Abstract
Drug discovery for obesity treatment, particularly bodily slimming, is a topic of timely importance that requires continued investigation, as the current therapies have limited efficacy with many adverse effects. Obesity is associated with adipose tissue expansion, where the size and number of adipocytes increase. Over the past few decades, high-throughput/content screening (HTS/HCS) has been carried out on morphological changes in adipose tissues and adipocytes for the development of anti-obesity therapies. Increased understating of current adipocyte-based HTS/HCS technology will facilitate drug screening for obesity and weight control.
Keywords
Abbreviation
2Dtwo-dimensional
3Dthree-dimensional
ADSCadipose-derived stem cell
AGPAT21-acylglycerol-3-phosphate O-acyltransferase 2
ASCadipose derived mesenchymal stromal cell
BAbeige adipocyte
BATbrown adipose tissue
BMIbody mass index
BODIPYboron-dipyrromethene
BSCL2BSCL2 lipid droplet biogenesis associated, seipin
CAS9CRISPR-associated protein 9
CRISPRclustered regularly interspaced short palindromic repeats
COXcyclooxygenase
DAPI4’,6-diamidino-2-phenylindole
GFPgreen fluorescent protein
HCShigh content drug screening
HTShigh throughput drug screening
iPSCinduced pluripotent stem cell
JARID2jumonji and AT-rich interaction domain containing 2
LDlipid droplet
TFtranscription factor
UCPuncoupling protein
WATwhite adipose tissue
1. Introduction
Obesity is recognized as a worldwide health problem [
[1]
], and nearly a third of the world's population is now classified as overweight or obese [[2]
]. People are concerned about not only obesity but also bodily slimming as a cause of health complaints. Body mass index (BMI) is currently used to classify obesity, defined as a BMI ≥ 30 kg/m2 [[1]
]. Human obesity measured by the BMI alone is insufficiently sensitive to measure a person's risk; their fat distribution, as reflected by the waist circumference and waist-to-hip ratio, plays at least as prominent a role [[3]
]. The waist circumference has proven to be a better marker of abdominal fat accumulation than the BMI [[4]
].1.1 Obesity and adipose tissue characteristics
Obesity is highly associated with the expansion of subcutaneous and visceral adipose tissue, including an increase in the number and size of adipocytes [
[5]
]. Human adipose depots can be classified into white adipose tissue (WAT) and brown adipose tissue (BAT), which are responsible for energy storage and body temperature maintenance, respectively [[6]
]. Brown adipocytes contain an extremely high number of mitochondria with the specific membrane protein uncoupling protein (UCP)1, which uncouples respiration from adenosine triphosphate synthesis and provokes energy dissipation in the form of heat [[7]
]. Recently, beige adipocytes (BAs) were identified as distinct thermogenic fat cells [[8]
]. The adipose tissue, considered an endocrine organ, plays a central role in the control of metabolism [[9]
], particularly in regulating the whole-body energy, glucose homeostasis, and insulin sensitivity [[10]
]. Moreover, the central fat, mostly intra-abdominal fat, is more medically important than the subcutaneous truncal fat [[3]
]. In this context, candidate pharmaceuticals that decrease the number or size of visceral adipocytes hold great potential in anti-obesity drug discovery.1.2 Current obesity treatment
Research into and development of pharmaceutical means of body weight control are promising. However, current anti-obesity drugs have many disadvantages, and their effects are often unsatisfactory, failing to meet the market demand [
[11]
,[12]
]. Yet, the risk and price of bariatric surgery, the most effective long-term treatment for severe obesity, are much higher than those of oral therapy [[11]
], which means that the introduction of a novel weight-loss therapy resulting from drug discovery has great safety and economic value.1.3 Current in vitro models to mimic obesity and its limitations
Cultures of the mouse (e.g., 3T3-L1 and C3H/10T1/2) and human (e.g., PAZ6, and LiSa-2)-differentiated preadipocyte cell lines are extensively used as in vitro models [
[13]
]. White, beige, and brown adipocytes differ in their size, morphology, handling of lipids, and thermogenic capacity [[13]
]. Primary cells, such as mouse embryonic fibroblasts and human adipose-derived stem cells (ADSCs), have been established to mimic white, beige, and brown adipocytes [[13]
]. Isolated rat mature adipocytes from rat epididymal fat pads have also been used as a classical in vitro WAT adipocyte model [[14]
]. To circumvent the limitations of two-dimensional (2D) cultures, methods have been developed to investigate mature adipocytes directly, including scaffold-embedded adipocytes, ceiling cultures, adipose tissue explant cultures, etc [[13]
]. To get closer to mature cells in native tissues, spheroid cultures have been developed for differentiation of progenitor cells or culture of differentiated cells in three-dimensional (3D) [[13]
], including 3T3-L1-based spheroid cultures, human ADSC-based 3D cultures, and stromal vascular fraction-based 3D cultures [15
, 16
, 17
, 18
].The mature WAT adipocytes in vivo contain a large unilocular lipid droplet (LD), but in vitro differentiation of primary and cell line-derived preadipocytes usually leads to the accumulation of small multilocular LDs [
[13]
]. Compared to 2D culture, 3D-cultured cells exhibit higher differentiation levels, as indicated by increased adipocyte gene expression and unilocular LD formation, much more closely to the adipocyte function and morphology observed in vivo [[13]
]. For human adipocytes, 3D culture systems could yield valuable insights into adipocyte biology discovery [[19]
]. However, 3D cultures are technically more challenging and time-consuming because, compared to 2D cultures [[13]
], individual aggregates are easily damaged and require the use of expensive plates or scaffolds.1.4 Relevant endpoints of adipogenesis
Adipogenesis represents a potential anti-obesity target during adipocyte differentiation [
[20]
]. Obesity is characterized by increased adipose tissue mass, and the expansion of fat depots can be driven by an increase in adipocyte size (hypertrophy) or by the differentiation of precursors during adipogenesis to form new adipocytes (hyperplasia) [[5]
]. Adipocyte differentiation (as assessed by LD accumulation) and brown adipocyte formation (as assessed by the formation of UCP1-positive cells) were both used to estimate the targets involved in anti-obesity treatment [[21]
]. While WAT is the major energy reserve in adults, there has been a dramatic increase in the incidence of obesity resulting from an excess of WAT [[22]
]. The candidate drugs which could directly reduce the number and size of intracellular LDs in WAT adipocytes should be a potential target in anti-obesity drug discovery. On the other hand, pharmacological agents that increase BAT thermogenesis may be considered a therapy for obesity and diabetes, which may help to achieve long-lasting weight loss and an improved metabolic profile [[23]
]. Upregulation of UCP1 by genetic manipulations or pharmacological agents was demonstrated to reduce obesity and improve insulin sensitivity [[24]
]. In addition, lipolysis, expression of adiponectin, and leptin should be considered indirect markers for anti-obesity drug screening, using an in vitro adipocyte model [25
, 26
, 27
].1.5 New technologies that can favor the screening of new compounds
High-throughput/content screening (HTS/HCS) is a useful drug development tool that has been introduced in recent years [
[28]
]. High-throughput drug screening combined with image analysis technology quickly and efficiently screens potential candidates by analyzing changes in their cell morphology. High-throughput/content screening has the power to analyze intracellular LD accumulation, which is related to the increased cell size of adipocytes. It can accurately measure changes in cell numbers by quantifying the nuclei of adipocytes. As such, HTS/HCS is critical to analyzing the expansion of adipose tissue, and could thus be applied in anti-obesity drug discovery.2. Current status of adipocyte-based HTS/HCS for anti-obesity drug discovery
We aimed to assess the current status of HTS/HCS for the development of anti-obesity drugs through analysis of the accumulation of intracellular lipid droplets in adipocytes. We screened the database PubMed (https://pubmed.ncbi.nlm.nih.gov/) of existing literature as of January 10, 2022, with the following inclusion criteria: HTS/HCS technology, adipocyte-based morphological analysis, and using an HTS/HCS reader; and with the following exclusion criteria: not for obesity treatment, not for drug screening or discovery, and not written in English (Table S1). The search strategy is set out in Table S2. We included 61 references in the selection process (Table S3) and summarized our findings after reviewing them. Our findings can be divided into those relating to drug screening and technologies of HTS/HCS, and others concerning the new analytical HTS/HCS methods.
2.1 Technologies of adipocyte-based HTS/HCS for anti-obesity drug discovery
There are several adipocyte models of HTS/HCS for anti-obesity drug discovery, as shown in Table 1 [
29
, 30
, 31
, 32
, 33
, 34
, 35
, 36
]. For the extensively used murine 3T3-L1 preadipocyte line, plasmids constructed of peroxisome proliferator-activated receptor gamma isoform-2 promoter-derived enhanced green fluorescent protein (GFP) reporter and adiponectin promoter-derived luciferase reporter were used to generate the cell models applied in HTS/HCS for anti-obesity drug discovery [[29]
,[30]
]. For the precursor cell models, primary human ADSCs were used to represent a physiologically relevant cell system to use in screening for anti-obesity drug discovery [[31]
]. Adipose-derived mesenchymal stromal cells (ASCs) and induced pluripotent stem cell (iPSC)-derived cells were also suggested for application in high-throughput drug screening, particularly in obesity research [[32]
,[33]
]. For the newly defined BAs, Singh et al. provided an HTS/HCS model of BAs from human adipose-derived stem/stromal cells in the serum-free medium, indicating the potential utility of BAs as a tool for the identifying drugs to treat metabolic diseases [[34]
]. However, for drug discovery, HTS/HCS requires a large number of adipocytes. Using different stem cell- or progenitor cell-derived adipocytes may lead to inconsistency in the analysis of the cause of cell differentiation. Further investigation is required to identify the ideal adipocyte model in HTS/HCS.Table 1Technologies of HTS/HCS for anti-obesity drug discovery.
| Author | Year | The model | Cell type | Novelty | Application |
|---|---|---|---|---|---|
| The adipocyte models | |||||
| Singh et al. | 2020 | The generation of beige adipocytes (BAs) from human adipose derived stem/stromal cells (ADSCs) in serum-free medium | Human ADSC-derived beige adipocytes | The authors show the therapeutic utility of BAs in a platform for high-throughput drug screening. | The potential utility of BAs was as a cell therapeutic and as a tool for the identification of drugs to treat metabolic diseases. |
| Yuan et al. | 2019 | A novel adipocyte quantification algorithm, named Fast Adipogenesis Tracking System (FATS), based on computer vision libraries | 3T3-L1 preadipocytes, adipose derived mesenchymal stem cells (ASCs), and induced pluripotent stem cell (iPSC)-derived cells | The FATS algorithm is versatile and capable of accurately detecting and quantifying the percentage of cells undergoing adipogenic and browning differentiation even under difficult conditions such as the presence of large cell clumps or high cell densities. | The FATS offered a universal and automated image-based method to quantify adipocyte differentiation of different cell lines in both standard and high-throughput workflows. |
| Eom et al. | 2018 | A mesenchymal cell has switched its fate to an adipogenic lineage | ASCs | Immuno-labelling against fatty acid binding protein-4 (FABP4), a lineage-specific marker of adipogenic differentiation, enabled visualization and quantification of differentiated cells. | The high-throughput FABP4 assay provides a quantitative assay for assessing the differentiation potential of patient-derived cells and is a robust tool for comparing different isolation and expansion methods. |
| Buehrer et al. | 2012 | Primary human adipose-derived stem cells represent a physiologically relevant cell system | primary human ASCs | By using the cells in a primary screen, the risk and cost of identifying artifacts due to interspecies variation and immortalized cell lines were eliminated. | Counterscreening with human primary stem cells from distinct adipose depots can be routinely performed to identify tissue-specific responses. |
| Hino et al. | 2011 | A reporter-based high-throughput screening (HTS) assay using insulin-resistant-mimic 3T3-L1 adipocytes | 3T3-L1 preadipocytes | The authors cloned 3T3-L1 preadipocytes stably expressing a vector that has a luciferase gene under the control of a human adiponectin promoter. | This HTS assay might be applicable to screening for other adipokine modulators that can be useful for the treatment of other conditions. |
| Li et al. | 2004 | A preadipocyte cell line expressing EGFP under the control of the promoter of adipocyte-specific expression gene peroxisome proliferator-activated receptor (PPAR) γ2 was generated | 3T3-L1 preadipocytes | A plasmid of pPPARgamma2-promoter-EGFP was constructed by inserting a 660bp sequence of mouse PPARgamma2 promoter into the Ase I and Kpn I sites of pEGFP-N3 and transferred into 3T3-L1 preadipocyte cells. | The cells were induced to differentiate and the expression of PPARγ2 was detected by the microscopic observation of EGFP and by RT-PCR assays. |
| The adipocyte culture methods | |||||
| Louis et al. | 2022 | A statistically validated high-throughput screening model by seeding human mature adipocytes from patients, encapsulated in physiological collagen microfibers | Abdominal human adipose tissues, breast adipose tissues | The drop tissues ensured the maintenance of adipocyte viability and functionality for controlling glucose and fatty acids uptake, as well as glycerol release. | The developed model of vascularized and non-vascularized adipose drop tissues should be compatible with high throughput formats for adipose-targeting drug screening, displaying a wide range of measurable functional outputs. |
| Hsiao et al. | 2016 | The differentiation of ADSCs to mature adipocytes in highly observable and highly handleable 3D fiber shaped constructs exhibiting morphologies and functions of native adipose tissues | ADSCs | Using the cell fiber technology, ADSCs were encapsulated in hydrogel microfibers, allowed to form into fiber-shaped constructs, and differentiated to mature unilocular adipocytes. | The adipocyte fibers make them suitable for biological studies, high-throughput drug screening/testing, and clinical applications. |
Several adipocyte-based HTS/HCS studies have applied innovative cell culture methods. Using cell fiber technology, ADSCs were transformed into mature unilocular adipocytes by encapsulating them in hydrogel microfibers, which allowed them to form fiber-shaped constructs [
[35]
]. Louis et al. provided an adipose drop model that adopted mature human adipocytes encapsulated in physiological collagen microfibers for HTS application in obesity-related drug screening [[36]
]. Taken together, the improvements in adipocyte culture methods may provide new insights into adipocyte-based HTS/HCS for anti-obesity drug discovery.2.2 Relevant studies of adipocyte-based HTS/HCS for anti-obesity drug discovery
The studies that applied HTS/HCS for anti-obesity drug discovery are listed in Table 2. Among these, some studies performed non-adipocyte HTS/HCS research that did not correspond with adipocyte-based morphological analysis, e.g., HeLa cells [
[37]
]. human embryonic kidney 293-T cells [[38]
], committed skeletal myoblasts C2C12 cells [[39]
], and COS-7 cells [[40]
]. In addition, an HTS/HCS reader was also applied to the analysis of morphological changes in adipocytes without drug screening [[41]
]. Lau et al. used an HTS/HCS reader to analyze and quantify the expression of specific intracellular proteins in adipocytes [[41]
]. It resolved the traditional dilemma of estimating the distribution and expression of intracellular proteins at the same time and may therefore indirectly replace immunoblotting. This approach indicates a new direction for the future application of HTS/HCS technology.Table 2Application of HTS/HCS for anti-obesity drug discovery.
| Author | Year | Library | Cell-based models | Adipocyte-based | Target screening | Image-based HTS/HCS | Description |
|---|---|---|---|---|---|---|---|
| Adipocyte-based HTS/HCS for anti-obesity drug discovery | |||||||
| Björk et al. | 2021 | A specific small interfering RNA pool | Human subcutaneous adipose tissue | yes | yes | yes | The role in the differentiation of human adipose tissue-derived stem cells (hASC) was investigated by an RNA interference (RNAi) screen. |
| Jiao et al. | 2019 | The lentiviral array (125 metabolic disease genes) | Human Simpson-Golabi-Behmel syndrome (SGBS) preadipocytes | yes | yes | yes | The top 125 genes were ablated in human pre-adipocytes via CRISPR/CAS9 and the resulting cellular phenotypes quantified during adipocyte differentiation with high-content microscopy and automated image analysis. |
| Qiu et al. | 2018 | FDA-approved Drug Library (Selleck, L1300) | Immortalized brown adipocytes derived from Ucp1-2A-GFP reporter mouses | yes | yes | yes | The authors have generated a Ucp1-2A-GFP reporter mouse, in which GFP intensity serves as a surrogate of the endogenous expression level of UCP1 protein; and immortalized brown adipocytes were derived from this mouse model and applied in drug screening. |
| Perdikari et al. | 2017 | InhibitorSelect Protein Kinase Inhibitor Library I, II, and III (Merck Millipore; catalog nos. 539744, 539745, and 539746), lentiviral shRNA knockdown | Mouse immortalized brown preadipocytes | yes | yes | yes | The authors performed lentiviral-mediated short hairpin knockdown or used pharmacological inhibitors in a high-content and high-throughput in vitro image-based screen. |
| Application of HTS/HCS in adipocytes for anti-obesity drug discovery | |||||||
| Shamsi et al. | 2020 | A protein library containing more than 5000 mammalian secreted peptides | Murine brown adipose tissue (BAT) preadipocytes | yes | yes | no | The authors identify two fibroblast growth factors (FGF), FGF6 and FGF9, as potent inducers of UCP1 expression in adipocytes and preadipocytes using high-throughput library screening of secreted peptides. |
| Hsu et al. | 2018 | 105 traditional Chinese herbs | 3T3-L1 preadipocytes | yes | yes | no | The authors performed an HTS assay to monitor galectin-12 promoter activity using 105 traditional Chinese herbs. |
| Imran et al. | 2018 | The compound library (854 compounds) | 3T3-L1 preadipocytes | yes | yes | no | The candidate was selected by screening as a small-molecule inducer of adipocyte differentiation among 854 candidates by using C3H10T1/2 mesenchymal stem cell. |
| Park et al. | 2015 | 560 natural product extracts (from Korea Chemical Bank) | Human and mouse11β-hydroxysteroid dehydrogenase (HSD) 1 overexpressing CHO-K1 cells or adipocytes | yes | yes | no | The authors conducted HTS of active natural product extracts from the Korea Chemical Bank, including Tanshinone I, Tanshinone IIA, and flavanone derivatives, and 2-and 3-phenyl-4H-chromen-4-one. |
| Söhle et al. | 2012 | A druggable siRNA library (containing 23,352 unique siRNAs targeting 7,784 human genes) | Subcutaneous human preadipocytes | yes | yes | no | The authors performed a high throughput siRNA screen with primary (pre)adipocytes, using a druggable siRNA library targeting 7,784 human genes. |
| Buehrer et al. | 2012 | A collection of 580 botanical extracts derived from Central Asian plants | human primary adipose-derived stromal/stem cells (ASCs) | yes | yes | no | Use of primary human ASCs represents a physiologically relevant cell system to screen for bioactive agents in the prevention and treatment of obesity and its related complications. |
| Hino et al. | 2011 | A library of approximately 100 000 small-molecule compounds | Insulin-resistant-mimic 3T3-L1 adipocytes | yes | yes | no | The authors report on the development of a reporter-based HTS assay using insulin-resistant-mimic 3T3-L1 adipocytes for the discovery of adiponectin secretion modulators. |
| Tiller et al. | 2009 | A cDNA clone collection with about 35,000 clones and a metabolic cDNA library with 200,000 clones | 3T3-L1 preadipocytes | yes | yes | no | An automated cDNA screen was established to identify secreted human proteins with an inhibitory effect on adipocyte differentiation and a potential inhibitory effect on adipose tissue growth. |
| Nayagam et al. | 2006 | A total of 147,000 compounds | 3T3-L1 preadipocytes | yes | yes | no | The authors identified compounds with sirtuin (SIRT) 1 activating and inhibiting potential using HTS. |
| The non-adipocyte HTS/HCS system for anti-obesity drug discovery | |||||||
| Jo et al. | 2021 | An approximately 3000-member in-house drug-like poly-heterocycle library constructed using a privileged substructure-based diversity-oriented synthesis (pDOS) strategy | HeLa cells | no | yes | yes | The authors performed image-based high content screening with a fluorogenic bioprobe (SF44), which visualizes cellular lipid droplets (LDs), to identify initial hit compounds. |
| Pettersson-Klein et al. | 2018 | The compound library (7,040 compounds) | Human embryonic kidney (HEK) 293-T cells transfected with the pEGFPC1-mPGC-1a1 plasmid | no | yes | yes | The authors designed a cell-based high-throughput screening (HTS) system to identify peroxisome proliferator-activated receptor-gamma coactivator (PGC)-1α1 protein stabilizers. |
| Nie et al. | 2017 | About 20,000 chemicals from the NIH library, Sigma LOPAC1280 library, ENZO fatty acid library (BML-2803), Bioactive lipid library (BML-2800), and S.D. lab in-house libraries. | Committed skeletal myoblasts C2C12 cells | no | yes | yes | The authors identified a retinoid X receptor (RXR) agonist, bexarotene (Bex), that efficiently converted myoblasts into brown adipocyte-like cells using high-throughput phenotypic screening to induce brown adipocyte reprogramming in committed myoblasts. |
| Jang et al. | 2012 | A library of low MW compounds was generated by individual synthesis to 95% purity at the Korean Research Institute of Chemical Technology (KRICT) | COS-7 cells | no | yes | yes | HTS was performed using a library of low MW compounds to identify the transcriptional co-activator with PDZ-binding motif (TAZ) modulators that enhance nuclear TAZ localization. |
| Application of an HTS/HCS reader on analysis of morphological changes in adipocytes | |||||||
| Lau et al. | 2019 | N/A | 3T3-L1 preadipocytes | yes | no | yes | Adipocytes determination by fluorescent dye (Nile Red) staining and protein expressions (PRDM16, PGC1-α, UCP1/2/3, and anti-FNDC5) of brown adipocyte markers were determined through HCS analysis. |
The concept of HTS/HCS was applied in several studies to adipocyte drug screening [
[29]
,[31]
,42
, 43
, 44
, 45
, 46
, 47
, 48
], but only a few performed adipocyte-based drug screening by quantifying and analyzing the intracellular lipid droplets with an HTS/HCS reader. Four selected studies used HTS/HCS to measure the intracellular lipid accumulation or expression of UCP1; therefore, they are included as the focus of this review study [[21]
,49
, 50
, 51
]. As can be seen, no study has focused on quantifying the size and number of lipid droplets. However, in the years to come, this will increase the complexity of analysis, confirming that the accumulation of lipid droplets in adipocytes could elevate the accuracy of drug screening.2.3 Current use of HTS/HCS technologies to screen and develop anti-obesity drugs
We found that few studies have attempted to perform adipocyte-based HTS/HCS for anti-obesity drug discovery. A summary of HTS/HCS based on an analysis of the morphological changes in adipocytes is shown in Table 3. Björk et al. systematically screened adipogenic human transcription factors (TFs) that are relevant for metabolic conditions [
[49]
]. In human ADSCs, boron-dipyrromethene (BODIPY)-labeled neutral lipids were assayed using the Acumen eX3 imager, to determine the siRNAs targeted to 148 TFs [[49]
,[52]
]. RNAi screening identified 39 genes involved in adipocyte differentiation, where 11 genes were novel, including the novel hit TF jumonji and AT-Rich interaction domain containing 2 (JARID2) [[49]
]. Meanwhile, Jiao et al. sought to identify novel disease-gene interactions during adipocyte differentiation in an in vitro Simpson-Golabi-Behmel syndrome pre-adipocyte model, using BODIPY staining for intracellular lipid accumulation and 4’,6-diamidino-2-phenylindole (DAPI) for nuclei [[50]
]. Morphometric measurements were taken using an Opera Phenix High-Content Screening System [[50]
]. A group of 14 gene ablations via CRISPR/CAS9, including two lipodystrophy genes BSCL2 Lipid Droplet Biogenesis Associated, Seipin (BSCL2), and 1-acylglycerol-3-phosphate O-acyltransferase 2 (AGPAT2), in human pre-adipocytes were selected based on decreased lipid accumulation in 125 candidates, indicating that a morphometric approach in adipocytes can resolve multiple cellular mechanisms for metabolic disease loci [[50]
]. Both of these studies explored the genes involved in the regulation of adipocyte differentiation, which may represent new targets for future anti-obesity drug discovery.Table 3Summary of adipocyte-based HTS/HCS for anti-obesity drug discovery.
| Authors | Year | Library | Adipocyte Type | Optimization | Detector | Hits | DOI |
|---|---|---|---|---|---|---|---|
| Björk et al. | 2021 | siRNAs containing he order numbers and sequences of the 148 selected transcriptional regulators | Human adipose-derived stem cells (hADSCs) | Bodipy (neutral lipids), Hoechst (DNA) | Acumen eX3 imager (TTP Labtech) | 39 genes that affected fat cell differentiation in vitro | 10.1210/endocr/bqab096 |
| Jiao et al. | 2019 | Selected 125 genes (3 sgRNA/gene) | Simpson Golabi Behmel Syndrome (SGBS) pre-adipocytes | Bodipy (lipid), DAPI (nuclei) | Opera PhenixTM (PerkinElmer) | A group of 14 genes characterized by decreased lipid accumulation | 10.1016/j.molmet.2019.03.001 |
| Qiu et al. | 2018 | FDA-approved library (∼1000 drugs) | Murine Ucp1-2A-GFP preadipocytes | GFP and DAPI (nuclei) | Cellomics ArrayScan VTI (Thermo Fisher Scientific) | 42 drugs that may activate UCP1 expression in brown adipocytes | 10.1016/j.ebiom.2018.10.019 |
| Perdikari et al. | 2017 | InhibitorSelect Protein Kinase Inhibitor Library I, II, III, and lentiviral shRNA knockdown (total 657 kinases) | Immortalized brown preadipocyte (IBA) | Hoechst (nuclei), LD540 (lipid droplets), second Alexa 488–conjugated rabbit antibody (UCP1) | Operetta (PerkinElmer) | 19 kinases (brown adipocyte proliferation), 19 kinases (brown adipocyte differentiation), and 152 kinases (brown adipocyte formation) | 10.1126/scisignal.aaf5357 |
By analysis of UCP1 expression, Qiu et al. generated immortalized brown adipocytes derived from UCP1-2A-GFP reporter mice to screen for FDA-approved drugs that may activate endogenous UCP1 expression in adipocytes [
[51]
]. They found 42 drugs, including sunitinib, clofazimine, acetylcysteine, and several inhibitors to cyclooxygenase (COX), that may activate UCP1 expression and potentially treat obesity in brown adipocytes by using a Cellomics ArrayScan reader [[51]
]. Further to this, Perdikari et al. tested a total of 657 kinases to analyze the effect of kinase signaling on brown adipocyte formation, using an inhibitor library and lentiviral short hairpin RNA knockdown in immortalized brown preadipocyte [[21]
]. Lipid droplets and expression of UCP1 were analyzed by fluorescent (LD540) and immunofluorescent (Alexa 488-conjugated secondary antibody) staining [[21]
]. In total, 190 kinases were identified that either stimulated or inhibited brown adipocyte proliferation, differentiation, or formation, e.g., 5’-AMP-activated protein kinase (AMPK) promoted the formation of brown adipocytes abundant in UCP1 [[21]
]. These studies highlighted potential candidates and pharmacological mechanisms in the development of anti-obesity therapies. Taken together, only one study pointed to a candidate list of FDA-approved drugs (42 in total) that may have an indirect anti-obesity effect through activation of UCP1 expression [[51]
]. Further candidates that may be able to reduce intracellular lipid droplet accumulation and prevent the expansion of adipose tissue must be investigated using adipocyte-based HTS/HCS.3. Perspectives and conclusion
3.1 Perspectives
We identified all existing studies related to HTS/HCS technology in adipocytes. Most studies that claimed to screen drugs by analyzing the morphological changes of adipocytes did not use an HTS/HCS reader (Table 2). Some studies used an HTS/HCS reader for drug screening but not for the adipocyte model. We suggest that adipocytes should be adopted for the screening of anti-obesity drugs because the number and size of LDs in adipocytes are believed to be directly related to obesity and slimming [
[5]
]. Conversely, if drug screening is carried out in a non-adipocyte model, the results may not be directly applicable to obesity treatment.The four studies we found that met the criteria of our search in the literature did not explain the development of HTS/HCS experiments (Table 3). Three studies provided original data but did not specify the design and overall process of drug screening, nor did they indicate what experimental results were used to set the planting density, treatment concentration, and treatment time of adipocytes [
[21]
,[49]
,[50]
]. From the information provided in the studies, it is difficult to understand how the drug screening platform was established. The lack of complete data for different experiments prevented big data sharing in the HTS/HCS drug screening process.Through RNA interference and HTS/HCS technology, we were able to detect the potential pharmacological action mechanism of new drugs [
[49]
,[50]
]. The novel hits AMPK (a regulator of brown adipocyte formation), JARID2 (regulating the expression of multiple fat cell phenotype-specific genes), BSCL2 (protein-protein interaction with perilipin 1), and AGPAT2 (genetic interaction with CCAAT enhancer binding protein alpha) were discovered in the involvement of adipogenesis and adipocyte metabolism [[21]
,[49]
,[50]
]. However, when using marketed drugs or small molecule inhibitors for drug screening, it should be remembered that the use of effective drug screening at the cellular level may not have a direct effect on the human body [[21]
,[51]
]. For example, sunitinib (multi-target receptor tyrosine kinase inhibitor), clofazimine (phospholipase activator), acetylcysteine (reactive oxygen species inhibitor), and several cox inhibitors were found to have potential in obesity treatment [[51]
]. However, clofazimine and acetylcysteine act as prodrugs [[53]
,[54]
]. Drugs that need to be metabolized by the patient should be excluded as far as possible.In the selection of adipocyte types, immortal cell lines are relatively stable, and stem cell-derived adipose cells are closer to the real state of adipocytes. Although both have advantages and disadvantages, it is suggested that testing be started from the cell line to find drugs that regulate lipid metabolism, followed by classifying them biologically and pharmacologically, analyzing the most likely candidate drugs according to the pharmacological mechanism or action target, and then conducting follow-up analyses. At this time, we can consider screening with a stem cell-derived or 3D-cultured adipocyte model and repeating the procedure several times if there is no ideal drug. In addition, to overcome a heterogeneous condition of obesity with polygenic aetiology associated with neurobehavioral, endocrine, and metabolic causes, primary isolated mature adipocytes from different individuals might be a potential in vitro model for high-throughput anti-obesity drug screening [
[12]
].The dosage and time of drug treatment in the process of HTS/HCS drug screening are important because drugs that can rapidly reduce LDs in adipocytes are very likely to have cytotoxicity to cells. The drugs screened may not have clinical application value when cytotoxicity cannot be ruled out. Therefore, the drug concentration and treatment time should be thoroughly tested, and the most suitable conditions should be found from multiple measurements. Given the limited number of HTS/HCS experiments, no single candidate may be able to play its full pharmacological role, and a group of candidates should be selected for further investigation.
The LDs of adipocytes are one of the most difficult quantitative cell images to analyze. Quantifying the number of adipocytes that are positive for LD and nucleus staining by automated image is a common method of analysis in HTS/HCS screening for anti-obesity drugs [
[21]
]. Unsupervised clustering of the morphologic profiles could be used to determine the potential group of hits, based on identifying LDs and nuclei, including related intensity, morphology, and texture parameters [[50]
]. Strictly standardized mean difference and mean fold change have also been used to select the candidates, in comparison with a negative control group [[49]
]. However, it is difficult and time-consuming to find an appropriate section of LDs suspended in the cytoplasm and focus automatically, and the effect may not be satisfactory, requiring a manual retest. Liquid droplets blocked by the nucleus are especially easily ignored. It may be assumed that the ratio of LDs blocked by the nucleus of each drug is similar, but candidates with obvious changes in the number of cells still need further attention in the analysis. It is worth noting that different thresholds should not be set during the experiment. It is meaningful to compare each treatment group with the control group after taking photos and analyzing them under the same conditions. In addition, for UCP1, identifying UCP1-positive cells and the ratio of GFP/DAPI signal have both been used to quantify the degree of brown adipocyte formation [[21]
,[51]
]. Although the expression of UCP1 fluorescence is relatively easy to measure, it is a surrogate endpoint. Therefore, it is necessary to decide how to select an adipocyte model according to the research purpose and the drug screening platform design.The HTS/HCS reader is difficult to obtain, and the cost is high, making its use uneconomical for a single study. However, entrusting the analysis to another group or body costs much more than doing it oneself. Thus, the problem of the instrument should be considered at the outset of the research. A confocal HTS/HCS reader can calculate three-dimensional volume, but the relative time is long and the file is extremely large. Even if it uses automatic focusing, it may fail to focus, owing to the small number of cells. For a confocal HTS/HCS reader, which consumes a lot of time and material resources, the results may not be good. Compared with the confocal reader, a non-confocal one has time to scan all areas of each well in 96/384-well plates, therefore it can be used to quantify and analyze the cell survival rate, which is also an advantage. Whichever type is chosen, it is advisable to check again manually, especially in groups with large changes in the number of cells.
In the quantification of intracellular LDs [
[55]
], we are able to calculate the average number of LDs or the total amount of fluorescence in adipocytes. However, if the number of cells is too small or too high, the conclusion will be incorrect. Even with HTS/HCS technologies, it is essential to manually confirm whether treatment with each drug is well photographed in each analysis. Sometimes, the procedure will be repeated many times because the image of a cell treated with some drugs is unsatisfactory, and there is no way to compare it with other drugs under the same screening conditions. After small-scale screening, large-scale screening is performed, requiring continuous testing. The Z-factor, a screening window coefficient, is used to measure whether experiments with small samples can be used in large experiments [[56]
]. If the image after drug treatment cannot be correctly analyzed, even after exhaustive efforts, this should be explained in the results and discussion. In large drug screening, many candidates do not meet the criteria for image analysis, but most studies ignore this point. Internal validation should be properly carried out, the ratio of effective analysis should be presented, and continuous testing should meet the condition of the most candidates to screen and analyze drugs more effectively. In addition, it is recommended that a viability dye, such as calcein AM, be added to ensure that the cells are still in a healthy state when reducing intercellular LDs. In this way, the selected candidates are better supported by the research.3.2 Conclusions
There have been few large-scale drug screening studies over the past few decades. Current adipocyte-based HTS/HTS cannot fully reflect the lipid content of adipocytes, which suggests that HTS technology still needs to be improved. However, combining HTS/HCS with 3D culture and confocal imaging is time-consuming and laborious, and subsequently leads to inconsistency in the analysis of data. The possible direction of the development of HTS/HCS for obesity treatment may be to quantify intracellular LD accumulation using adipocyte-based morphological analysis. Looking to the future, developing HTS/HCS technology will be critical to improving anti-obesity drug screening, and we should focus on screening specific candidates in a small-scale, economic, and efficient way.
Availability of data and materials
All data generated or analyzed during this study are included in the references and the supplementary materials.
Funding
This research was partially supported by the School of Pharmacy, Shanghai University of Medicine and Health Sciences (from Leo Tsui).
Conflict of interests
The author declares that he has no competing financial interests.
Appendix. Supplementary materials
References
- Obesity-induced changes in adipose tissue microenvironment and their impact on cardiovascular disease.Circ Res. 2016; 118: 1786-1807
- The epidemiology of obesity.Metabolism. 2019; 92: 6-10
- Human obesity: an evolutionary approach to understanding our bulging waistline.Diabetes Metab Res Rev. 2001; 17: 347-362
- Abdominal obesity and the metabolic syndrome: contribution to global cardiometabolic risk.Arterioscler Thromb Vasc Biol. 2008; 28: 1039-1049
- Adipogenesis and metabolic health.Nat Rev Mol Cell Biol. 2019; 20: 242-258
- Brown adipose tissue: Recent insights into development, metabolic function and therapeutic potential.Adipocyte. 2012; 1: 13-24
- Uncoupling protein 1 of brown adipocytes, the only uncoupler: a historical perspective.Front Endocrinol. 2011; 2: 85
- The common and distinct features of brown and beige adipocytes.Trends Endocrinol Metab. 2018; 29: 191-200
- Adipose tissue in control of metabolism.J Endocrinol. 2016; 231: R77-R99
- Adipose tissue regulates insulin sensitivity: role of adipogenesis, de novo lipogenesis and novel lipids.J Intern Med. 2016; 280: 465-475
- Current and future drug targets in weight management.Pharm Res. 2011; 28: 1792-1818
- Anti-obesity drug discovery: advances and challenges.Nat Rev Drug Discov. 2022; 21: 201-223
- In vitro and ex vivo models of adipocytes.Am J Physiol Cell Physiol. 2021; 320: C822-C841
- Metabolism of isolated fat cells. I. Effects of hormones on glucose metabolism and lipolysis.J Biol Chem. 1964; 239: 375-380
- A surface-tethered spheroid model for functional evaluation of 3T3-L1 adipocytes.Biotechnol Bioeng. 2014; 111: 174-183
- Engineered adipose tissue from human mesenchymal stem cells maintains predefined shape and dimension: implications in soft tissue augmentation and reconstruction.Tissue Eng. 2005; 11: 556-566
- Production of a new tissue-engineered adipose substitute from human adipose-derived stromal cells.Biomaterials. 2007; 28: 2850-2860
- Using a 3D culture system to differentiate visceral adipocytes in vitro.Endocrinology. 2015; 156: 4761-4768
- Three-dimensional adipocyte culture: the next frontier for adipocyte biology discovery.Endocrinology. 2015; 156: 4375-4376
- Adipogenesis as a potential anti-obesity target: a review of pharmacological treatment and natural products.Diabetes Metab Syndr Obes. 2021; 14: 67-83
- A high-throughput, image-based screen to identify kinases involved in brown adipocyte development.Sci Signal. 2017; : 10
- Understanding adipocyte differentiation.Physiol Rev. 1998; 78: 783-809
- Brown fat as a therapy for obesity and diabetes.Curr Opin Endocrinol Diabetes Obes. 2010; 17: 143-149
- UCP1: its involvement and utility in obesity.Int J Obesity. 2008; 32: S32-S38
- Regulation of lipolysis in adipocytes.Annu Rev Nutr. 2007; 27: 79-101
- New insight into adiponectin role in obesity and obesity-related diseases.Biomed Res Int. 2014; 2014658913
- Leptin and Obesity: Role and Clinical Implication.Front Endocrinol. 2021; 12585887
- Impact of high-throughput screening in biomedical research.Nat Rev Drug Discov. 2011; 10: 188-195
- High-throughput screening for small-molecule adiponectin secretion modulators.J Biomol Screen. 2011; 16: 628-636
- The labeling of 3T3-L1 preadipocyte cells with enhanced green fluorescent protein.Sheng Wu Gong Cheng Xue Bao. 2004; 20: 607-609
- Tools for the identification of bioactives impacting the metabolic syndrome: screening of a botanical extract library using subcutaneous and visceral human adipose-derived stem cell-based assays.J Nutr Biochem. 2012; 23: 519-525
- Visualization and Quantification of Mesenchymal Cell Adipogenic Differentiation Potential with a Lineage Specific Marker.J Vis Exp. 2018;
- Fast Adipogenesis Tracking System (FATS)-a robust, high-throughput, automation-ready adipogenesis quantification technique.Stem Cell Res Ther. 2019; 10: 38
- Human beige adipocytes for drug discovery and cell therapy in metabolic diseases.Nat Commun. 2020; 11: 2758
- 3D Tissue Formation of Unilocular Adipocytes in Hydrogel Microfibers.Adv Healthc Mater. 2016; 5: 548-556
- High-throughput drug screening models of mature adipose tissues which replicate the physiology of patients' Body Mass Index (BMI).Bioact Mater. 2022; 7: 227-241
- Phenotypic discovery of SB1501, an anti-obesity agent, through modulating mitochondrial activity.ChemMedChem. 2021; 16: 1104-1115
- Small molecule PGC-1α1 protein stabilizers induce adipocyte Ucp1 expression and uncoupled mitochondrial respiration.Mol Metab. 2018; 9: 28-42
- Brown Adipogenic Reprogramming Induced by a Small Molecule.Cell Rep. 2017; 18: 624-635
- TM-25659 enhances osteogenic differentiation and suppresses adipogenic differentiation by modulating the transcriptional co-activator TAZ.Br J Pharmacol. 2012; 165: 1584-1594
- Novel discovery of Averrhoa bilimbi ethanolic leaf extract in the stimulation of brown fat differentiation program in combating diet-induced obesity.BMC Complement Altern Med. 2019; 19: 243
- FGF6 and FGF9 regulate UCP1 expression independent of brown adipogenesis.Nat Commun. 2020; 11: 1421
- Galectin-12 is involved in corn silk-induced anti-adipogenesis and anti-obesity effects.Am J Chin Med. 2018; 46: 1045-1063
- A pivotal role of AMPK signaling in medicarpin-mediated formation of brown and beige.Biofactors. 2018; 44: 168-179
- Identification of a novel 11β-HSD1 inhibitor from a high-throughput screen of natural product extracts.Pharmacol Res. 2015; 102: 245-253
- Identification of new genes involved in human adipogenesis and fat storage.PLoS One. 2012; 7: e31193
- Effects of TWEAK (TNF superfamily member 12) on differentiation, metabolism, and secretory function of human primary preadipocytes and adipocytes.Endocrinology. 2009; 150: 5373-5383
- SIRT1 modulating compounds from high-throughput screening as anti-inflammatory and insulin-sensitizing agents.J Biomol Screen. 2006; 11: 959-967
- An RNAi screening of clinically relevant transcription factors regulating human adipogenesis and adipocyte metabolism.Endocrinology. 2021; : 162
- Discovering metabolic disease gene interactions by correlated effects on cellular morphology.Mol Metab. 2019; 24: 108-119
- Screening of FDA-approved drugs identifies sutent as a modulator of UCP1 expression in brown adipose tissue.EBioMed. 2018; 37: 344-355
- Effects of selected bioactive food compounds on human white adipocyte function.Nutr Metab. 2016; 13: 4
- Mode of Action of Clofazimine and Combination Therapy with Benzothiazinones against Mycobacterium tuberculosis.Antimicrob Agents Chemother. 2015; 59: 4457-4463
- High-dose oral N-acetylcysteine, a glutathione prodrug, modulates inflammation in cystic fibrosis.Proc Natl Acad Sci U S A. 2006; 103: 4628-4633
- Analysis and quantification of oxidized low-density lipoprotein-induced lipid droplets in macrophages through high-content screening: application for antiatherogenic drugs discovery.Assay Drug Dev Technol. 2019; 17: 223-230
- A simple statistical parameter for use in evaluation and validation of high throughput screening assays.J Biomol Screen. 1999; 4: 67-73
Article info
Publication history
Published online: August 07, 2022
Accepted:
August 5,
2022
Received in revised form:
July 15,
2022
Received:
May 22,
2022
Identification
Copyright
© 2022 The Author(s). Published by Elsevier Inc. on behalf of Society for Laboratory Automation and Screening.
User license
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
