Tripterygium wilfordii

1.1. Sources

Tripterygium wilfordii (of the family Celastraceae) is a chinese herb that is referred to as lei gong teng but more well known as Thunder God Vine.^[1] The plant is known as toxic although the root pulp is seen as pharmacologically active against inflammatory and autoimmune diseases, and it appears the root extracts have some traditional usage against Rheumatoid Arthritis and (to use diction common to traditional medicine) for 'dispeling wind and eliminating dampness, dredging collaterals and relieving pain, reducing swelling and easing pain'.^[2][3] Although the vine being associated with adverse effects even at moderate doses (diarrhea, headache, nausea and infertility), and it has been noted that inadequate processing techniques can lead to vine contamination of root extracts.^[1]

Other names that are attributed to this herb beyond 'Thunder God Vine' are yellow vine wood, Gelsemium elegans, vegetable insecticide and red medicine.

Tripterygium wilfordii is a Traditional Chinese Medicine for inflammatory disorders where the root is used, and the vine appears to have toxic properties (despite this medicine sometimes being referred to as Thunder God Vine)

1.2. Composition

• Celastrol and Triptolide, the main bioactives^[1][4]

• The alkaloids (1.6% total) wilforine (78.3mg/100g), wilforgine (47.5mg/100g), wilfordine (26.5mg/100g) and wilforidine (4mg/100g)^[2]

• Wilforide A

• Demethylzeylasteral^[5]

The main three bioactives (Celastrol and Triptolide for bioactivity, Wilforine as well) are concentrated in ethanolic extracts and can be further concentrated in Ethyl Acetate (EtOAc) extractions.^[6] Sodium Carbonate extracts can then be used to concentration Celastrol while Triptolide and Wilforine are left in the ethyl acetate layer.^[6]

1.3. Structure

Celastrol itself is a pentacylic triterpenoid but is further classified as a quinone methide triterpene.^[1] Two carbons on the structure (C2 and C6, labelled below) demonstrate high susceptibility toward a nucleophilic attack^[7] and can form covalent Michael adducts with nucleophilic thiol groups of cysteine residues and underlies many mechanisms of Celastrol with enzymes.^[8][9] Reactions with the 'quinine methide' of Celastrol refer to these nucleophilic attacks and subsequent adducts.

Celastrol can work via forming adducts with Cystine groups (involved in the structure of proteins) due to positions on its backbone that are prone to accepting an electron

Triptolide has been synthesized into a prodrug known as PG490-88 which has been patented^[10] and is under investigation for anti-cancer effects.^[11]

1.4. Extractions and Formulations

Solid Lipid Nanoparticles (SLNs) are made from solid lipids and approximately 50–1000nm in width that serve as drug carriers,^[12] and this delivery system has been used in numerous studies as an attempt to mitigate hepatotoxic side effects of Triptolide.^{[13][14][15][16]}

SLNs appears to be a drug delivery formulation which may mitigate some toxicity associated with Thunder God Vine

Sodium Bicarbonate extractions, which appear to concentrate Celastrol at the expense of Triptolide and Wilforine, are associated with a reduced LD₅₀ (median lethal dose) to 1,210mg/kg (95% CI of 1,098–1,322mg/kg) which appears to be slightly safer than Tripterygium Glycosides with an LD₅₀ of 257mg/kg (95% CI of 227-287 mg/kg).^[6]

2.1. Metabolism

Triptolide is metabolized into four monohydroxylated metabolites in liver microsomes (rat) and three in human liver microsomes, with CYP3A4 playing a major role in metabolism of Triptolide into its metabolites and CYP2C19 playing a lesser role; the one metabolite mediated by CYP2C19 being the metabolite not detectedin human liver microsomes.^[11]

2.2. Enzymatic Interactions

Demethylzeylasteral is able to inhibit UGT1A6 (Ki 0.6μM; IC₅₀ 15.2+/-0.6μM) and UGT2B7 (Ki 17.3μM; IC₅₀ 62.5+/-1.6μM) via competitive inhibition, with 100μM of the molecule reducing the activity of UGT1A6 and UGT2B7 to 0.8+/-0.1% and 14.8+/-1.1% of baseline value (with a third isoform, UGT1A9, being reduced to 61.3+/-5.6%).^[5]

3.1. Mechanisms

IKKα and IKKβ are both inhibited in a dose-dependent manner in vivo following Celastrol administration, and it appears Celastrol targets the Cys-179 residue via its quinine methide (reducing the quinine methide to form Dihydrocelastrol abolishes this effect).^[8] This binding prevents degradation of IκBα and subsequently inhibits NF-kB activity (as IκBα is a suppressor of NF-kB) but as Celastrol does not inhibit ReIA-induced NF-kB activity the inhibitory effects appear to be solely downstream of IKKβ (although this inhibition blocks LPS, PMA, and TNF-α induced NF-kB activity which is via IKKs).^[8] There may not be a direct inhibition of IKKα in a similar manner, but inhibition of TAK1 (which positively regulates IKKs and inhibition of TAK1 resulting in suppression of IKKS) has been noted with Celastrol,^[17][18] and inhibition of an HSP90-Cdc37 complex^[9] appears to also influence IKKs and may be a novel regulator.^[19]

At least one study in prostatic cancer cells noted that inhibition of chymotrypsin-like proteosomal activity caused an increase in IκBα,^[7] the accumulation of which could also suppress NF-kB activity.^[1] Induction of HSP70, noted due to possibly lowering the activation requirements of its promoter, has been noted with Celastrol;^[20] HSP70 can reduce the association of NEMO to IKK, NEMO being a protein that associates with IKKs (IKKα-IKKβ-NEMO complex) and works with HSP-Cdc37 to induce IKK activation.^[21]

Indirect inhibitor of NF-kB secondary to IKKβ inhibition, and may inhibit the IKK complex via (a lot of) indirect means beyond direct inhibition of IKKβ. Appears to be effective at very low concentrations

No influence on AP-1 binding is noted.^[8]

3.2. Allergies

Celastrol can bind to ERK proteins and prevent association with FcεRI receptors, which are expressed on mast cells and mediate anti-allergic responses.^[22]

3.3. Macrophages

In macrophages (RAW 264.7) incubated with 300-1000ng/mL Celastrol note absolute suppression of LPS-induced nitric oxide and TNF-α release with dose-dependent effects at lower concentrations with 10ng/mL noting a reduction and 100ng/mL being statistically significant.^[8]

3.4. T-Cells

Triptolide (bioactive of Thunder God Vine) has been noted to inhibit IL17 and IFNγ production and secretion secondary to inhibiting Th17 cell differentiation,^[23][24] which has been thought to underlie protection from hepatic ischemia/reperfusion injury; this study gave 0.1mg/kg Triptolide injections for a week prior to hepatic ischemia reduced the increase in IL17 (demonstrated to be critical due to IL17 antibodies also being protective) due to less Th17 (CD4+ positive lymphocytes) which was thought to be due to inhibition of STAT3 (required for Th17 differentiation).^[25]

Interleukin-6 (IL6) is able to activate STAT3^[26] which further acts on the retinoid orphan receptor γ-T (RORγt)^[27] which induced Th17 differentiation without significantly affecting T_reg cells in general;^[28] this inefficacy on Treg cells despite inhibition of Th17 cells has been noted following Triptolide administration to mice both after ischemia^[25] and IL6-induced STAT3 activation, which Triptolide inhibits after injections of 0.07mg/kg over 8 weeks.^[29]

Inhibition of Th17 cell activation may underlie therapeutic efficacy in T-cell dependent inflammatory conditions, such as colonic inflammation^[29] and experimental rheumatoid arthritis.^[23]

Can inhibit differentiation of Treg cells into Th17 (CD4+ lymphocytes that secrete IL17) and be selectively immunosuppressive on activated T-cells, which may be a therapeutic mechanism of action; this is possibly mediated via STAT3 inhibition

3.5. Interventions

Administration of 10-30mg/kg Celestrol (injections) noted dose-dependent suppression of Carrageenan-induced Myeloperoxidase (MPO), TNF-, and PGE₂ production by 90.7+/-2.1%, 66.5+/-3%, and 93.0+/-2.0% at the higher dose.^[8]

3.6. Rheumatoid Arthritis

In regards to rat and mouse studies, one study using solely alkaloids from the root (ethanolic extraction) that had rheumatoid arthritis induced with injections of Complete Freund’s adjuvant (pro-arthritic toxin) noted that oral ingestion of 0.25-1g/kg total alkaloids over 28 days showed dose-dependent reductions in paw edema (underperforming the active control of 7.6mg/kg Dexamethasone) suppressed the serum cytokines IL-6, IL-8, and TNF-α (0.5-1mg/kg same potency as active control) and improved histology of the joints (lesser potency than active control).^[2]

It has been noted that Celastrol can induce Yap1 activity secondary to interactions with Yap1's carboxy-terminal redox center.^[30]

5.1. Liver

Cytochrome P450 reductase is involved in protective effects against Triptolide, and its knockout in animals augments hepatoxicity secondary to increasing its bioavailability (by reducing metabolism).^[31] Dexamethasone, a corticosterone, can reduce toxicity secondary to inducing the activity of CYP3A^[32] and drug delivery systems that bypass the liver reduce toxicity somewhat.^[33]

CYP3A appears to metabolize triptolide to less toxic metabolites; with more activity of CYP3A being protective and less activity (inhibition; which is common to many supplements) can augment toxicity

6.1. Mechanisms

Mechanistically, Celastrol appears to be able to prevent association between HSP90 and Cdc37 via binding to one of three cystiene residues on the N-terminal of Cdc37 (and physically preventing the ability of HSP90 to bind to Cdc37) and does not inhibit the ATP binding site of HSP90 yet is reversible (both of those mechanisms being common to the antibiotic Geldanamycin; a known HSP90 inhibitor).^[9] Secondary to indirect inhibition of HSP90-Cdc37, downregulation of oncogenic protein kinases are noted (prostatic cancer cells).^[9]

Association with p23 has also been noted with Celastrol but also noted with Dihydrocelastrol (and possibly not mediated by the quinine methide), and this association induces a amyloid-like fibril formation which prevents association of HSP90 to p23.^[34]

May directly inhibit HSP90-Cdc37 complexation, and may alter p23 function in regards to HSP90

One study has noted both hyperphosphorylation of heat shock transcription factor-1 (HSF1) with lower concentrations augmenting the response of this promoter to heat shock stress, suggesting that Celastrol may reduce the heat threshold required to activate HSF1.^[20] This results in induction of HSP70 (Heat-Shock Protein 70).^[20]

The inhibition of NF-kB also appears to underlie the prevention of TNF-induced induction of metastatic (MMP9, COX2, and ICAM1), antiapoptotic (IAP1, IAP2, Bcl-2, Bcl-XL, c-FLIP, and survivin), and proliferative (cyclin D1 and COX-2) gene products; this may underlie how Celastrol can synergisically enhance Doxorubicin, Palcitaxel, and TNF-induced apoptosis.^[18] This NF-kB inhibition does not appear to be cell-type specific, as it was demonstrated in Lung adenocarcinoma (H1299), bladder (253JBV), myeloma (U266) and embryonic kidney cells (A293), and appears to be secondary to preventing TNF-induced IκBα degradation and phosphorylation, and prevented TAK1/TAB1-induced NF-κB–dependent reporter gene expression.^[18]

6.2. Breast

One study in mice where invasive breast cancer cells (MDA-MB-435) were injected into mice injections of 30mg/kg Celastrol were given every other day for 7 weeks, a 60+/-5% inhibition of tumor size was noted.^[8]

Triptolide appears to be anti-proliferative in both estrogen responsive (MCF-7) and estrogen non-responsive (MDA-MB-435) cells, although with greater potency in MCF-7; this appeared to be related to varying effects on p53, where an upregulation of wild type p53 was noted in MCF-7 cells concurrent with a downregulation of ERα (dose dependent manner up to 40nM) while the mutant p53 in MDA-MD-435 cells was downregulated.^[35]

6.3. Prostate

Celastrol has been shown to inhibit proteasomal chymotrypsin activity in prostatic cells (IC₅₀ 2.5μmol/L; 80% inhibition at 5μmol/L) and can induce accumulation of ubiquitinated proteins (prostatic cancer cells); it does not appear to inhibit trypsin-like or PGPH-like proteasomal activities significantly.^[7] This mechanism is similar to Bortezomib, an anti-cancer drug, and was previously noted with flavonoid compounds with a ketone group on the C ring (Quercetin and Kaempferol^[36]); this inhibition of proteasomal activity caused apoptosis in incubated cancer cells (PC-3) by 55% and (LNCaP) 40% at 2.5μmol/L, which was associated with accumulation of IκBα, Bax, and p27 and caspase-3 release and PARP cleavage.^[7]

Androgen receptor expression in LNCaP cells is attenuated with 2.5μmol/L Celastrol and nearly abolished at 5μmol/L.^[7]

In mice injected with PC-3 prostatic tumors was able to suppress 65% and 82% of the subsequent tumor growth over the next 30 days (1mg/kg and 3mg/kg respectively) with significance noted at days 1-3 (up to 70% inhibition); a reduction of proteasamol activity to 25% of control was confirmed in vivo via tumor biopsy.^[7] When injected with C4-2B tumor cells, 35% inhibition over 3 days is noted and suppression of androgen receptor expression confirmed via biopsy.^[7]

6.4. Osteosarcoma

One study in W256 osteosarcoma cells noted that isolated Celastrol was able to induce apoptosis with an IC₅₀ of 0.43+/-0.1μmol/L, which outperformend Partheolide (5+/-1μmol/L); the active constituent of Feverfew.^[17] This is secondary to caspase release and DNA fragmentation, and thought to be from inhibiting TAK1 phosphorylation (parthenolide ineffective on this protein) and abolishing IKKβ binding to IKKγ at 1umol/L and preventing activation of the IKK complex and subsequent NF-kB activation (mechanism shared with parthenolide).^[17] Migration of W256 cells was abolished in vitro with 0.5umol/L Celastrol, and injections of 1mg/kg Celastrol to mice bearing W256 tumors was able to significantly reduce metastasis (from absolute to 11%), reducing trabecular bone loss, and reducing osteolytic lesions by up to 75%; similar efficacy was seen with 1mg/kg Parthenolide.^[17]

6.5. Melanoma

When tested in vitro, Celastrol is weakly selective for tumor cells rather than healthy melanoma cells with IC₅₀ values of growth inhibition being 2.32+/-0.04µM (B16 melanoma cell), 2.12+/-0.04µM (SK-MEL-2 melanoma cell) and 3.78+/-1.75µM (normal HaCaT cell line).^[37]

Apoptosis appears to be mediated in Melanoma cells in the concentration range of 1-3uM with an increase in cell count accumulated in the sub-G1 phase and increase DNA fragmentation; indicative of apoptosis, which appears to be a ROS mediated caspase-3 release and PARP cleavage.^[37] This study also noted that AIF plays a role (apoptosis independent of capsases^[38]) which was ROS dependent, and attenuates signalling via PI3K/AKT.^[37]

An inherent problem with using Thunder God Vine for any supplemental purpose is that the bioactive components that exert benefit are the same compounds with potential toxicity, making 'safe' extractions of this supplement also ineffective.

7.1. Triptolide

Triptolide, a main bioactive of Tripterygium Wilfordii, appears to have a low therapeutic threshold and potential for toxicity; the LD₅₀ (median lethal dose) following injections in rats is a mere 0.86mg/kg.^[39] In HBZY-1 (rat mesangial) cells, Triptolide has a therapeutic ratio of approximately 11.1 (the EC₅₀ of inhibiting inflammation, 20.7+/-2.1nM, being 11.1-fold less than the dose inducing 50% cell death) and in macrophages this therapeutic ratio is 1.6.^[6]

7.2. Celastrol

Celastrol appears to have a therapeutic threshold ratio of 3.6 in rat mesangial cells (the dose exerting 50% inflammation inhibition, 1040+/-40nM, being 3.6-fold lower than that inducing 50% cell death) while in macrophages this therapeutic threshold is reduced to 1.4.^[6]

7.3. Case Studies

Various case studies associated with usage of Tripterygium wilfordii include Kaposi sarcoma after 10 years usage of 2g powder daily (containing 470mcg Triptolide),^[40]