1,4-dihydroxy-2-naphthoic Acid Induces Apoptosis in Human Keratinocyte: Potential Application for Psoriasis Treatment

Psoriasis, which affects approximately 1–3% of the population worldwide, is a chronic inflammatory skin disorder characterized by epidermal keratinocytes hyperproliferation, abnormal differentiation, and inflammatory infiltration. Decrease in keratinocyte apoptosis is a specific pathogenic phenomenon in psoriasis. Chinese herbs have been used for the treatment of psoriasis in China showing promising effect in clinical trials. A traditional Chinese medicine has relatively fewer side effects with longer remission time and lower recurrence rate. The extract of Rubia cordifolia L. (EA) was previously found by us to induce HaCaT keratinocytes apoptosis. In this study we identified one of the components in Rubia cordifolia L., the anthraquinone precursor 1,4-dihydroxy-2-naphthoic acid (DHNA), induces HaCaT keratinocytes apoptosis through G0/G1 cell cycle arrest. We have also demonstrated that DHNA acts through both caspase-dependent and caspase-independent pathways. Besides, cytotoxicity and IL-1α release assays indicate that DHNA causes less irritation problems than dithranol, which is commonly employed to treat psoriasis in many countries. Since DHNA possesses similar apoptotic effects on keratinocytes as dithranol but causes less irritation, DHNA therefore constitutes a promising alternative agent for treating psoriasis. Our studies also provide an insight on the potential of using EA and DHNA, alternatively, as a safe and effective treatment modality for psoriasis.


Introduction
Psoriasis, affecting 1-3% of the population worldwide [1], is a common chronic immune disorder characterized by thickened red plaques with an overlying silver-white scale. The most common type of psoriasis is psoriasis vulgaris which accounts for 90% of the cases [2]. Affecting both sexes equally and people of all ages; psoriasis causes significant impacts on the quality of life [3]. Conventional treatments of psoriasis such as topical, phototherapy, and systemic treatment are based on severity of disease. Since an estimated 75% of psoriatic patients have mild-to-moderate disease, topical treatments remain the most widely used [1]. The costs for treating patients with psoriasis are substantial, estimated to be US $1.6-3.2 billion/year and $1,130 to 6,650 per patient/year [4,5]. With recent introduction of biological agents, the estimated cost can range from US $13000 to 30000/patient annually [6].
Psoriasis has three principal histological features, including epidermal hyperplasia or abnormal differentiation, dilated blood vessels in dermis, and predominantly infiltration of leucocytes into dermis causing inflammation. It is considered a T-helper 1 (Th1) disease based on the increase in cytokines of the Th1 pathway such as interferon gamma, interleukin (IL) 2 and 12, as found in psoriatic plaques [2,3]. However, a strong evidence also indicated keratinocytes contributed to the disease [7] and keratinocytes is viewed as another major player of this chronic inflammatory disease [8]. For instance, decrease in skin cells apoptosis is suggested as a specific pathogenic phenomenon in psoriasis [9,10]. Acanthosis in psoriatic skin is resulted from diminished epidermal apoptosis [11], while induction of apoptosis is 2 Evidence-Based Complementary and Alternative Medicine involved in the regression of psoriatic hyperplasia [9,10]. Therapies based on the regulation of keratinocyte proliferation are potentially useful in treatment of psoriasis because the restored homeostatic control of keratinocyte growth and differentiation is crucial for recovery from psoriatic to normal epidermis [12]. For example, established therapies like dithranol, vitamin-D 3 analogs, and low-dose methotrexate induce apoptosis [11].
Chinese herbs have been used for the treatment of psoriasis in China, and clinical trials have also been carrying out [13,14] with promising evidence of effectiveness [15,16]. In traditional Chinese medicine (TCM) theories, psoriasis is classified into three categories: blood heat, blood dryness, and blood stasis [13]. The rationale for using TCM is related to their actions on clearing internal heat, activating blood to remove stasis, and strengthening any deficiencies. The literature review also describes the mechanism of the TCM treatment in psoriasis is through inhibition of keratinocyte proliferation and induction of apoptosis [17]. Besides, therapeutic actions of these herbs include anti-inflammatory, modulation of cytokine production, or inhibition of angiogenesis which are relevant in reducing the severity of psoriasis [13,17]. Although Western medicine is practiced primarily for the treatment of psoriasis, patients' dissatisfaction with conventional treatment or adverse effects after application were the most common reasons for switching to alternative therapies like herbal remedies. Besides, reviews also reported that TCM has relatively fewer side effects, a longer remission time, and a lower recurrence rate than the Western therapeutic agents [18]. Consistent with this finding, one study showed that the recurrence rate of psoriasis in TCM treated group was significantly lower than Daivonex (calcipotriol) treated group while the clinical efficacy was similar [19].
In our previous studies it was demonstrated that the extract of Rubia cordifolia L. could induce keratinocyte apoptosis in a psoriasis-relevant HaCaT cell models [12,20,21]. In this study one of the components in Rubia cordifolia L., the anthraquinone precursor DHNA (Figure 1), was found to induce HaCaT keratinocytes apoptosis through G0/G1 cellcycle arrest, activation of caspases as well as in a caspase independent manner. Moreover, results from cytotoxicity and IL-1 release assay suggest DHNA causes less irritation problems than dithranol, which its use has declined because of its substantial skin irritation problems, but is still a popular and important topical treatment for psoriasis among European countries. Since DHNA possesses similar apoptotic effects on keratinocytes as dithranol but causes less irritation, these results suggest DHNA may be developed into a less-irritating topical agent alternative to dithranol for the treatment of psoriasis. In addition, the ethyl acetate extract of the root of Rubia cordifolia L. and DHNA may be used together in a sequential therapy as a safe and effective therapeutic alternative for the treatment of psoriasis.     ). As TUNEL assay is a sensitive method which detects single-strand breaks in high molecular weight DNA, it was able to show that 30 M (6.13 g/mL) DHNA treatment induced apoptosis in HaCaT cells as early as 24 h while other apoptotic markers are still not available or detectable (e.g., externalization of PS, morphology).      On the other hands, no active form of caspase-9 was found and Bid was not cleaved. There were also no significant changes in expression levels of Bcl-2, Bcl-xL, Bid, and FADD (Figure 9(a)). When treatment time was increased to 24 h, activated caspase-9 and cleaved Bid could then be detected (Figure 9(b)).

DHNA Induced Caspase Independent Apoptosis in HaCaT
Cells. In vehicle and medium control cells, immunofluorescence staining revealed the localization of AIF in mitochondria and its absence in nucleus (Figures 10(a) and 10(b)), 3.10. IL-1 Release Assay. In vehicle and medium control, no IL-1 release from NCTC 2544 cells was detected ( Figure 12). DHNA significantly induced the release of this cytokine at a much higher concentration than dithranol (60 M or 12.25 g/mL for DHNA versus 2 M or 0.45 g/mL for dithranol; ∼30 times) ( Figure 12). Compared with vehicle control, a significant induction of IL-1 release could only be observed at higher concentrations of DHNA tested (60 to 100 M or 12.25 to 20.42 g/mL) (Figure 12(a)). In contrast, dithranol induced significant release of this cytokine at all concentrations tested (Figure 12(b)

Discussion
Psoriasis is recognized as an organ specific autoimmune disease triggered by an activated cellular immune system [22]. As the exact aetiology and pathogenesis remain unknown, it has been argued for many years whether psoriasis is a disorder of the immune system or disorder of the skin. Evidence indicates keratinocytes contributed to the disease [7]. Based on several lines of evidence, we believe that keratinocytes may play a major and important role in the pathogenesis and development of psoriasis. It would be interesting to note that the presence of T cells in skin lesions does not mean it initiates the disease. Although psoriasis is thought to be a Th1 induced disorder, psoriasis develops in patients infected by the human immunodeficiency virus with the same frequency as in the general population [23]. In psoriasis, the expression levels of many gene products are altered, compared with unaffected skin. Among these altered gene products, the expression levels of messenger RNAs transcribed by keratinocytes are much higher than T cells due to the overwhelming number of keratinocytes in lesions [24]. Next, keratinocytes can synthesize many cytokines and chemokines which could induce chemotaxis that attracts and activates leukocytes found in epidermis of psoriatic lesions such as ICAM-1, CD40 and HLA-DR [24,25]. Increased levels of CXCL9/MIG, CXCL10/IP-10, CXCL11/I-TAC [24], and RANTES/CCL5 [26] synthesized by keratinocytes in psoriasis lesions as well as the BRAK/CXCL14 that are upregulated in nonlesional skin of psoriasis patients [24], may also activate and initiate the migration of mononuclear leukocytes/T cells/monocytes to the psoriatic lesions. Besides, keratinocytes not only produce but also respond to the growth factors and cytokines produced themselves [27]. For example the production of transforming growth factor (TGF ) by epidermal keratinocytes is greatly elevated in psoriasis [27]. This results in increased transcription of hyperproliferating keratin K6/16 genes and the activation of keratinocytes, which produce and respond to immunological signals like interleukins 1/6 and growth factors like TGF / and epidermal growth factor (EGF). Subsequently, a series of signal cascades is activated that contributes to the development and persistence of psoriatic lesion. Moreover, the response of psoriatic keratinocytes to T cell cytokine in vitro is different from normal keratinocytes. For example, the basal keratinocytes of psoriatic uninvolved skin are susceptible to the growth stimulatory effect of T cell cytokine IFN-, while this is not in normal skin [28]. Therefore, keratinocytes from a psoriatic environment exist in a preactivated state that is highly responsive to signaling. The special conditions or changes which result in predisposition of activation in those keratinocytes may be the reason for later development to psoriatic plaque under the influence of immunologic stimulation. In this regard, there are genetically engineered mice models that target epidermal keratinocytes and are sufficient to initiate psoriasis like phenotype [29,30] such as the IKK2 knockout mice [29,30] and the JunB/c-Jun double knockout mouse [23], and both of them resemble many features of psoriasis that are T cell independent. Taken together, keratinocytes participate in maintaining a chronically perpetuating immune response that sustains psoriasis, and changes in keratinocytes may even trigger this disease in the skin of predisposed individuals. Therefore, it is believed that epidermal alterations may account for the initiation of skin lesions in psoriasis and thus restoring hyperproliferative keratinocytes to normal    would be beneficial in this disease. Particularly, inhibition of keratinocytes proliferation and induction of keratinocyte apoptosis have long been considered as targets of antipsoriatic strategies [31], and the discovery of new antiproliferative and keratinocyte differentiation-modulating agents remains a worthy aspect in psoriasis research nowadays.
Earlier we identified the extract of the root of Rubia cordifolia L. induces apoptosis in a psoriasis-relevant HaCaT keratinocyte [12,20,21]. In particular, hydroxyanthraquinones are the major bioactive compounds in R. cordifolia roots [32,33]. In this study, we attempt to identify the compound which presents in Rubia cordifolia L. that may be responsible for the observed effects of the herb extract on HaCaT cells. During screening the anthraquinone precursor, DHNA was found to time and dose dependently inhibit keratinocytes proliferation. Compared with dithranol, DHNA was less potent in inhibiting keratinocytes growth (∼4 times on HaCaT; ∼12 times on NCTC 2544). However, DHNA exhibited much less cytotoxicity on fibroblasts (∼8 times on Hs-68) and melanocytes (∼10 times on PIG1) while was comparable to EA extract of Rubia cordifolia L. (Table 1). It is worth noting dithranol is safe, effective and has been a mainstay treatment of psoriasis for over 80 years; however, its use has declined because it causes substantial skin irritation as well as staining of skin, clothing, and furniture [34][35][36]. For topical treatment of psoriasis, skin irritation is a common side effect [1,4,35]. Skin irritation is a local arising, reversible, and nonimmunological inflammatory reaction that is caused by contact and stimulation with irritants [37,38]. Irritants can cause epidermal damage and initiate the release of IL-1 from keratinocytes as a primary event in skin defense [39]. Subsequently, secondary mediators (such as release of chemokines) and morphological alterations (vasodilation and infiltration of immune cells) are induced with finally the onset of typical symptoms such as oedema, erythema, itching as well as pain [37,40]. Despite its substantial skin irritation problem, dithranol is still popular among European countries as an important (second choice) therapy [35,41]. With respect to the underlying mechanisms of action of dithranol in psoriasis, dithranol can inhibit the growth of mouse and human keratinocytes as well as psoriatic keratinocytes in vivo; it was also demonstrated that it induces apoptosis in keratinocytes via the mitochondrial pathway [42]. In this regard, identification of new treatment which can achieve similar apoptotic effects of dithranol while with less irritation side effect would be of great interest.
Traditionally skin irritation assay is conducted in animals using Draize animal test [43] which has been introduced for more than 60 years as a standard for the evaluation of skin irritating properties of chemicals [44]. However for ethical reasons, the Cosmetics Directive (Directive 76/768/EEC) and European Directive 2003/15/EEC impose a permanent ban of animal testing for cosmetic products and ingredients from 2009 in Europe. As the ethical issues and regulatory context are similar around the world [45], hence, there is a great effort aiming to reduce and replace animal studies. The development and use of alternative in vitro methods is encouraged [40,45,46]. In vitro skin irritation studies have been used to predict the skin irritation potential of different chemicals [47,48]. The most commonly used parameter to assess irritant responses is the in vitro measurement of cell viability [37]. It is a suitable endpoint to evaluate the irritation potential of any given substances, as cytotoxicity is known to trigger irritation processes [38]. One in vitro skin irritation model, the "single cell assay", involves a single-cell type cultured in plastic dishes under conventional culture condition [40]. Previous studies suggested that cultured human keratinocytes can be used to predict irritation caused by various surfactants [49] and there is a good correlation between in vitro cytotoxicity and in vivo skin irritation potential in human [50,51]. Therefore, single cell assay and cytotoxicity assay have been employed in many studies to predict the skin irritation potential of chemicals. For example, human keratinocyte (HaCaT, NCTC 2544) monolayer cultures have been used as an in vitro model to predict potential skin irritation with cytotoxicity as endpoints [39,[52][53][54]. Therefore in our study, we also used single cell assay with cytotoxicity to evaluate the skin irritation potential of DHNA and dithranol. For better prediction of irritation potential, apart from cell viability, additional biomarkers should also be incorporated in the study. As mentioned, IL-1 is an important mediator which initiates inflammation in the skin. Only followed cell injury, IL-1 is exclusively released from leaky keratinocytes while in intact epidermis, IL-1 is naturally eliminated by desquamation [37]. In both in vitro and in vivo systems the release of IL-1 is accepted as an early marker of irritation [39,54]. Therefore, together with cell viability test, measuring the distribution of cytokines (release of IL-1 ) is an effective and promising approach to predict chemicals as potential irritants [39]. The use of keratinocytes with these endpoints as in vitro skin irritation model has already been justified [55]. Because of its epidermal location and its ability to synthesize different inflammatory mediators, and there is a good correlation between in vitro response of keratinocytes cultures and in vivo skin irritation data [53], human keratinocytes become a relevant biological target and, hence, represents an excellent cell model in assessing skin irritants [53]. In EU, the Cosmetics Directive (76/768/EEC) imposes that cosmetic products put on the market within the Community must not cause damage to human health when they are applied under normal of use. Before new cosmetic products are launched and marketed, the potential of its ingredients or the final product in causing skin irritation needs to be evaluated as part of the safety assessment [56]. As mentioned, this directive also bans animal testing on cosmetic products. To replace animal testing the cosmetic and toiletry industry in Europe represented by the European, Cosmetic Toiletry and Perfumery Association (COLIPA), has adopted the use of in vitro test systems to predict potential irritants [37]. Since undifferentiated keratinocytes may synthesize higher level of IL-1 and its production is decreased as keratinocytes are gradually differentiated [57], therefore the undifferentiated human keratinocytes (NCTC 2544 cells) are used to evaluate the release of this inflammatory cytokine upon treatments with DHNA or dithranol. Our results demonstrated DHNA is predicted to be less irritating than dithranol. First, IC 50 values from cell viability assay revealed DHNA was less cytotoxic than dithranol (Table 1). On the other hand, both compounds induced the release of proinflammatory cytokine IL-1 from keratinocytes; however, DHNA induced the release of this cytokine at a much higher concentration than dithranol (∼30 times, Figure 12). Besides, at IC 50 of DHNA on NCTC 2544 keratinocytes no significant level of IL-1 was released (Table 1, Figure 12(a)), while dithranol could readily induce significant release of this cytokine at concentrations below its IC 50 on NCTC 2544 keratinocytes (Table 1, Figure 12(b)). Since DHNA has lower skin irritation potential than dithranol, therefore if DHNA can exert similar apoptotic effects on keratinocytes as dithranol, it could be a promising 14 Evidence-Based Complementary and Alternative Medicine layers which contact more directly with topically applied psoriasis treatment, subsequent studies were focused on this cell model. An early event during apoptosis is externalization of PS which is normally confined to the internal leaflet of plasma membrane [58]. This was confirmed in DHNA treated HaCaT cells by annexin V/PI staining (Figures 3(a)-3(d)). A later step in apoptosis is DNA fragmentation that results from the activation of endonucleases [59]. Apoptotic cell shrink, the chromatin becomes pyknotic and compacted, and nucleus breaks up (karyorrhexis). Cell appears to bud (membrane blebbing), which later become apoptotic bodies and phagocytosed by macrophages [60]. Hoechst stained HaCaT cells treated with DHNA clearly revealed chromatin condensation (pyknotic) and nuclear fragmentation (karyorrhexis) which are commonly observed in ongoing apoptotic cells (Figure 2). The Cell Death Detection ELISA plus assay also demonstrated that DHNA treatment increased DNA fragmentation in HaCaT cells by detecting histone-associated DNA fragments ( Figure 6). Moreover, induction of apoptosis in HaCaT cells by DHNA was further confirmed with TUNEL technique by detecting DNA strand breaks (Figure 7). It is suggested PS exposure precedes the nuclear changes that define apoptosis as well as the loss of membrane integrity [58]. Our results agreed with this observation that externalization of PS can be detected first (e.g., 60 M DHNA at 24 h treatment), with the decrease in mitochondrial membrane, potential and morphological change thereafter. Taken together, it is confirmed that the apoptotic induction of HaCaT cells by DHNA is time and dose dependent.
As induction of apoptosis may be cell-cycle related [61,62], therefore cell-cycle analysis was performed on DHNA treated HaCaT cells. Our investigation showed that DHNA treated HaCaT cells were arrested at G0/G1 phase of the cell cycle with induction of sub-G0/G1 population, which indicates apoptotic cells ( Figure 5). Moreover, the resulting G0/G1 arrest was correlated with changes in the expression of major cell cycle proteins as analyzed by theWestern blot. The cyclin dependent kinases (Cdk) are serine/threonine kinases which are important in controlling cell cycle in eukaryotic cells [63][64][65]. Cyclins form heterodimeric complexes with specific Cdk at different points in cell cycle to phosphorylate target proteins and hereby promote cell cycle progression. In G1/S phase transition, cyclin D/Cdk4 and cyclin D/Cdk6 are the first two complexes to become active and involved in early G1 phase, while cyclin E/Cdk2 is formed in late G1 phase. Later, activated cyclin A/Cdk2 complexes drive cells into S phase. DHNA treatment down-regulated the level of Cyclin A/D1/D3 and Cdk4 which are required for entry into S phase in HaCaT cells (Figure 8). The inactive form of Cdk2 (Thr14 and Tyr15 phosphorylated, 34 kDa) increased while there were no obvious change in the active form of Cdk2 (Thr160 phosphorylated, 33 kDa). Besides, DHNA treatment also dose-dependently increased the expression of cyclin dependent kinase inhibitor p21, which binds and inhibits the activity of cyclin/Cdk complexes and blocks progression through G1/S phase in cell cycle. In psoriasis, epidermal hyperplasia is characterized by an increase of percentage of normally quiescent basal keratinocytes in the proliferative phases of cell cycle [66]. In particular, cyclin D1 is overexpressed in keratinocytes and is associated with epithelial hyperproliferation in psoriasis [67]. Our findings imply DHNA may be able to alter abnormal expression of the cell cycle proteins, favor G0/G1 arrest, and apoptosis in psoriatic keratinocytes which demonstrated decreased apoptosis.
To study the molecular mechanisms underlying DHNA induced HaCaT cells apoptosis, expression level of major caspases and their substrates were examined using the Western blot analysis. DHNA treatment for 12 hours increased the expressions of Fas receptor and the active form of caspase-3/7/8 in HaCaT cells. Besides, the DNA repair enzyme poly (ADP-ribose) polymerase (PARP) was cleaved and inactivated (Figure 9(a)). These results suggest, DHNA induces apoptosis through the death receptor mediated pathway. Ligation of death receptor (e.g., Fas) of the tumor necrosis factor receptor family with their ligands causes rapid death inducing signaling complex (DISC) formation and recruits Fas-associated death domain (FADD), which in turn recruits and processes procaspase-8 into a mature form. Activated caspase-8 then directly activates effector caspase (e.g., caspase 3) to trigger apoptosis [68,69]. Moreover, mitochondrial permeabilization indicated by the decrease in transmembrane potential was detected in DHNA-treated HaCaT cells (Figure 4). The loss of mitochondrial transmembrane potential is usually associated with cytochrome and other apoptotic factors released from the mitochondria. It is suggested that opening of the mitochondrial permeability transition pore (PTP) complex results in a decrease in MMP and cytochrome release [70]. In mitochondrial apoptotic pathway, permeabilization of mitochondrial outer membrane results in the release of apoptogenic factors (e.g., cytochrome ) from intermembrane space into cytoplasm. In cytoplasm, cytochrome forms complex (apoptosome) with apoptotic protease activating factor-1 (Apaf-1) and procaspase-9. The resulting apoptosome activates caspase-9 by cleavage, and activated caspase-9 in turn activates procaspase-3 to mediate apoptosis. The mitochondrial permeabilization is mainly regulated through Bcl-2 family proteins that consist of antiapoptotic (e.g., Bcl-2, Bcl-xL) and proapoptotic members (e.g., Bax). By interacting with PTP complex, the Bcl-2 proteins can regulate mitochondrial permeabilization. In addition, proapoptotic Bcl-2 proteins can also form pores in the outer mitochondrial membrane without involving PTP [70]. Besides, Bcl-2 proteins also provide a link between the death receptor and mitochondrial mediated apoptosis. When cleaved by activated caspase-8, the BH3-only proteins Bid can also trigger cytochrome release by translocation to mitochondria. Since activated caspase-9 and cleaved Bid is only observed after 24-hours-treatment of DHNA (Figure 9(b)), it is postulated that DHNA acts on death receptor first then triggers mitochondrial pathway and, hence, death receptor signaling occurs upstream the mitochondrial events. Further studies are required to confirm our speculation; nevertheless, the role of caspase is confirmed in our model, since the pancaspase inhibitor Z-VAD-FMK attenuated DHNA-induced HaCaT cells apoptosis (Figure 3(e)). The results from caspase inhibition assay showed that DHNA induced apoptosis in the HaCaT cells could only be partially blocked with Z-VAD-FMK. Since the caspase inhibitor Z-VAD-FMK used in the experiments is an irreversible inhibitor, therefore caspases alone were not solely responsible for the DHNA induced apoptosis. It would be interesting to note that DHNA treatment results in the loss of mitochondrial transmembrane potential in HaCaT cells ( Figure 4). In this regard, the mitochondrial proteins apoptosis inducing factor (AIF) and endonuclease G (endoG) have been implicated in the caspase independent apoptosis [71,72]; AIF and endoG released from mitochondria can trigger caspase independent apoptotic cell death [73] and contribute to nuclear chromatin condensation and DNA fragmentation [71]. AIF is normally confined to mitochondria. Upon apoptosis induction, AIF translocates from the mitochondrial intermembrane space to the nucleus. In nucleus, AIF causes chromatin condensation and large DNA fragmentation of around 50 kbp [74]. EndoG is mainly present in mitochondrial intermembrane space but it can also be found in nucleus [75]. It is suggested endoG is involved in mitochondrial DNA synthesis as well as apoptosis. Similar to AIF, endoG translocates to the nucleus during apoptosis. When released from mitochondria, endoG cleaves DNA into nucleosomal fragments independent of caspases [76]. Our results provide additional evidences that DHNA also mediates caspase independent apoptosis in HaCaT cells, as AIF and endoG were shown to be translocated to nucleus after DHNA treatment (Figures 10 and 11), where they directly caused chromatin condensation, DNA fragmentation and mediated apoptosis by a caspase independent mechanism [76,77].

Evidence-Based
Despite the presently available treatment options, one should bear in mind that psoriasis is a chronic disease and no available therapies are curative in nature. Therefore, disease inevitably recurs when therapy is discontinued [78].
Although most of the therapies in treating psoriasis have significant side effects [79], each medication has its own advantages and limitations. For optimal management, a given treatment requires an initial quick clearing of symptoms followed by long-term maintenance therapy [80]. This involves the sequential use of therapeutic agents which aim at improving the overall outcome of psoriasis treatment as well as to minimize the long-term toxicity of any particular treatment. The strategy includes 3 steps: (i) the clearing stage that employs short-term use of rapidly acting agents; (ii) the transitional stage involves gradual dose decreasing of the rapidly acting agent with the introduction of a safer maintenance agent; and (iii) the maintenance stage with the safer maintenance agent to ensure long-term control with minimal side effects. Our previous and present study suggest that, the ethyl acetate extract of the root of Rubia cordifolia L. (EA) and DHNA may be used together in this sequential therapy, in which EA could be effective at producing rapid clearing of psoriatic lesions with the fact that its potency is comparable to dithranol and more potent than DHNA in inhibiting keratinocytes proliferation, whereas DHNA may be better suited for maintenance therapy for its milder irritation effect compared with dithranol.

Conclusions
As the exact aetiology and pathogenesis of psoriasis is still unknown and current available treatments have many limitations, the pursuit and development of alternative therapeutic options remain an ongoing scientific effort. In this study we identified DHNA, similar to the herb extract we previously identified, induces apoptosis in HaCaT keratinocytes through G0/G1 cell cycle arrest. DHNA also demonstrated to act through both caspase-dependent and caspase-independent pathways of apoptosis. These results indicate DHNA possesses similar apoptotic effects on keratinocytes as dithranol, and at the same time, DHNA shows far less irritating actions as indicated by less cytotoxicity and less potent to induce the release of proinflammatory cytokine IL-1 . Since dithranol is an important and popular therapy among European countries, DHNA therefore constitutes a promising alternative agent for treating psoriasis. In addition, our study provides an insight on the potential of using the ethyl acetate extract of the root of Rubia cordifolia L. (EA) and DHNA alternatively in a sequential therapy as a safe and effective treatment modality for psoriasis. Future studies on EA and DHNA will further strengthen our knowledge in the mechanisms underlying their antipsoriatic effects and further help the development of drugs for the treatment of psoriasis.