Tooth-Fairy vs. the Easter Bunny

This post was earlier cross-posted at Leonid Schneider's site, hence the unfrivolous tone. The version there is improved by Leonid's editing, background details and frame-story

Dr Harriet Hall is credited with coining the unkind label "Tooth-Fairy Science" to describe the phenomenon of elaborate performative sciency-looking investigations, determining the behaviour of imaginary forces and entities. "Cargo-Cult Science" is an acceptable alternative. But Dr Hall did not envisage the possibility of fictional cross-overs and team-ups, "Tooth-Fairy vs. the Easter Bunny Science", where the explanatory variables in a study expand to include all possible interactions among multiple imaginary forces or entities.

Case in point: these string figures from network-analysis software are persuasive in their interlocking-gearwheel elegance, though if the goal of the analysis was to allow an underlying coherence to emerge from an inchoate sea of Protein / Protein interaction details, then Big Data has let us down again. The context is the use of a herbal diuretic to reduce ascites in tumour-xenografted mice, i.e. fluid distension as the carcinoma metastases eat into the linings of their abdominal cavities (the mice would still die, because carcinoma, but without so much inflation). Observing a contradiction between (1) the codified wisdom of 16th-century Chinese herbalists (who rated liquorice and spurge as a toxic pair of plant products, never to be used together) and (2) the combination of the two products in the recipe for Gansui-Banxia-Tang, Zhang et al (2016) [21] deduced that the variable, unstandardised blends of chemicals in the two plants must be antagonistic in one ratio and synergistic in another. Thus they unleashed the full armamentarium of Systems Biology to characterize this postulated interaction.

I claim no expertise in this methodology but I do confess to wondering whether the plotting software would come up with equally convincing conceptual gearwheels if the input tables of protein interactions and herbal constituents were replaced with random numbers. A greater cause for concern, though, is that the histology slides of liver and kidney tissue -- illustrating various levels of toxicity for each antagonistic / synergistic ratio -- overlap substantially with the figures in a previous paper by the same authors on the same liquorice / spurge combination (Lin et al 2016 [18]) but labeled differently. I am also concerned that some hipster brewery will come out with a Gansui-Banxia-Tang Antagonism Ale using liquorice and spurge as substitutes for hops.

[21] was retracted in August 2018 for "some unreliable data in the experimental validations", although the retraction is not recorded in the journal's webpages.

This is our introduction to the oeuvre of Dr Na Lin, of the Institute of Chinese Materia Medica (part of the Chinese Academy of Chinese Medical Sciences) and the Academy of Integrative Medicine at Fujian University of Traditional Chinese Medicine. From Zhang et al (2013) onwards, a number of her papers have followed a similar in-silico network-analytical approach, with the laudable goal of harmonising the diagnostic constructs of classical TCM with the radically different conceptual framework of contemporary global medicine. With their plethora of networks and scarcity of new data, these papers need not concern us here, for they have not drawn fire from the vast disembodied intellects who frequent and contribute to PubPeer. They do invite questions, though... if the classificatory constructs of TCM are so helpful in choosing appropriate treatments, and if TCM treatments are so efficacious, why is it necessary to synthesize or enhance so much of the visual evidence in the digital realm?

As I’ve asked before, if #TCM works, why do proponents have to fake so much research?

— Cheshire (@Thatsregrettab1) March 31, 2020

Dr Lin's own contributions to PubPeer discussions are concentrated in this thread. The study in question (Xu et al 2014a [8]) is one of a series in which classical TCM recipes were tested as estrogen substitutes, Hormone Replacement Therapy for rodents with no ovary function (due to immaturity or ovariectomy), to address the major health crisis of menopausal mice. In this case the recipe was QBMR, Qibaomeiran... the rationale for expecting QBMR to "antagonize estrogen decline" was its therapeutic rubric as "kidney strengthening". The illustration in question compares two purportedly different smears of vagina epithelial cells. Hat tip to Dr Elisabeth Bik for noticing this, and many of the other shenanigans I cite below.

Lin and co-author Ying Xu offered explanations for the identical portions of the two panels (and the clear evidence that one or other is a manipulated copy) which amounted to restatements of the method.

#9 Ying Xu

Thanks for your suggestions. The vaginal smear method to examine vagina cells is classical to study estrogen like activity and is simple and easy to prepare, which was recorded in the literature books of pharmacological experimental methodology and estrogen activity research references. The general steps to prepare the smear are as below:

1. washed rat's vagina using small amount of physiological saline to obtain vaginal epithelial and dried.
2. Fixed the vaginal lavage with 95% ethanol for 10 min.
3. Stained vaginal epidermal cells with Methylene Blue for 10 min.
4. Observed under microscopy the keratinized vaginal cells which indicates the estrus of rats.

Thanks

Na Lin & Ying Xu

These failed to convince any readers, but PubPeer readers don't get a vote on retractions and the real target for persuasion was the journal's editors. Rejuvenation Research is published by Mary Ann Lieber Inc and edited by Aubrey de Grey.

A companion paper reported that "Short-term QBMR exerts estrogenic activities without side effects" (Xu et al 2015 [16]). Despite the promise of no side effects, something seriously wrong happened to the tissues of the murine subjects, turning their antibody-stained histology slices into the pictorial equivalent of a three-minute techno track that was extended and looped into a 15-minute dance-floor remix.




The Western Blots were also affected.

The earliest in this series of Mouse Menopause collaborations with Ying Xu were a pair, administering Ginseng Root to ovariectomised mice (Xu et al 2014b [9]) and to immature mice (Jie et al 2015 [13]). One stained tissue slice appears in both papers but conceivably it came from a mouse marked by immaturity and ovariectomy.

Dance-floor-remix microphotographs are a recurring feature in this series. In Xu et al (2016a) [19], Veratrum Nigrum paired up with Ginseng root to fight crime estrogen decline, and something happened to Figure S3.

We are now back in the territory of Tooth-Fairy-Meets-the-Easter-Bunny Science, with studies of the Incompatibility Mechanisms through which Veratrum Nigrum antagonises other plant roots. In Xu et al (2019a) [26], incompatibility with Paeony roots "focused on estrogen-estrogen receptor pathways", while in Xu et al (2018) [24] it "inhibits the estrogenic activity of Salvia miltiorrhiza Bunge". In short, V. nigrum is almost as antagonistic as our host Dr Schneider.

Perhaps the scanner broke down when the authors were preparing Figures for [26], causing the ER(α)- and ER(β)-stained microphotographs to resemble exercises in brick-laying.



Only a fraction of the repeating graphic motifs are marked. Other Figures have lower resolution but even so they seem to have arisen through a process of crystallisation rather than biological activity.

I have my own theory for how V. nigrum might inhibit the effects of other herbs, a suggestion requiring no network mapping of protein-protein-interaction; i.e., it makes the patient vomit them up. But I am not a TCM practitioner. Anyway, [26] is best read in conjunction with Xu et al (2019b) [27], a companion paper devoted to the estrogen-receptor expression up-regulation of Paeony root in isolation. For the crystallisation had progressed further there and in the regularity of their repeating features, the Figures could be scenes from Ballard's masterpiece.



By the same token, [24] is best read in conjunction with its companion paper Xu et al (2016b) [20], addressing how Salvia miltiorrhiza Bunge in isolation "increases estrogen level without side effects". Again, the crystallisation of vagina histology is not considered as a side effect.

It may be that the Western Blots of Fig 6C in [20] are of more interest [left]. The 11 lanes in each WB are unlabeled, so it is unclear how they relate to the eight conditions in the quantitative analyses derived from them. At least when the same blots appear in Fig 6B of [24] [right] to illustrate different experiments, there are 11 experimental conditions so the relationship is straightforward.



This whole series of papers built to a crescendo in Zhao et al (2019) [28], where QingYan Formula was the potion under scrutiny for its hormone-replacement potential (QingYan Formula was also the topic of Zheng et al 2019 [25]). These panels do not depict the foreheads and hairlines of teenage zombies, with teenage acne covered with nearly-invisible Elastoplasts... in fact the colored rectangles and cartouches are there to draw attention to repeated clusters of cells.

The generous approach to the image in this series of papers is to explain the less-repetitive examples as authentic microphotographs, representative of actual experiments, that have felt the touch of the digital airbrush for aesthetic reasons. On the other hand, they could be 100% digital creations, constructed in such haste that a few careless glitches and wallpaper mismatches remain... this raises the possibility that other images are equally bogus, but assembled more carefully. It is all a question of whether to trust the authors' promise that data were collected as claimed... authors, you'll recall, who are currently playing dumb and flatly denying any problems with their photoshopped vaginal epithelial-cell smear from [8].

#13 Ying Xu

Thanks for your constant concerns. Currently what we can tell is that the cells are not identical from the difference elements inside shown in the pictures as we discussed. Your concerns are interesting to us too, the chance for such similarities of cells in appearance may be less as you insisted but it does happen in practice. As to why the similarity of cells in this picture is remarkable and how it happens, frankly speaking, currently we have no knowledge of it and also don’t know how to reappear it because our experiment purpose did not focus on the details of similarities but to study the estrous of rats. Therefore, your kind understandings on this concern are really appreciated and we will take your concerns into consideration in our future research.

Thanks

Na Lin and Ying Xu

Now one could argue that all these PubPeer threads are a waste of brains and bandwidth, since TCM is too easy a target and not worth the trouble of critiquing, like shooting fish in a barrel. This may be true but the fish are valuable and well-funded. The production of this series of papers earned US$ 82,000 for Ying Xu for 2016-2019, in the form of Grant #81573632 from the National Natural Science Foundation of China. The conversion of this amount into $$$ per image forgery is left as an exercise for the reader.

I'll belatedly begin at the beginning. Na Lin appeared on the international-journal scene as first author of two papers: Lin, Sato & Ito (2001) [1], and Lin et al (2003) [2], with Akira Ito from the School of Pharmacy at Tokyo University of Pharmacy and Life Sciences as corresponding and senior author. Presumably Lin was a guest or visiting researcher at TUPLS. [1] and [2] dwelt respectively on triptolide (which appears frequently in the TCM literature as a phytochemical in search of an application) and on nobiletin from citrus peel, both in the context of joint inflammation... finding biomedical applications for citrus fruit flavenoids was a research interest of Prof. Ito at the time, and a source of funding.


The two phytochemicals were studied in a single experiment, if the common lanes in the papers' illustrative Northern Blots are any guide; then salami-sliced for separate publication. Other Northern and Western Blots are also composite in nature, with clear vertical splice-lines (and horizontal ones too) hinting that lanes shown as contiguous were not originally so... for all I know, that may have been fine by the standards of image presentation in that benighted era. though Ito's other publications appear to be blameless.



A later paper with similar authorship details is more problematic: Lin et al (2007) [3]. Even at this early date, Figure 6B(b) "NF-kB p65 mRNA in cartilage" displays all the hallmarks of digital enhancement, and glares disapprovingly at the reader in what I choose to believe is a homage to Magritte.



In Fig 1C a slice of mouse knee appears in the right-hand panel to illustrate how well the triptoline treatment alleviated the arthritis induced by collagen injection. Its condition is indeed similar to the Control slice shown in the left-hand panel. In fact they are overlapping details of a single image (with a horizontal flip and a slight rotation) .

But wait, there's more! The same image appears a third time in Fig 3D of Lin et al (2008) [4] (rotated through 90°) to illustrate how well nobiletin alleviates collagen-induced arthritis.

No papers featuring Akira Ito as corresponding author have appeared for the last few years and it may be that he or she has retired, leaving no-one to correspond with. The leadership of the Tokyo University of Pharmacy and Life Sciences have not yet responded to an invitation to comment on these papers.

Anyway, these early papers provide a convenient segue to a second significant theme in Na Lin's research oeuvre: the use of different TCM preparations to treat bone and joint disease (in mice). I have not looked into the funding for this project but suffice to say that an aging population and (more to the point) an aging leadership cadre can enormously boost the importance of curing arthritis and osteoporosis.

For all the apparent solidity of vertebrate skeletons, they are in a perpetual state of flux, with one cell population eroding the bone in a delicate balance with a second set constantly reinforcing it. Rodent physiology seems to lend itself to easy disruption of this balance, in turn creating the opportunity to alleviate the resulting disease model with the herbal recipe of choice. Also there are stem-cells involved, and Tooth-Fairy Science becomes even more sciency with the addition of stem-cells. Welcome, then, to Xu et al (2014c) [10] where Sanmiao Formula restored the state of rat cartilage after surgical trauma and forestalled the onset of osteoarthritis.

Multiplicated large and small clusters of cells betray the use of the clone-stamp tool to produce these microphotographs; either that, or very tiny cats left trails of pawprints across them.

In Guo et al (2016) [17] the condition was rheumatoid arthritis and the cure was Guizhi-Shaoyao-Zhimu decoction (GSZD). Two panels of Fig 7 were the same apart from a vertical stretch.

Kong et al (2015a) [14] and Kong et al (2015b) [15] are companion papers, both deploying saponins from Anemone flaccida against bone resorption. The inhibition of activation of cellular messenger pathways is the explanatory mechanism for a Tooth-Fairy in the form of the abrogation or merely the suppression of osteoclast differentiation. The funding body will be reassured by extensive repurposing of WB loading controls in a show of thrift to prove that research grants were not wasted.

Xu et al (2014d) [11] and Xu et al (2014e) [12] added a refinement to Tooth Fair Science by introducing the concept of Medicinal Guides, i.e. by postulating that the Tooth Fairy has assistants which guide stem cells to the place in the body where they activity is needed (a task they are incapable of performing by themselves). Rats were afflicted with surgically-induced osteoarthritis, as in [10], but in these two papers it was treated with Ermiao recipe, with Angelicae pubescentis root and Herba Asari Mandshurici (Xixin) respectively in the role of Spirit-Guide helping stem cells to home from bone marrow to the site of the trauma. A distiller of my acquaintance swears by Angelica root as a gin botanical but she has never mentioned Xixin.

The similarities among the Figures of the papers are extensive. The miniature cats have left paw-prints all through them. Close inspection hints at the amount of work invested in constructing the images, so one can understand the authors' desire to get maximum usage from them.

Broaching the topic of bone-marrow stem (stromal) cells gives me an excuse to mention Kong et al (2016) [22] and Kong et al (2017) [23], where we learn of the Janus-faced nature of Huogu formula. For its aqueous fraction and ethyl alcohol fraction both promote the differentiation of those stromal cells, via the BMP and Wnt signaling pathways, but in different directions (osteogenic and lipogenic differentiation respectively). What concerns us here is the overlaps between scenes of cell cultures, within and between the two studies. No rats were injured during their construction.

All this has made demands of your patience but it is building up to a climax. Jiang et al (2014a) [5] administered steroids to rats while alleviating the resulting femural-head osteonecrosis, with ABE (Achyranthes bidentata extract). I have concerns about Fig 2, where bone-marrow slices from different conditions overlap... but the overlaps are not identical, with artisanal hand-drawn changes to one or the other.

In good news for steroid-abusing rodents, it turns out that Pravastatin also prevents steroid-induced femural-head osteonecrosis in rats, by a different cellular-messaging pathway (Jiang et al 2014b [6]). I would have more confidence in the biological veracity of the illustrations if it were not for the repeated graphic components.

Mirabile dictu, that second messaging pathway also enables Huogu Ⅰ formula to block steroid-induced femural-head osteonecrosis in rats (Jiang et al 2014c [7])! And the Figures in [7] are just wild. If you only follow one link to a PubPeer thread to see all the Figures omitted here, make it this thread.



Someone spent a lot of time stitching these quilts together from scraps of leopard-print fabric.

H/t Phycodrys Adamsiae

It is as if they are daring the peer-reviewers of the Journal of Traditional Chinese Medicine: "Yes, we made all of this up. Are you gonna do anything about it?"

For some reason, the loading-control band used to calibrate the levels of four proteins in Fig 5(b) of [6] appears in different reflections and rotations. Maybe it was dropped on the floor repeatedly. The same GAPDH loading control also appears (reflected and rotated) in Fig 4A-C of [7], although the lanes are necessarily different.



The loading-control band in Fig 7C of [5] has its own secret life.

Finally we come to Fig 2 of [7]. An avenue of bonsai trees, caught in car headlights? No, these are "representative images of micro-CT reconstructed 3-D microangiography of proximal femur" for differently-treated groups, constructed in Photoshop if the repeated fractal elements are any guide.

Conceivably all three osteonecrosis-preventing drugs could have been studied in a single experiment, then salami-sliced for separate publication, allowing the appearance of the same Control and Steroid-only (Model) images in Fig 2 of [6]. But the results for Huogu 15 g/kg and Prava 20 mg/kg are also more similar than one might expect.

Who predicted that treatment with Achyranthes bidentata extract will again, produce identical Bonsai trees reconstructions of femur vasculature in Fig 4A of [5]?

I have not yet exhausted the scientific investigation of Tooth Fairies in Na Lin threads at PubPeer. The same paradigm of Injury-model / TCM therapy was applied to antipyretic treatments, using yeast-injection pyrexia. Rats underwent various surgical interventions so that the resulting chronic pain could be treated with Wu-Tou decoction; or were purged with Epsom salts so that the diarrhea could be treated with rhubarb. But bored now. I prepared a spreadsheet for interested readers.

If any reader wants something more extreme, there is always Chinmedomics.

SOURCES

[1]. “Triptolide, a novel diterpenoid triepoxide from Tripterygium wilfordii Hook. f., suppresses the production and gene expression of pro-matrix metalloproteinases 1 and 3 and augments those of tissue inhibitors of metalloproteinases 1 and 2 in human synovial fibroblasts“, Na Lin, Takashi Sato, Akira Ito. Arthritis & Rheumatism (2001) doi: 10.1002/1529-0131(200109)44:9<2193::aid-art373>3.0.co;2-5

[2]. “Novel anti-inflammatory actions of nobiletin, a citrus polymethoxy flavonoid, on human synovial fibroblasts and mouse macrophages“, Na Lin, Takashi Sato, Yuji Takayama, Yoshihiro Mimaki, Yutaka Sashida, Masamichi Yano, Akira Ito. Biochemical Pharmacology (2003) doi: 10.1016/s0006-2952(03)00203-x

[3]. “Triptolide, a diterpenoid triepoxide, suppresses inflammation and cartilage destruction in collagen-induced arthritis mice“, Na Lin, Chunfang Liu, Cheng Xiao, Hongwei Jia, Keisuke Imada, Hao Wu, Akira Ito. Biochemical Pharmacology (2007) doi: 10.1016/j.bcp.2006.08.027

[4]. “Nobiletin, a citrus polymethoxy flavonoid, suppresses gene expression and production of aggrecanases-1 and -2 in collagen-induced arthritic mice“, Keisuke Imada, Na Lin, Chunfang Liu, Aiping Lu, Weiheng Chen, Masamichi Yano, Takashi Sato, Akira Ito. Biochemical and Biophysical Research Communications (2008) doi: 10.1016/j.bbrc.2008.05.171

[5]. “Achyranthes bidentata extract exerts osteoprotective effects on steroid-induced osteonecrosis of the femoral head in rats by regulating RANKL/RANK/OPG signaling“, Yini Jiang, Yanqiong Zhang, Weiheng Chen, Chunfang Liu, Xiaomin Li, Danni Sun, Zhenli Liu, Ying Xu, Xia Mao, Qiuyan Guo, Na Lin. Journal of Translational Medicine (2014a) doi: 10.1186/s12967-014-0334-7

[6]. “Pravastatin prevents steroid-induced osteonecrosis in rats by suppressing PPARγ expression and activating Wnt signaling pathway“, Yini Jiang, Yanqiong Zhang, Haojun Zhang, Bin Zhu, Ping Li, Chao Lu, Ying Xu, Weiheng Chen, Na Lin. Experimental Biology and Medicine (2014b) doi: 10.1177/1535370213519215

[7]. “Huogu I formula prevents steroid-induced osteonecrosis in rats by down-regulating PPARγ expression and activating Wnt/LRP5/β-catenin signaling“, Yini Jiang, Daobing Liu, Xiangying Kong, Ying Xu, Weiheng Chen, Na Lin. Journal of Traditional Chinese Medicine (2014c) doi: 10.1016/s0254-6272(14)60100-x

[8]. “Treatment with qibaomeiran, a kidney-invigorating Chinese herbal formula, antagonizes estrogen decline in ovariectomized rats“, Ying Xu, Xiao-ping Ma, Jie Ding, Zhen-li Liu, Zhi-qian Song, Hong-ning Liu, Na Lin. Rejuvenation Research (2014a) doi: 10.1089/rej.2014.1557

[9]. “Treatment with Panax ginseng antagonizes the estrogen decline in ovariectomized mice“, Ying Xu, Jie Ding, Xiao-Ping Ma, Ying-Hui Ma, Zhi-Qiang Liu, Na Lin. International Journal of Molecular Sciences (2014b) doi: 10.3390/ijms15057827

[10]. “Sanmiao formula inhibits chondrocyte apoptosis and cartilage matrix degradation in a rat model of osteoarthritis“, YING XU, GUO-JING DAI, QIAN LIU, ZHEN-LI LIU, ZHI-QIAN SONG, LI LI, WEI-HENG CHEN, NA LIN. Experimental and Therapeutic Medicine (2014c) doi: 10.3892/etm.2014.1862

[11]. “Effect of Ermiao Recipe|二妙方| with medicinal guide Angelicae Pubescentis Radix on promoting the homing of bone marrow stem cells to treat cartilage damage in osteoarthritis rats“, Ying Xu, Guo-jing Dai, Qian Liu, Xiao-ping Ma, Li Li, Wei-heng Chen, Na Lin. Chinese Journal of Integrative Medicine (2014d) doi: 10.1007/s11655-014-1761-2

[12]. “Effect of Ermiao Fang with Xixin|Herba Asari Mandshurici| on bone marrow stem cell directional homing to a focal zone in an osteoarthritis rat model“, Ying Xu, Guojing Dai, Qian Liu, Hongwei Zhu, Weiheng Chen, Ping Zhang, Tiejun Zhao, Na Lin. Journal of Traditional Chinese Medicine (2014e) doi: 10.1016/s0254-6272(15)30050-9

[13]. “Estrogenic effect of the extract of Renshen (Radix Ginseng) on reproductive tissues in immature mice“, Ding Jie, Xu Ying, Ma Xiaoping, An Jinna, Yang Xiudong, Liu Zhiqiang, Lin Na. Journal of traditional Chinese medicin (2015) doi: 10.1016/s0254-6272(15)30125-4

[14]. “Total saponin from Anemone flaccida Fr. Schmidt abrogates osteoclast differentiation and bone resorption via the inhibition of RANKL-induced NF-κB, JNK and p38 MAPKs activation“, Xiangying Kong, Wenbin Wu, Yue Yang, Hongye Wan, Xiaomin Li, Michun Zhong, Hongyan Zhao, Xiaohui Su, Shiwei Jia, Dahong Ju, Na Lin. Journal of Translational Medicine (2015a) doi: 10.1186/s12967-015-0440-1

[15]. “Triterpenoid Saponin W3 from Anemone flaccida Suppresses Osteoclast Differentiation through Inhibiting Activation of MAPKs and NF-κB Pathways“, Xiangying Kong, Yue Yang, Wenbin Wu, Hongye Wan, Xiaomin Li, Michun Zhong, Xiaohui Su, Shiwei Jia, Na Lin. International Journal of Biological Sciences (2015b) doi: 10.7150/ijbs.12296

[1+]. “Short-time QiBaoMeiRan Formula Treatment Exerts Estrogenic Activities without Side Effects on Reproductive Tissues in Immature Mice“, Ying Xu, Xiao-ping Ma, Jin-na An, Zi-jia Zhang, Jie Ding, Ya-kun Qu, Zhen-li Liu, Na Lin. Scientific Reports (2015) doi: 10.1038/srep17436

[17]. “Guizhi-Shaoyao-Zhimu decoction attenuates rheumatoid arthritis partially by reversing inflammation-immune system imbalance“, Qiuyan Guo, Xia Mao, Yanqiong Zhang, Shuqin Meng, Yue Xi, Yi Ding, Xiaocun Zhang, Yuntao Dai, Xia Liu, Chao Wang, Yuting Li, Na Lin. Journal of Translational Medicine (2016) doi: 10.1186/s12967-016-0921-x

[18]. “Effect of Glycyrrhiza on the Diuretic Function of Euphorbia kansui: An Ascites Mouse Model“, Ya Lin, Yanqiong Zhang, Erxin Shang, Wenfang Lai, Hongwei Zhu, Yuhua Fang, Qingxia Qin, Haiyu Zhao, Na Lin. Evidence-based Complementary and Alternative Medicine (2016) doi: 10.1155/2016/7620817

[19]. “Effect of the Interaction of Veratrum Nigrum with Panax Ginseng on Estrogenic Activity In Vivo and In Vitro“, Ying Xu, Jie Ding, Jin-na An, Ya-kun Qu, Xin Li, Xiao-ping Ma, Yi-min Zhang, Guo-jing Dai, Na Lin. Scientific Reports (2016a) doi: 10.1038/srep26924

[20]. “Salvia miltiorrhiza bunge increases estrogen level without side effects on reproductive tissues in immature/ovariectomized mice“, Ying Xu, Ting Chen, Xin Li, Ya-kun Qu, Jin-na An, Hong-xia Zheng, Zi-jia Zhang, Na Lin. Aging (2016b) doi: 10.18632/aging.101145

[21]. “Revealing the Effects of the Herbal Pair of Euphorbia kansui and Glycyrrhiza on Hepatocellular Carcinoma Ascites with Integrating Network Target Analysis and Experimental Validation“, Yanqiong Zhang, Ya Lin, Haiyu Zhao, Qiuyan Guo, Chen Yan, Na Lin. International Journal of Biological Sciences (2016) doi: 10.7150/ijbs.14151

[22]. “Aqueous Fraction of Huogu Formula Promotes Osteogenic Differentiation of Bone Marrow Stromal Cells Through the BMP and Wnt Signaling Pathways“, Xiangying Kong, Xiaomin Li, Cun Zhang, Liuluan Zhu, Hongye Wan, Jia Zhu, Cuiling Liu, Hongchang Su, Qingxia Qin, Weiheng Chen, Na Lin. Rejuvenation Research (2016) doi: 10.1089/rej.2015.1795

[23]. “Ethyl acetate fraction of Huogu formula inhibits adipogenic differentiation of bone marrow stromal cells via the BMP and Wnt signaling pathways“, Xiangying Kong, Xiaomin Li, Cun Zhang, Liuluan Zhu, Chunfang Liu, Qingxia Qin, Cuiling Liu, Qianqian Wang, Jia Zhu, Xuan Wu, Hongye Wan, Weiheng Chen, Na Lin. International Journal of Biological Sciences (2017) doi: 10.7150/ijbs.18430

[24]. “Veratrum nigrum inhibits the estrogenic activity of salvia miltiorrhiza bunge in vivo and in vitro“, Ying Xu, Ting Chen, Xin Li, Qu Ya Kun, Jin Na An, Hong Xia Zheng, Yuan Zhao, Zi Jia Zhang, Na Lin. Phytomedicine (2018) doi: 10.1016/j.phymed.2018.03.038

[25]. “Long-time qingyan formula extract treatment exerts estrogenic activities on reproductive tissues without side effects in ovariectomized rats and via active ER to ERE-independent gene regulation“, Hong-Xia Zheng, Yuan Zhao, Ying Xu, Zi-Jia Zhang, Jing-jing Zhu, Yuan-Fang Fan, Na Lin. Aging (2019) doi: 10.18632/aging.102035

[26]. “Incompatibility Mechanism Between Radix Paeoniae Alba and Veratrum nigrum Focusing on Estrogen-Estrogen Receptor Pathway in Immature/Ovariectomized Mice“, Ying Xu, Xin Li, Ting Chen, Ya-kun Qu, Hong-xia Zheng, Zi-jia Zhang, Yuan Zhao, Na Lin. Rejuvenation Research (2019a) doi: 10.1089/rej.2017.2026

[27]. “Radix Paeoniae Alba increases serum estrogen level and up-regulates estrogen receptor expression in uterus and vagina of immature/ovariectomized mice“, Ying Xu, Xin Li, Ting Chen, Ya-kun Qu, Hong-xia Zheng, Zi-jia Zhang, Yuan Zhao, Na Lin. Phytotherapy Research (2019b) doi: 10.1002/ptr.6205

[28]. “Estrogenic Effect of the Extract of QingYan Formula on Reproductive Tissues in Immature Mice“, Yuan Zhao, Hong-Xia Zheng, Ying Xu, Na Lin. Evidence-based Complementary and Alternative Medicine (2019) doi: 10.1155/2019/5493714