semiring which is called Clifford semiring and showed that a semiring S is a Clifford semiring if and only if it is a strong distributive lattice of skew-rings [20]. What's more, as a further generalization, they proved that a semiring S is a generalized Clifford semiring if and only if it is a strong b-lattice of skew-rings. Sen and Maity had also extended completely regular semigroups to completely regular semirings by giving a gross structure theorem: A semiring S is completely regular semiring if and only if it is a b-lattice of completely simple semirings [21]. Pastijn and Guo used inspiration for the study of semirings which are disjoint unions of rings in theory developed for completely regular semiring [26]. Maity and Ghosh also extended completely regular semiring to quasi completely regular semirings, and show that S is quasi completely regular semirings if and only if S is an idempotent semiring of quasi skew-rings [13,14]. Since the ideas of transversals are important to study algebraic structures which is useful in the study of semigroups structure [15,16,23,24]. In 2022, Huang et al. introduced some special semiring transversals into semirings and extended the results of completely regular semirings [9].

    On the other hand, in the regular semigroups with inverse transversals, split orthodox semigroups are not only special regular semigroups with inverse transversals but also one of the origins of inverse transversals. In [12], D. B. McAlister and T. S. Blyth introduced a kind of semigroups which are called split orthodox semigroups and used a band, an inverse semigroup, and Munn morphism to give a structure theorem for them. El-Qallali studied the split quasi-adequate semigroups whose idempotents are commutative and extended the result of split orthodox semigroups [5]. Li extended split orthodox semigroups to split P-regular semigroups [22].

    To develop new ways to study semirings, we will study a kind of semirings called split additively orthodox semirings which have the property like split orthodox semigroups. It is also a class of special semirings with transversals. In this paper, after obtaining some properties theorems of such semirings, we obtain a structure theorem for them by idempotent semirings, additively inverse semirings, and Munn semigroup. Consequently, the corresponding results of Clifford semirings and generalized Clifford semirings in [20], and split orthodox semigroups in [12] are also extended and strengthened.

    For the terminology and notions not given in this paper, the reader is referred to [1,8].

    Firstly, we claim that the following theorem will be frequently used without further mention.

    Theorem 2.1. (Miller-Clifford theorem) [8]

    (1) Let e and f be D-equivalent idempotents of a semigroup S. Then each element a of ReLf has a unique inverse a in RfLe, be such that aa=e and aa=f.

    (2) Let a,b be elements of a semigroup S. Then abRaLb if and only if RbLa contains an idempotent.

    In this section, we will list some elementary results of split bands and split orthodox semigroups. The following results are all due to D. B. McAlister and T. S. Blyth. For convenience, throughout this section the letter D will always denote the Green's relation on a band B.

    Definition 2.1. (Definition 1.1 [12]) Let B={Bα:αY} be a band with structure semilattice Y and D-classes the rectangular bands Bα. If :BB/D is the natural morphism then we shall say that B is split if there is a morphism π:B/DB such that π=idB/D. Such a morphism π will be called a splitting morphism.

    Definition 2.2. (Definition 1.2 [12]) Let B={Bα:αY} be a band. Then by a skeleton of B shall mean a subset E={xα:αY} such that xαBα for every αY and xαxβ=xαβ=xβxα for all α,βY.

    Lemma 2.1. (Lemma 1.3 [12]) A band B is split if and only if it has a skeleton. If π:B/DB is splitting morphism then \rm{Im} π is a skeleton of B.

    We recall that the relation

    γ={(x,y)T×T:V(x)=V(y)}

    on an orthodox semigroup T turns out to be the smallest inverse semigroup congruence on T. Moveover, on B the band of idempotents of T, γ is the same to D.

    Definition 2.3. (Definition 1.4 [12]) Let T be an orthodox semigroup and let :TT/γ be natural morphism. Then we shall say that T is split if there is a morphism π:T/γT such that π=idT/γ.

    Definition 2.4. (Definition 1.5 [12]) Let T be an orthodox semigroup with band of idempotents B. Suppose that E is a D-transversal of B in that E meets every D-class only once. Then we define the span of E by

    Sp(E)={aT:(e,fE)eRaLf}.

    Lemma 2.2. (Lemma 1.6 [12]) Let T be an orthodox semigroup with band of idempotents B. Suppose that E is a D-transversal of B in that E meets every D-class only once. Then Sp (E) meets every γ-class of T exactly once.

    We denote the unique inverse of aSp(E) in Sp(E) by a. Moreover, note that e=aa and f=aa. Then

    Theorem 2.2. (Theorem 1.7 [12]) Let T be an orthodox semigroup with band of idempotents B and suppose that B has a skeleton E. Then the following statements are equivalent:

    (1) There is an inverse subsemigroup S of T that meets every γ-class of T exactly once and has E as semilattice of idempotents;

    (2) aEaE for every a Sp (E);

    (3) Sp (E) is a subsemigroup of T.

    Moreover, if (1) holds, then necessarily S= Sp (E).

    B is a band with a skeleton E. eE. Let θ be a band isomorphism between subbands of B of the form eBe. We say θ is skeleton-preserving if it satisfies that

    fθEfE,fDomθ.

    What is more, as shown in [12], if denote Dom(θ) and Im(θ) by eθBeθ and fθBfθ respectively, then

    eθϕ=(fθeϕ)θ1 and fθϕ=(fθeϕ)ϕ.

    Now, denote by TB the set of skeleton-preserving isomorphisms θ between subbands of B of the form eBe, where eE.

    Lemma 2.3. (Lemma 2.1 [12]) Let B be a band with a skeleton E. Then TB is an inverse semigroup.

    Lemma 2.4. (Lemma 2.2 [12]) Let B be a band with a skeleton E. For θTB, define ˉθ:BB as following:

    bB,bˉθ=(eθbeθ)θ.

    Then ¯θϕ=ˉθˉϕ, θ,ϕTB.

    Given θTB, let θ:eθBeθfθBfθ. It is clear that EDomθ=eθE and ECodθ=fθE. Moreover, θ induces an isomorphism ˆθ:eθEfθE. For every xeθE, we have xˆθ=xθ. The assignment ^:θˆθ is then a morphism from TB to TE (the Munn semigroup of semilattice E).

    Suppose now that S is an inverse semigroup with semilattice of idempotents E, let μ:aμa be a morphism from S to TE.

    If there exists a morphism θ:STB making the following diagram commutative,

    we call it is a triangulation of μ. Denote aθ=θa, then ˆθa=μa, μa:aaEaaE,eaea,eaaE.

    Lemma 2.5. (Lemma 2.3 [12]) Let S be an inverse semigroup with semilattice of idempotents E and let B be a band with skeleton E. For every aS, let the domain and codomain of μa be eaE (so that ea=aa) and faE (so that fa=aa). Let θ be a triangulation of μ. Then given a,bS and e,f,u,vB such that eLea, fRfa, uLeb, vRfb, we have

    e(fu)ˉθaLeab and (fu)ˉθbvRfab.

    Corollary 2.1. (Corollary 2.4 [12]) If

    W=W(B,S,θ)={(e,a,f)B×S×B:eLea,fRfa}

    then the prescription

    (e,a,f)(u,b,v)=(e(fu)ˉθa,ab,(fu)ˉθbv)

    defines a binary operation on W.

    Theorem 2.3. (Throrem 2.5 [12]) W(B,S,θ) is an orthodox semigroup whose band of idempotents is isomorphic to B.

    Theorem 2.4. (Theorem 2.7 [12]) The orthodox semigroup W=W(B,S,θ) is split and W/γS.

    Theorem 2.5. (Theorem 2.8 [12]) Let T be a split orthodox semigroup with band B of idempotents. If π:T/γT is a splitting morphism then the set E=B Imπ of idempotents of Imπ is a skeleton of B and Sp (E)= Imπ. Moreover, if θ: Im πTB is given by aθ=θa, where the domain of θa is aaBaa, the codomain of θa is aaBaa and bθa=aba, then θ is a triangulation of μ: Im πTB and

    TW(B,Imπ,θ).

    Let (S,+,) be a semiring. Then the Green's relation on (S, +) denoted by +R, +L and +H. The set of additive idempotents and the set of multiplicative idempotents of S are denoted by E+(S) and E(S) respectively.

    We also denote the additive inverse of xS by x, (x) by x, and all the additive inverses of x by +V(x) respectively.

    Lemma 3.1. Let (S,+,) be a semiring. x,yS, if x+V(x), then (xy)=xy and (yx)=yx.

    Proof: Clearly.

    In this section, we introduce the concept of split additively orthodox semirings. For convenience, T is denoted the additively orthodox semiring whose additive idempotents forms an idempotent semiring always. The letter +D will always denote the Green's relation on (+E(S),+). Since +D is congruence on (+E(S),+,) and every +D-class is an idempotent semiring for which the additive reduct is a rectangular band. Such idempotent semiring will be called additive rectangular idempotent semiring in this paper.

    T is an additively orthodox semiring, so (T, +) is an orthodox semigroup. Then we have the following corollary.

    Corollary 3.1. The relation

    γ={(x,y)T×T:+V(x)=+V(y)}

    is the smallest additively inverse semiring congruence on T.

    {Proof: }x,y,aT,xγy, then +V(x)=+V(y). (ax)+V(ax), (ax)=ax. Since x+V(x)=+V(y), then

    ax+ay+ax=a(x+y+x)=ax

    and

    ay+ax+ay=a(y+x+y)=ay.

    So +V(ax)+V(ay). Similarly, we can show the other side. Hence, γ is the semiring congruence on T.

    Moreover, γ is an additively inverse semiring congruence on T, since T/γ is an additively inverse semiring.

    Now, let ρ to be a semiring congruence on T such that T/ρ is an additively inverse semiring. For (x,y)γ, let a+V(x)(=+V(y)). Then both xρ and yρ are inverses of aρ in the additively inverse semiring T/ρ. So xρy which implies that γρ.

    Therefore, γ is the smallest additively inverse semiring congruence on T.

    Moveover, on an idempotent semiring I, γ is the same to +D, i.e. I/γ=I/+D which is b-lattice.

    Example 3.1. Let S1={0,a,b} whose additive and multiplicative Cayley tables as following:

    Then (S1,+,) forms an additively inverse semiring.

    Let I1={p,q} and Λ1={u,v} whose additive and multiplicative Cayley tables as following:

    It is easy to verify that (I1,+,) and (Λ1,+,) form idempotent semirings. The additive reduct of (I1,+,) is a left zero band, and its multiplicative reduct is a semilattice. The additive reduct of (Λ1,+,) is a right zero band, and its multiplicative reduct is a semilattice.

    Then the direct product T=I1×S1×Λ1 forms an additively regular semiring, and E+(T)=I1×E+(S1)×Λ1 is an idempotent semiring, hence T is an additively orthodox semiring.

    (i,x,y)T,

    (k,y,l)+V((i,x,j))y+V(x)y=x.

    It implies that

    (i,x,y)γ(k,y,l)y=x,(i,x,y),(k,y,l)T.

    So, T/γ={I1×{x}×Λ1|xS1}S1. Therefore, γ is the smallest additively inverse semiring congruence on T.

    Definition 3.1. Let T be an additively orthodox semiring whose additive idempotents forms an idempotent semiring and :TT/γ be natural morphism. Then we shall say that T is split if there is a morphism π:T/γT such that π=idT/γ.

    Some additively orthodox semirings are neither completely regular semirings nor split additively orthodox semirings, see the following example.

    Example 3.2. Let S1 as shown in Example 1. Then it is also a completely regular semiring. Although the set E+(S1)={0,b} is an ideal of S1, (S1,+,) is not a generalized Clifford semiring [20].

    Let I2={p,q} and Λ2={u,v} whose additive and multiplicative Cayley tables as following:

    We can verify that (I2,+,) and (Λ2,+,) form semirings. The additive reduct of (I2,+,) is a left zero band, and its multiplicative reduct is a group. The additive reduct of (Λ2,+,) is a right zero band, and its multiplicative reduct is a group.

    Then the direct product T=I2×S1×Λ2 forms an additively regular semiring, and E+(T)=I2×E+(S1)×Λ2 forms its subsemiring, hence T is an additively orthodox semiring. But E(T)={p}×E+(S1)×{u}, according to Lemma 2.5 in [21], thus (T,+,) is not a completely regular semiring. Since E+(T) is not an idempotent semiring, T is not a split additively orthodox semiring. Moreover, {(p,x,u)|xS1} is an additively inverse semiring transversal of T [9].

    A split additively orthodox semiring may not be a completely regular semiring, see the following example.

    Example 3.3. Let S2={0,i,a,e,f,x,y} whose additive and multiplicative Cayley tables as following:

    It is easy to verify that (S2,+,) forms an additively inverse semiring, but +Hx={x} is not a skew-ring, so S2 is not a completely regular semiring or a generalized Clifford semiring.

    Let I1 and Λ1 as shown in Example 3.1. Then the direct product T=I1×S2×Λ1 forms a additively regular semiring, and E+(T)=I1×E+(S2)×Λ1 is an idempotent semiring, hence T is an additively orthodox semiring. Additionally, T/γ={I1×{x}×Λ1|xS2}. Let π:T/γ{p}×S2×{u},I1×{x}×Λ1(p,x,u). It is easy to verify that π is a split morphism, so T is a split additively orthodox semiring. Moreover, Imπ is an additively inverse semiring transversal of T. But ImπS2 which means that it is not a generalized Clifford semiring transversal of T. In addition, +H(p,x,u)={(p,x,u)} is not a skew-ring, so T is not a completely regular semiring.

    A split additively orthodox semiring may be also a completely regular semiring, see the following example.

    Example 3.4. Let T as shown in Example 3.1. So T is an additively orthodox semiring, and T/γ={I1×{x}×Λ1|xS1}S1. Let π:T/γ{p}×S1×{u},I1×{x}×Λ1(p,x,u). It is easy to verify that π is a split morphism, so T is a split additively orthodox semiring. What is more, Imπ={(p,x,u)|xS1} is an additively inverse semiring transversal of T. But ImπS1 which means that it is not a generalized Clifford semiring transversal of T. Meanwhile, T is a completely regular semiring.

    Theorem 3.1. Let T be an additively orthodox semiring with a generalized Clifford semiring transversal S. Then the following statements are equivalent:

    (1) T is a completely regular semiring;

    (2) T is a split additively orthodox semiring.

    Proof: (1)(2): T is an additively orthodox semiring which means that E+(T) forms a band. Moreover, E+(T)E(T), since T is a completely regular semiring. Let π:T/γS,γxx,x+V(x)S. It is easy to verify that π is a split morphism.

    (2)(1): By Theorem 4.1 in [9], T is a b-lattice of additively orthodox semirings with skew-ring transversals. For each additively orthodox semiring with skew-ring transversals Sα, Sα is an additively completely simple semiring by Theorem 3.1 in [9]. Moreover, E+(T)E(T). Hence, Sα is a completely simple semiring. Therefore, T is a b-lattice of completely simple semiring which means that T is a completely regular semiring.

    By Definition 2.1 and Definition 3.1, if there is a split morphism π on the idempotent semiring I, then π is a split morphism on band (I,+).

    Definition 3.2. Let I={Iα:αY} be an idempotent semiring, where Y is a b-lattice. Then by a skeleton of I shall mean a subset E={xα:αY} such that xαIα for every αY, and xα+xβ=xα+β=xβ+xα and xαxβ=xαβ for all α,βY.

    Example 3.5. Let Y={0,e} whose additive and multiplicative Cayley tables as following:

    It is easy to verify that (Y,+,) forms a b-lattice. Let I1 and Λ1 as shown in Example 3.1. Then the direct product I=I1×Y×Λ1 forms an idempotent semiring.

    Additionally, I/γ={I1×{x}×Λ1|xY}Y. Denote I1×{x}×Λ1 by γx, then I=γ0γe.

    Let E={(p,0,u),(p,e,u)}. Then (p,0,u)γ0 and (p,e,u)γe. Moreover,

    (p,0,u)+(p,e,u)=(p,e,u)=(p,e,u)+(p,0,u),

    and

    (p,0,u)(p,e,u)=(p,0e,u)=(p,0,u),

    where (p,0,u)γ0e

    Therefore, E is a skeleton of I.

    By Definition 2.2 and Definition 3.2, we easily get that if E is a skeleton of I then (E, +) is a skeleton of (I, +).

    Lemma 3.2. An idempotent semiring I is split if and only if it has a skeleton. If π:I/+DI is splitting morphism then Imπ is a skeleton of I.

    Proof: By Lemma 2.1, Definition 3.1, and Definition 3.2, it is clear.

    From now on, we explore the structure of split additively orthodox semirings.

    Lemma 4.1. Let T be a split additively orthodox semiring whose additive idempotents forms an idempotent semiring I. Then E=I Imπ is a skeleton of I.

    Proof: Imπ is an additively inverse subsemiring of T, and meets every γ-class of T exactly once. As a result, E=IImπ is the set of all the additive idempotents of Imπ forms b-lattice, and E meets every +D-class of I exactly once, hence it is a skeleton of I.

    The span of E is also the important set:

    Sp(E)={aT:(e,fE)e+Ra+Lf}.

    By Lemma 2.2, the following lemma is obtained directly.

    Lemma 4.2. Let T be a split additively orthodox semiring whose additive idempotents forms an idempotent semiring I. Then Sp (E) meets every γ-class of T exactly once.

    Noting that for every aSp(E), exist e,fE satisfy that e+Ra+Lf, then +Le+Rf contains an additive inverse a of a, and aSp(E). By Lemma 4.2, we get that a is the unique inverse of a in Sp(E), which is denoted by a hereafter. Moreover, e=a+a and f=a+a. Actually, we can get more from the following theorem.

    Theorem 4.1. Let T be an additively orthodox semiring with an idempotent semiring of additive idempotent I and suppose that I has a skeleton E. Then the following statements are equivalent:

    (1) There is an additively inverse subsemiring S of T that meets every γ-class of T exactly once and has E as b-lattice of additive idempotents;

    (2) a+E+aE for every a\rm{Sp}(E);

    (3) Sp (E) is a subsemiring of T.

    Proof: (1)(2): By Theorem 2.2, it is clear.

    (2)(3): By Theorem 2.2, Sp(E) is close under addition. Now we show that it is close under multiplication. Given a,bSp(E), there exist ea,eb,fa,fbE such that ea+Ra+Lfa, and eb+Ra+Lfb. Since +R and +L are multiplicative congruence on S, then eaeb+Rab+Lfafb. Since eaeb,fafbE, it follows that Sp(E) is a subsemiring of T.

    (3)(1): If Sp(E) is a subsemiring, by the remarks following Lemma 9, Sp(E) is an additively orthodox subsemiring which meets every γ-class exactly once and has E as its set of additive idempotents.

    By Lemma 2.2, the following corollary is obtained directly.

    Corollary 4.1. Let T be a split additively orthodox semiring whose additive idempotents forms an idempotent semiring I. Then Sp (E)= Im π.

    For every aImπ, let ea=a+a, fa=a+a. Define θa:ea+I+eafa+I+fa by xθa=a+x+a. According to the proof of Theorem 2.5 (i.e. Theorem 2.8 in [12]), θa is a skeleton-preserving isomorphism from (ea+I+ea,+) to (fa+I+fa,+) satisfies that

    (fI)fθEfE.

    Denote all the skeleton-preserving isomorphisms φ between the subbands such as e+I+e of (I, +) by +TI, where eI. Then define the addition on +TI as composition of maps, i.e. e(φ+ϕ)(eφ)ϕ. By Lemma 2.3, we get that (+TI,+) is an inverse semigroup.

    Lemma 4.3. Let T be a split additively orthodox semiring whose additive idempotents forms an idempotent semiring I. E=I Imπ. TE is the Munn semigroup of (E,+). For μaTE, define θ: Imπ+TI by aθ=θa, where the domain of θa is a+a+I+a+a, and the codomain is a+a+I+a+a. Moreover, bθa=a+b+a, then θ is a triangulation of μ: ImπTE.

    Proof: According to the proof of Theorem 2.3 (i.e. Theorem 2.5 in [12]), the mapping θ:Imπ+TI defined by aθ=θa is a morphism from (Imπ,+) to (+TI,+). By the definition of θa, we obtain that ˆθa=μa, where μaTE. TE is the Munn semigroup of (E,+), and define eμa=a+e+a which is an isomorphism from (a+a+E,+) to (a+a+E,+). As a result, θ is a triangulation of μ:ImπTE.

    Moreover, we can extend each θ+TI to a mapping ˉθ:II by defining

    (bB)bˉθ=(eθ+b+eθ)θ.

    And define the addition between two maps as composition of maps. By Lemma 4, we get that

    (θ,ϕTB)¯θ+ϕ=ˉθ+ˉϕ.

    Theorem 4.2. Let S be an additively inverse semiring with b-lattice of additive idempotents E and I be an idempotent semiring with skeleton E. For every aS, let the domain and codomain of μaTE be ea+E (so that ea=a+a) and fa+E (so that fa=a+a). Let θ be a triangulation of μ. Then given a,bS and e,f,u,vI such that e+Lea, f+Rfa, u+Leb, v+Rfb, we have

    (1) (e+(f+u)ˉθa)+Lea+b and ((f+u)ˉθb+v)+Rfa+b;

    (2) ab+H(ab) and eab=fab=eafb=faeb=eaeb=fafb;

    (3) eu+Leab and fv+Rfab.

    Proof: (1) By Lemma 2.5, it is clear.

    (2) Since ea+La+Rfa and fb+Lb+Reb, then eafb+Lab+Rfaeb. Similarly, we can get that eafb+Rab+Lfaeb. By Lemma 6, ab=(ab)=ab. So eafb+L(ab)+Lfaeb and eafb+R(ab)+Rfaeb, that is eafb+H(ab)+Hfaeb. Similarly, we get that eaeb+Hab+Hfafb. Consequently, ab+H(ab) and eab=fab=eafb=faeb=eaeb=fafb.

    (3) Since e+Lea and u+Leb, then eu+Leaeb=eab. Similarly, fv+Rfab.

    Theorem 4.3. Let T be a split additively orthodox semiring whose additive idempotents forms an idempotent semiring I. For any a,b Imπ and e,f,u,vI such that e+Lea, f+Rfa, u+Leb, v+Rfb, then

    (e+a+f)(u+b+v)=eu+ab+fv.

    Proof: Firstly, on the one hand, since f+L(e+a+f)+Re and v+L(u+b+v)+Ru, then fv+L(e+a+f)(u+b+v)+Reu. On the other hand, since e+Lea, f+Rfa, u+Leb and v+Rfb, then eu+Leab and fv+Rfab, so fv+L(eu+ab+fv)+Reu. Hence, (e+a+f)(u+b+v)+H(eu+ab+fv).

    Secondly, since e+Lea+La and f+Rfa+Ra, then

    e+a+f+a+e+a+f=e+a+a+a+f=e+a+f

    and

    a+e+a+f+a=a+a+a=a,

    so a+V(e+a+f), i.e. a is the unique additive inverse of e+a+f in Imπ. Similarly, (u+b+v)=b and (eu+ab+fv)=(ab). Thus,

    ((e+a+f)(u+b+v))=(e+a+f)(u+b+v)=(e+a+f)b=(e+a+f)b=ab=(ab).

    Consequently, (e+a+f)(u+b+v)=eu+ab+fv as required.

    Theorem 4.4. Let T be a split additively orthodox semiring whose additive idempotents forms an idempotent semiring I. For any a,b,c Imπ and e,f,u,v,g,hI such that e+Lea, f+Rfa, u+Leb, v+Rfb, g+Lec, h+Rfc, then satisfies the following four equations:

    ge+g(f+u)ˉθa=ge+(hf+gu)ˉθca,
    h(f+u)ˉθb+hv=(hf+gu)ˉθcb+hv,
    eg+(f+u)ˉθag=eg+(fh+ug)ˉθaoc,

    and

    (f+u)ˉθbh+vh=(fh+ug)ˉθbc+vh.

    Proof: For any a,b,cImπ and e,f,u,v,g,hI such that e+Lea, f+Rfa, u+Leb, v+Rfb, g+Lec, h+Rfc. On the one hand, by Theorem 4.2, we get that

    (e+(f+u)ˉθa)+Lea+b and ((f+u)ˉθb+v)+Rfa+b,

    then

    g(e+(f+u)ˉθa)+Lecea+b and h((f+u)ˉθb+v)+Rfcfa+b,

    that is

    (ge+g(f+u)ˉθa)+Lec(a+b) and (h(f+u)ˉθb+hv)+Rfc(a+b),

    where

    ec(a+b)=c(a+b)+(c(a+b))=c(a+b)+c(a+b)

    and

    fc(a+b)=(c(a+b))+c(a+b)=c(a+b)+c(a+b).

    So,

    ge+g(f+u)ˉθa=ge+g(f+u)ˉθa+ec(a+b)=ge+g(f+u)ˉθa+c(a+b)+c(a+b)=ge+g(f+u)ˉθa+c(a+b)+fc(a+b)+c(a+b)=ge+g(f+u)ˉθa+c(a+b)+h(f+u)ˉθb+hv+fc(a+b)+c(a+b)=ge+g(f+u)ˉθa+c(a+b)+h(f+u)ˉθb+hv+c(a+b)=(g+c+h)(e+(f+u)ˉθa+a+b+(f+u)ˉθb+v)+c(a+b)=(g+c+h)(e+(fa+f+u+fa)θa+a+b+(eb+f+u+eb)θb+v)+c(a+b)=(g+c+h)(e+(a+fa+f+u+fa+a)+a+b+(b+eb+f+u+eb+b)+v)+c(a+b)=(g+c+h)(e+(a+f+u+a)+a+b+(b+f+u+b)+v)+c(a+b)=(g+c+h)(e+a+f+u+(a+a)+(b+b)+f+u+b+v)+c(a+b)=(g+c+h)(e+a+f+u+(b+b)+(a+a)+f+u+b+v)+c(a+b)=(g+c+h)(e+a+f+u+f+u+b+v)+c(a+b)=(g+c+h)(e+a+f+u+b+v)+c(a+b).

    On the other hand, since +L and +R are multiplicative congruence on T, then

    ge+Lecea=eca,
    hf+Rfcfa=fca,
    gu+Leceb=ecb,
    hv+Rfcfb=fcb.

    By Theorem 4.2, we get that

    (ge+(hf+gu)ˉθ(ca))+Leca+cb and ((hf+gu)ˉθcb+hv)+Rfca+cb,

    that is

    (ge+(hf+gu)ˉθca)+Leca+cb and ((hf+gu)ˉθcb+hv)+Rfca+cb,

    where

    eca+cb=ca+cb+(ca+cb)=ca+cb+(c(a+b))=c(a+b)+c(a+b)

    and

    fca+cb=(ca+cb)+c(a+b)=(c(a+b))+c(a+b)=c(a+b)+c(a+b).

    So,

    ge+(hf+gu)ˉθ(ca)=ge+(hf+gu)ˉθ(ca)+eca+cb=ge+(hf+gu)ˉθ(ca)+ca+cb+(ca+cb)=ge+(hf+gu)ˉθ(ca)+ca+cb+fca+cb+(ca+cb)=ge+(hf+gu)ˉθ(ca)+ca+cb+(hf+gu)ˉθcb+hv+fca+cb+(ca+cb)=ge+(fca+hf+gu+fca)θ(ca)+ca+cb+(ecb+hf+gu+ecb)θcb+hv+(ca+cb)=ge+(ca+hf+gu+(ca))+ca+cb+(cb+hf+gu+cb)+hv+(ca+cb)=ge+ca+hf+gu+((ca)+ca)+(cb+(cb))+hf+gu+cb+hv+(ca+cb)=ge+ca+hf+gu+(cb+(cb))+((ca)+ca)+hf+gu+cb+hv+(ca+cb)=ge+ca+hf+gu+ecb+fca+hf+gu+cb+hv+(ca+cb)=ge+ca+hf+gu+hf+gu+cb+hv+(ca+cb)=ge+ca+hf+gu+cb+hv+(ca+cb)=(g+c+h)(e+a+f)+(g+c+h)(u+b+v)+(ca+cb)=(g+c+h)[(e+a+f)+(u+b+v)]+(ca+cb)=(g+c+h)(e+a+f+u+b+v)+(ca+cb).

    Therefore,

    ge+g(f+u)ˉθa=ge+(hf+gu)ˉθca

    as required. Similarly, we can show the following three equations:

    h(f+u)ˉθb+hv=(hf+gu)ˉθcb+hv,
    eg+(f+u)ˉθag=eg+(fh+ug)ˉθaoc,

    and

    (f+u)ˉθbh+vh=(fh+ug)ˉθbc+vh.

    Theorem 4.5. Let S be an additively inverse semiring with b-lattice of additive idempotents E and I be an idempotent semiring with skeleton E. For every aS, let the domain and codomain of μaTE be ea+E (so that ea=a+a) and fa+E (so that fa=a+a). Let θ be a triangulation of μ. Then given a,bS and e,f,u,vI such that e+Lea, f+Rfa, u+Leb, v+Rfb, g+Lec, h+Rfc, and satisfies the following four equations:

    ge+g(f+u)ˉθa=ge+(hf+gu)ˉθca,
    h(f+u)ˉθb+hv=(hf+gu)ˉθcb+hv,
    eg+(f+u)ˉθag=eg+(fh+ug)ˉθaoc,

    and

    (f+u)ˉθbh+vh=(fh+ug)ˉθbc+vh.

    Define two binary operations

    (e,a,f)+(u,b,v)=(e+(f+u)ˉθa,a+b,(f+u)ˉθb+v)

    and

    (e,a,f)(u,b,v)=(eu,ab,fv)

    on

    W=W(I,S,θ)={(e,a,f)I×S×I:e+Lea,f+Rfa}.

    Then W is a split additively orthodox semiring whose additive idempotents form an idempotent semiring which is isomorphic to I, and W/γS.

    Conversely, every split additively orthodox semiring is of the form W(I,S,θ).

    Proof: By Theorem 2.3, (W,+) is an orthodox semigroup. And the associativity of multiplication is clear. We only need to prove the distributivity of the semiring W. Give (e,a,f),(u,b,v),(g,c,h)W, by the four equations,

    (g,c,h)[(e,a,f)+(u,b,v)]=(g,c,h)(e+(f+u)ˉθa,a+b,(f+u)ˉθb+v)=(g(e+(f+u)ˉθa),c(a+b),h((f+u)ˉθb+v))=(ge+g(f+u)ˉθa,ca+cb,h(f+u)ˉθb+hv)=(ge+(hf+gu)ˉθca,ca+cb,(hf+gu)ˉθcb+hv)=(ge,ca,hf)+(gu,cb,hv).

    Thus the distributivity on left is hold. And the distributivity on right can be proved similarly. Hence W is additively orthodox semiring as required.

    By the proof of Theorem 2.3 (i.e. Theorem 2.5 in [12]), we get that +E(W)={(e,a,f)W:a+E(S)}, moreover, the map ϕ defined from +E(W) to I by (e,a,f)ϕ=e+f is bijective and preserves addition. So we need to show it also preserves multiplication well. For any (e,a,f),(u,b,v)+E(W), since a,b+E(S) then a=a and b=b, we find that ea=a+a=a+a=a=a+a=fa and eb=b+b=b+b=b=b+b=fb, so e+La+Rf and u+Lb+Rv. On the one hand, by Theorem 2.1, we get that e+R(e+f)+Lf and u+R(u+v)+Lv, so eu+R(e+f)(u+v)+Lfv since +R and +L are both multiplicative congruence on I. On the other hand, we get that eu+Lab+Rfv, then eu+R(eu+fv)+Lfv by Theorem 2.1. Therefore, (e+f)(u+v)+H(eu+fv) which means that (e+f)(u+v)=eu+fv. Hence,

    [(e,a,f)(u,b,v)]ϕ=(eu,ab,fv)ϕ=eu+fv=(e+f)(u+v)=(e,a,f)ϕ(u,b,v)ϕ.

    Thus ϕ is an isomorphism as require.

    By the proof of Theorem 2.3 (i.e. Theorem 2.5 in [12]), we get that

    (e,a,f)γ(u,b,v)a=ba=b,

    and the mapping π:W/γW given by γ(e,a,f)π=(ea,a,fa) preserves addition and satisfies that π=idW/γ. Now we will show that it also preserves multiplication:

    γ(e,a,f)πγ(u,b,v)π=(ea,a,fa)(eb,b,fb)=(eaeb,ab,fafb)=(eab,ab,fab)=γ(e,a,f)(u,b,v)π.

    Finally, it is clear that

    W/γImπS,

    the second isomorphism being that given by (ea,a,fa)a.

    Conversely, let T be a split additively orthodox semiring with the idempotent semiring I of additive idempotents, and mapping π:T/γT be a splitting morphism. By Lemma 4.1 and Corollary 4.1, the set E=IImπ of additive idempotents of Imπ is a skeleton of I and Sp(E) = Imπ.

    Moreover, define θ: ImπTI by aθ=θa, where the domain of θa is a+a+I+a+a, the codomain of θa is a+a+I+a+a and bθa=a+b+a. By Lemma 4.3, θ is a triangulation of μ:ImπTE. By Theorem 4.4, W(I,Imπ,θ) satisfies the four equations. We can therefore construct the split additively orthodox semiring W=W(I,Imπ,θ).

    Define the map ψ:WT by

    (e,a,f)ψ=e+a+f.

    By the proof shown in [12]. ψ is bijection and preserves addition.

    Since for any a,bImπ and e,f,u,vI such that e+Lea, f+Rfa, u+Leb, v+Rfb, then

    (e+a+f)(u+b+v)=eu+ab+fv.

    So it also preserves multiplication clearly.

    Therefore, ψ is a semiring isomorphism.

    As shown in [20], a generalized Clifford semiring S is not only an additively orthodox semiring, but also a strong b-lattice T of skew-rings Rα(αT), i.e. S=<T,Rα,ϕα,β>. Hence, S/γ=T. Let π:T/γS,αxRα. It is easy to verify that π is a split morphism, so S is a split additively orthodox semiring.

    Remark 1. From Theorem 4.5, we can see that the class of split additively orthodox semiringsis actually not only a general extension of the class of Clifford semirings and generalized Clifford semirings studied in [20], but also a general extension of the class of split orthodox semigroups in [12].

    In this paper, we introduce and explore split additively orthodox semirings. Some property theorems are obtained, and a structure theorem is established by using idempotent semirings, additively inverse semirings, and Munn semigroup. It not only extends and strengthens the corresponding results of Clifford semirings and split orthodox semigroups but also develops a new way to study semirings.

    This work was supported in part by the Dongguan Science and Technology of Social Development Program (2022), in part by the NNSF of China (11801239, 12171022).

    All authors declare no conflicts of interest in this paper.

    [1] King ME, Rowe JM (2007) Recent developments in acute myelogenous leukemia therapy. Oncologist 12 Suppl 2: 14-21.
    [2] Robak T, Wierzbowska A (2009) Current and emerging therapies for acute myeloid leukemia. Clin Ther 31 Pt 2: 2349-2370.
    [3] Steele VE, Boone CW, Dauzonne D, et al. (2002) Correlation between electron-donating ability of a series of 3-nitroflavones and their efficacy to inhibit the onset and progression of aberrant crypt foci in the rat colon. Cancer Res 62: 6506-6509.
    [4] Cardenas M, Marder M, Blank VC, et al. (2006) Antitumor activity of some natural flavonoids and synthetic derivatives on various human and murine cancer cell lines. Bioorg Med Chem 14: 2966-2971. doi: 10.1016/j.bmc.2005.12.021
    [5] Li Y, Fang H, Xu W (2007) Recent advance in the research of flavonoids as anticancer agents. Mini Rev Med Chem 7: 663-678. doi: 10.2174/138955707781024463
    [6] Singh M, Kaur M, Silakari O (2014) Flavones: an important scaffold for medicinal chemistry. Eur J Med Chem 84: 206-239. doi: 10.1016/j.ejmech.2014.07.013
    [7] Chabot GG, Touil YS, Pham MH, et al. (2010) Flavonoids in cancer prevention and therapy: chemistry, pharamcology, mechanisms of action, and perspectives for cancer drugdiscovery. In: Moulay A, editor. Alternative and complementary therapies for cancer. Springer US, 583-612.
    [8] Hou DX, Kumamoto T (2010) Flavonoids as protein kinase inhibitors for cancer chemoprevention: direct binding and molecular modeling. Antioxid Redox Signal 13: 691-719. doi: 10.1089/ars.2009.2816
    [9] Ravishankar D, Rajora AK, Greco F, et al. (2013) Flavonoids as prospective compounds for anti-cancer therapy. Int J Biochem Cell Biol 45: 2821-2831. doi: 10.1016/j.biocel.2013.10.004
    [10] Li-Weber M (2009) New therapeutic aspects of flavones: the anticancer properties of Scutellaria and its main active constituents Wogonin, Baicalein and Baicalin. Cancer Treat Rev 35: 57-68. doi: 10.1016/j.ctrv.2008.09.005
    [11] Liesveld JL, Abboud CN, Lu C, et al. (2003) Flavonoid effects on normal and leukemic cells. Leuk Res 27: 517-527. doi: 10.1016/S0145-2126(02)00265-5
    [12] Newcomb EW (2004) Flavopiridol: pleiotropic biological effects enhance its anti-cancer activity. Anticancer Drugs 15: 411-419. doi: 10.1097/01.cad.0000127332.06439.47
    [13] Cheng S, Gao N, Zhang Z, et al. (2010) Quercetin induces tumor-selective apoptosis through downregulation of Mcl-1 and activation of Bax. Clin Cancer Res 16: 5679-5691. doi: 10.1158/1078-0432.CCR-10-1565
    [14] Fathi AT, Karp JE (2009) New agents in acute myeloid leukemia: beyond cytarabine and anthracyclines. Curr Oncol Rep 11: 346-352. doi: 10.1007/s11912-009-0047-x
    [15] Blum W, Phelps MA, Klisovic RB, et al. (2010) Phase I clinical and pharmacokinetic study of a novel schedule of flavopiridol in relapsed or refractory acute leukemias. Haematologica 95: 1098-1105. doi: 10.3324/haematol.2009.017103
    [16] Karp JE, Smith BD, Resar LS, et al. (2011) Phase 1 and pharmacokinetic study of bolus-infusion flavopiridol followed by cytosine arabinoside and mitoxantrone for acute leukemias. Blood 117: 3302-3310. doi: 10.1182/blood-2010-09-310862
    [17] Zeidner JF, Foster MC, Blackford AL, et al. (2015) Randomized multicenter phase II study of flavopiridol (alvocidib), cytarabine, and mitoxantrone (FLAM) versus cytarabine/daunorubicin (7+3) in newly diagnosed acute myeloid leukemia. Haematologica 100: 1172-1179. doi: 10.3324/haematol.2015.125849
    [18] Bauvois B, Puiffe ML, Bongui JB, et al. (2003) Synthesis and biological evaluation of novel flavone-8-acetic acid derivatives as reversible inhibitors of aminopeptidase N/CD13. J Med Chem 46: 3900-3913. doi: 10.1021/jm021109f
    [19] Quiney C, Dauzonne D, Kern C, et al. (2004) Flavones and polyphenols inhibit the NO pathway during apoptosis of leukemia B-cells. Leuk Res 28: 851-861. doi: 10.1016/j.leukres.2003.12.003
    [20] Piedfer M, Bouchet S, Tang R, et al. (2013) p70S6 kinase is a target of the novel proteasome inhibitor 3,3'-diamino-4'-methoxyflavone during apoptosis in human myeloid tumor cells. Biochim Biophys Acta 1833: 1316-1328. doi: 10.1016/j.bbamcr.2013.02.016
    [21] Bauvois B, Dauzonne D (2006) Aminopeptidase-N/CD13 (EC 3.4.11.2) inhibitors: chemistry, biological evaluations, and therapeutic prospects. Med Res Rev 26: 88-130.
    [22] Bouchet S, Tang R, Fava F, et al. (2016) The CNGRC-GG-D(KLAKLAK)2 peptide induces a caspase-independent, Ca2+-dependent death in human leukemic myeloid cells by targeting surface aminopeptidase N/CD13. Oncotarget 7: 19445-19467.
    [23] Klobusicka M, Kusenda J, Babusikova O (2005) Myeloid enzymes profile related to the immunophenotypic characteristics of blast cells from patients with acute myeloid leukemia (AML) at diagnosis. Neoplasma 52: 211-218.
    [24] Taussig DC, Pearce DJ, Simpson C, et al. (2005) Hematopoietic stem cells express multiple myeloid markers: implications for the origin and targeted therapy of acute myeloid leukemia. Blood 106: 4086-4092. doi: 10.1182/blood-2005-03-1072
    [25] Piedfer M, Dauzonne D, Tang R, et al. (2011) Aminopeptidase-N/CD13 is a potential proapoptotic target in human myeloid tumor cells. Faseb J 25: 2831-2842. doi: 10.1096/fj.11-181396
    [26] Wickstrom M, Larsson R, Nygren P, et al. (2011) Aminopeptidase N (CD13) as a target for cancer chemotherapy. Cancer Sci 102: 501-508. doi: 10.1111/j.1349-7006.2010.01826.x
    [27] Dauzonne D, Folléas B, Martinez L, et al. (1997) Synthesis and in vitro cytotoxicity of a series of 3-aminoflavones. . Eur J Med Chem 32: 71-82. doi: 10.1016/S0223-5234(97)84363-2
    [28] Dauzonne D, Demerseman P (1990) A convenient synthesis of 3-chloro-3,4-dihydro-4-hydroxy-3-nitro-2-p henyl-2H-1-benzopyrans. Synthesis 1: 66-70.
    [29] Dauzonne D, Grandjean C (1992) Synthesis of 2-Aryl-3-nitro-4H-1-benzopyran-4-ones. Synthesis 7: 677-680.
    [30] Pham MH, Auzeil N, Regazzetti A, et al. (2007) Identification of new flavone-8-acetic acid metabolites using mouse microsomes and comparison with human microsomes. Drug Metab Dispos 35: 2023-2034. doi: 10.1124/dmd.107.017012
    [31] Ceccaldi A, Rajavelu A, Champion C, et al. (2011) C5-DNA methyltransferase inhibitors: from screening to effects on zebrafish embryo development. Chembiochem 12: 1337-1345. doi: 10.1002/cbic.201100130
    [32] Lanotte M, Martin-Thouvenin V, Najman S, et al. (1991) NB4, a maturation inducible cell line with t(15;17) marker isolated from a human acute promyelocytic leukemia (M3). Blood 77: 1080-1086.
    [33] Laouar A, Wietzerbin J, Bauvois B (1993) Divergent regulation of cell surface protease expression in HL-60 cells differentiated into macrophages with granulocyte macrophage colony stimulating factor or neutrophils with retinoic acid. Int Immunol 5: 965-973. doi: 10.1093/intimm/5.8.965
    [34] Laouar A, Villiers C, Sanceau J, et al. (1993) Inactivation of interleukin-6 in vitro by monoblastic U937 cell plasma membranes involves both protease and peptidyl-transferase activities. Eur J Biochem 215: 825-831. doi: 10.1111/j.1432-1033.1993.tb18098.x
    [35] Broker LE, Kruyt FA, Giaccone G (2005) Cell death independent of caspases: a review. Clin Cancer Res 11: 3155-3162. doi: 10.1158/1078-0432.CCR-04-2223
    [36] Wang ZB, Liu YQ, Cui YF (2005) Pathways to caspase activation. Cell Biol Int 29: 489-496. doi: 10.1016/j.cellbi.2005.04.001
    [37] Antczak C, De Meester I, Bauvois B (2001) Transmembrane proteases as disease markers and targets for therapy. J Biol Regul Homeost Agents 15: 130-139.
    [38] Bauvois B (2001) Transmembrane proteases in focus: diversity and redundancy? J Leukoc Biol 70: 11-17.
    [39] Bauvois B (2004) Transmembrane proteases in cell growth and invasion: new contributors to angiogenesis? Oncogene 23: 317-329. doi: 10.1038/sj.onc.1207124
    [40] Mina-Osorio P (2008) The moonlighting enzyme CD13: old and new functions to target. Trends Mol Med 14: 361-371. doi: 10.1016/j.molmed.2008.06.003
    [41] Antczak C, De Meester I, Bauvois B (2001) Ectopeptidases in pathophysiology. Bioessays 23: 251-260.
    [42] Grujic M, Renko M (2002) Aminopeptidase inhibitors bestatin and actinonin inhibit cell proliferation of myeloma cells predominantly by intracellular interactions. Cancer Lett 182: 113-119. doi: 10.1016/S0304-3835(02)00086-1
    [43] Winnicka B, O'Conor C, Schacke W, et al. (2010) CD13 is dispensable for normal hematopoiesis and myeloid cell functions in the mouse. J Leukoc Biol 88: 347-359. doi: 10.1189/jlb.0210065
    [44] Scaffidi C, Schmitz I, Zha J, et al. (1999) Differential modulation of apoptosis sensitivity in CD95 type I and type II cells. J Biol Chem 274: 22532-22538. doi: 10.1074/jbc.274.32.22532
    [45] Galluzzi L, Vitale I, Abrams JM, et al. (2012) Molecular definitions of cell death subroutines: recommendations of the Nomenclature Committee on Cell Death 2012. Cell Death Differ 19: 107-120. doi: 10.1038/cdd.2011.96
    [46] Galluzzi L, Kepp O, Krautwald S, et al. (2014) Molecular mechanisms of regulated necrosis. Semin Cell Dev Biol 35: 24-32. doi: 10.1016/j.semcdb.2014.02.006
    [47] Pasparakis M, Vandenabeele P (2015) Necroptosis and its role in inflammation. Nature 517: 311-320. doi: 10.1038/nature14191
    [48] Baritaud M, Boujrad H, Lorenzo HK, et al. (2010) Histone H2AX: The missing link in AIF-mediated caspase-independent programmed necrosis. Cell Cycle 9: 3166-3173. doi: 10.4161/cc.9.16.12887
    [49] Vanden Berghe T, Linkermann A, Jouan-Lanhouet S, et al. (2014) Regulated necrosis: the expanding network of non-apoptotic cell death pathways. Nat Rev Mol Cell Biol 15: 135-147.
    [50] Bonora M, Wieckowski MR, Chinopoulos C, et al. (2015) Molecular mechanisms of cell death: central implication of ATP synthase in mitochondrial permeability transition. Oncogene 34: 1475-1486. doi: 10.1038/onc.2014.96
    [51] Yuan Z, Long C, Junming T, et al. (2012) Quercetin-induced apoptosis of HL-60 cells by reducing PI3K/Akt. Mol Biol Rep 39: 7785-7793. doi: 10.1007/s11033-012-1621-0
    [52] Lee WJ, Hsiao M, Chang JL, et al. (2015) Quercetin induces mitochondrial-derived apoptosis via reactive oxygen species-mediated ERK activation in HL-60 leukemia cells and xenograft. Arch Toxicol 89: 1103-1117. doi: 10.1007/s00204-014-1300-0
    [53] Chen YC, Shen SC, Lee WR, et al. (2002) Wogonin and fisetin induction of apoptosis through activation of caspase 3 cascade and alternative expression of p21 protein in hepatocellular carcinoma cells SK-HEP-1. Arch Toxicol 76: 351-359. doi: 10.1007/s00204-002-0346-6
    [54] Hu C, Xu M, Qin R, et al. (2015) Wogonin induces apoptosis and endoplasmic reticulum stress in HL-60 leukemia cells through inhibition of the PI3K-AKT signaling pathway. Oncol Rep 33: 3146-3154.
    [55] Hsiao PC, Lee WJ, Yang SF, et al. (2014) Nobiletin suppresses the proliferation and induces apoptosis involving MAPKs and caspase-8/-9/-3 signals in human acute myeloid leukemia cells. Tumour Biol 35: 11903-11911. doi: 10.1007/s13277-014-2457-0
    [56] Budhraja A, Gao N, Zhang Z, et al. (2012) Apigenin induces apoptosis in human leukemia cells and exhibits anti-leukemic activity in vivo. Mol Cancer Ther 11: 132-142. doi: 10.1158/1535-7163.MCT-11-0343
    [57] Ruela-de-Sousa RR, Fuhler GM, Blom N, et al. (2010) Cytotoxicity of apigenin on leukemia cell lines: implications for prevention and therapy. Cell Death Dis 1: e19. doi: 10.1038/cddis.2009.18
    [58] Park C, Lee WS, Go SI, et al. (2014) Morin, a flavonoid from moraceae, induces apoptosis by induction of BAD protein in human leukemic cells. Int J Mol Sci 16: 645-659. doi: 10.3390/ijms16010645
    [59] Gao H, Liu Y, Li K, et al. (2016) Hispidulin induces mitochondrial apoptosis in acute myeloid leukemia cells by targeting extracellular matrix metalloproteinase inducer. Am J Transl Res 8: 1115-1132.
    [60] Cardenas MG, Blank VC, Marder MN, et al. (2012) 2'-Nitroflavone induces apoptosis and modulates mitogen-activated protein kinase pathways in human leukaemia cells. Anticancer Drugs 23: 815-826. doi: 10.1097/CAD.0b013e328353f947
    [61] Chang H, Lin H, Yi L, et al. (2010) 3,6-Dihydroxyflavone induces apoptosis in leukemia HL-60 cell via reactive oxygen species-mediated p38 MAPK/JNK pathway. Eur J Pharmacol 648: 31-38. doi: 10.1016/j.ejphar.2010.08.020
    [62] Rosato RR, Dai Y, Almenara JA, et al. (2004) Potent antileukemic interactions between flavopiridol and TRAIL/Apo2L involve flavopiridol-mediated XIAP downregulation. Leukemia 18: 1780-1788. doi: 10.1038/sj.leu.2403491
    [63] Delmulle L, Vanden Berghe T, Keukeleire DD, et al. (2008) Treatment of PC-3 and DU145 prostate cancer cells by prenylflavonoids from hop (Humulus lupulus L.) induces a caspase-independent form of cell death. Phytother Res 22: 197-203.
    [64] Wu PP, Kuo SC, Huang WW, et al. (2009) (-)-Epigallocatechin gallate induced apoptosis in human adrenal cancer NCI-H295 cells through caspase-dependent and caspase-independent pathway. Anticancer Res 29: 1435-1442.
    [65] Zhang Y, Yang ND, Zhou F, et al. (2012) (-)-Epigallocatechin-3-gallate induces non-apoptotic cell death in human cancer cells via ROS-mediated lysosomal membrane permeabilization. PLoS One 7: e46749. doi: 10.1371/journal.pone.0046749
    [66] Wang G, Wang JJ, Yang GY, et al. (2012) Effects of quercetin nanoliposomes on C6 glioma cells through induction of type III programmed cell death. Int J Nanomedicine 7: 271-280. doi: 10.2217/nnm.11.186
    [67] Liao H, Bao X, Zhu J, et al. (2015) O-Alkylated derivatives of quercetin induce apoptosis of MCF-7 cells via a caspase-independent mitochondrial pathway. Chem Biol Interact 242: 91-98. doi: 10.1016/j.cbi.2015.09.022
    [68] Lindsay CK, Gomez DE, Thorgeirsson UP (1996) Effect of flavone acetic acid on endothelial cell proliferation: evidence for antiangiogenic properties. Anticancer Res 16: 425-431.
    [69] Granci V, Dupertuis YM, Pichard C (2010) Angiogenesis as a potential target of pharmaconutrients in cancer therapy. Curr Opin Clin Nutr Metab Care 13: 417-422. doi: 10.1097/MCO.0b013e3283392656
    [70] Prasad S, Phromnoi K, Yadav VR, et al. (2010) Targeting inflammatory pathways by flavonoids for prevention and treatment of cancer. Planta Med 76: 1044-1063. doi: 10.1055/s-0030-1250111
    [71] Asensi M, Ortega A, Mena S, et al. (2011) Natural polyphenols in cancer therapy. Crit Rev Clin Lab Sci 48: 197-216. doi: 10.3109/10408363.2011.631268
    [72] Pham MH, Dauzonne D, Chabot GG (2016) Not flavone-8-acetic acid (FAA) but its murine metabolite 6-OH-FAA exhibits remarkable antivascular activities in vitro. Anti-Cancer Drugs 27: 398-406. doi: 10.1097/CAD.0000000000000341
    [73] Granja A, Pinheiro M, Reis S (2016) Epigallocatechin Gallate Nanodelivery Systems for Cancer Therapy. Nutrients 8: 307. doi: 10.3390/nu8050307
    [74] Bauvois B (2012) New facets of matrix metalloproteinases MMP-2 and MMP-9 as cell surface transducers: outside-in signaling and relationship to tumor progression. Biochim Biophys Acta 1825: 29-36.
    [75] Trujillo A, McGee C, Cogle CR (2012) Angiogenesis in acute myeloid leukemia and opportunities for novel therapies. J Oncol 2012: 128608.
    [76] Haouas H (2014) Angiogenesis and acute myeloid leukemia. Hematology 19: 311-323. doi: 10.1179/1607845413Y.0000000139
    [77] Klein G, Vellenga E, Fraaije MW, et al. (2004) The possible role of matrix metalloproteinase (MMP)-2 and MMP-9 in cancer, e.g. acute leukemia. Crit Rev Oncol Hematol 50: 87-100. doi: 10.1016/j.critrevonc.2003.09.001
    [78] Bouchet S, Tang R, Fava F, et al. (2014) Targeting CD13 (aminopeptidase-N) in turn downregulates ADAM17 by internalization in acute myeloid leukaemia cells. Oncotarget 5: 8211-8222. doi: 10.18632/oncotarget.1788
    [79] Bouchet S, Bauvois B (2014) Neutrophil Gelatinase-Associated Lipocalin (NGAL), Pro-Matrix Metalloproteinase-9 (pro-MMP-9) and Their Complex Pro-MMP-9/NGAL in Leukaemias. Cancers (Basel) 6: 796-812. doi: 10.3390/cancers6020796
  • Reader Comments
  • © 2016 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(6099) PDF downloads(1081) Cited by(3)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog