Xinysun Instrument (Shanghai) Co., Ltd.

Applications

Breadcrumb: Home > Applications > Lab Research

Gardenia Extract Against UVB Photo-Damage: UCL-3200M Dual-Dose Irradiation Protocol

Updated:2026-11-02Views:

In April 2025, a study from China Pharmaceutical University and Shanghai Junyu on gardenia extract against UVB photodamage appeared in Cosmetics. The exposure source recorded in the method section was a 302 nm UV crosslinker, which in the LUYOR product line is the UCL-3200M. This article unpacks the exposure scheme: why one study needed both 0.01 J/cm² for HaCaT cells and 0.4 J/cm² for a 3D epidermis model, why the low-dose end requires energy mode rather than timing, three operating details that decide whether the run works, and how the UCL-3200M and UCL-3500M divide the work. The recorded doses are 10 mJ/cm² and 400 mJ/cm², a fortyfold span.

In April 2025, the Laboratory of Skin Health and Cosmetic R&D Evaluation at China Pharmaceutical University, working with Shanghai Junyu Biotechnology, published a study in MDPI Cosmetics on gardenia fruit extract (Gardeniae Fructus extract, GFE) against UVB photodamage (Cosmetics 2025, 12(2), 72, open access). The method section records the exposure source as "UV crosslinker, Luyor (Shanghai)" at 302 nm - which in the LUYOR product line is the UCL-3200M .

From an instrumentation point of view, one feature of this study is worth unpacking: the same study used two doses - 10 mJ/cm² for the cell model and 400 mJ/cm² for the 3D epidermis model, a fortyfold difference. For one crosslinker to be accurate at both the 0.01 J/cm² end and the 0.4 J/cm² end, what matters is not "how many seconds" but the energy integral. This article reconstructs the exposure scheme, the dose logic and the operating points.

1. Why one study used two doses

The two doses correspond to two models, and the difference lies in how many layers light crosses before reaching living cells:

  • Two-dimensional cell model: HaCaT human keratinocytes grow as an adherent monolayer covered by a thin layer of PBS. With no stratum corneum in the way, UVB acts almost directly on the cells, and 10 mJ/cm² is already enough to push oxidative stress and inflammatory markers into a significant difference.
  • Three-dimensional epidermis model: a reconstructed human epidermis (Regenovo) cultured at the air-liquid interface, with a stratum corneum and a stratified structure. The stratum corneum is the main attenuating layer for UVB, so an insufficient dose never reaches the living cell layer, and 400 mJ/cm² is needed.

In other words, the dose difference is not arbitrary; it is derived backwards from the barrier structure of the model. That is practical for anyone doing efficacy evaluation: change the model and the dose has to be reset, and copying numbers from the literature often wastes a run.

2. The exposure protocol in the paper

Item Cell model (HaCaT) 3D epidermis model
Exposure source LUYOR UV crosslinker (Luyor, 302 nm) Same UVB source (the model is not restated in that section)
Wavelength 302 nm (UVB) 302 nm (UVB)
Dose 10 mJ/cm² (0.01 J/cm²) 400 mJ/cm² (0.4 J/cm²)
Medium during exposure Thin PBS layer over the cells Submerged in culture medium
Post-exposure handling Fresh medium, culture continued for 24 h GFE 50 μg/mL for 24 h, then readout
Readouts ROS, T-SOD, CAT, GSH-Px, MDA; IL-6, IL-1β FLG, LOR, IVL immunofluorescence (barrier proteins)

The paper gives the dose but not the exposure time or working distance. That is precisely the point: the authors recorded a dose, not a time. A crosslinker is a closed chamber and the samples sit at fixed positions inside it, so there is no adjustable "lamp distance" to report; dose is the number that genuinely belongs in the method section.

3. Why the 0.01 J/cm² end requires energy mode

UCL-3200M has an energy range of 0.000-9.999 J/cm², and 10 mJ/cm² converts to 0.010 J/cm², one thousandth of the range. That position is sensitive to the control method:

  1. The time is down to seconds. At the irradiance typical of 6 × 8 W lamps inside a chamber (a few mW/cm²), 10 mJ/cm² corresponds to an exposure of a few seconds. A person with a stopwatch who presses half a second early or late can introduce a dose error above 10%.
  2. The lamps age. For the same "5 seconds of exposure", a new lamp and a lamp two years old can deliver very different doses, and nothing shows on the outside.
  3. Low-dose experiments have no margin. Too low a dose and the injury does not develop, so the model group and the control group do not separate; too high a dose and the photodamage is excessive, so the protective effect of the test article is swamped. Either way the conclusion is distorted.

That is what energy mode is for: the microprocessor monitors UV intensity in the chamber during exposure and integrates over time, stopping as soon as the set dose accumulates. If the lamp dims it exposes longer; if intensity is high it stops earlier - what goes into the record is "10 mJ/cm²", a figure that can be checked, rather than "exposed it for a moment".

4. At 0.4 J/cm², what is saved is time arithmetic

400 mJ/cm² is 40 times the cell end, and the exposure moves from seconds to tens of seconds or minutes. On the face of it, time mode would work, but the same problems appear: 3D model batches are run across different days, lamp condition and chamber temperature change, and a time-based protocol does not line up between batches - which shows up as vague complaints such as "the FLG staining came out faint in this batch".

The chamber is 32×26×15 cm, so a full 12-well Transwell plate fits flat and is exposed in one pass, which reduces within-batch variation. Another advantage of energy mode at the high-dose end is that no conversion is needed: whether the target is 10 or 400 mJ/cm², the setting is the same physical quantity, and changing model or project does not mean feeling out the time again.

5. Beyond dose: three operating details that decide the outcome

  1. Keep the liquid layer thickness consistent. The paper covered the cell model with a thin PBS layer and submerged the 3D model in culture medium. Liquid attenuates UVB, so an extra millimeter of liquid removes part of the dose. The same batch must keep the same volume, and the volume belongs in the procedure.
  2. Change the medium immediately after exposure. Riboflavin, tryptophan and similar components of culture medium generate reactive oxygen species under UV, while PBS contains no such photosensitizers. The paper replaced the medium after exposure, which both restores nutrients and clears the products of medium photochemistry, so the source of damage stays clean.
  3. Keep placement fixed. The chamber has mirror stainless steel walls with aluminum diffuse reflectors, and the UV sensor sits above the chamber behind a quartz guard. The uniformity design reduces variation within the chamber, but where the sample sits is still a variable - fix the placement area, keep multi-plate samples near the center, use the same arrangement across batches and record it.

6. Composition and readouts: what the study measured

The composition of GFE was characterized by UPLC-MS/MS: 10 iridoids were identified, with geniposide prominent among them, alongside genipin-1-gentiobioside, gardenoside acid, shanzhiside methyl ester, deacetylasperulosidic acid and asperuloside. Extraction used 20 volumes of water at 80°C, twice for 2 hours each, followed by activated carbon decolorization, macroporous resin purification, ethanol elution and freeze-drying.

The readouts fall into four layers, from the surface inward:

Layer Marker Change in the UVB model Effect of GFE
Oxidative stress ROS (DCFH-DA) Increased Significantly suppressed at 12.5/25/50 μg/mL
Antioxidant enzymes T-SOD, CAT, GSH-Px Decreased Recovered effectively
Lipid peroxidation MDA Increased Reduced
Inflammatory cytokines IL-6, IL-1β (ELISA) Secretion increased Significantly suppressed
Barrier proteins FLG, LOR, IVL (3D model immunofluorescence) Expression downregulated Upregulated after 50 μg/mL for 24 h

7. Mechanism clues: the cell cycle keeps coming up

The paper cross-validated with proteomics and network pharmacology, and the numbers are worth noting:

  • Proteomics: between the control and UV groups, 234 proteins were upregulated and 186 downregulated, and KEGG enrichment pointed to the cell cycle; GSEA showed UVB activating the G2/M checkpoint. After GFE treatment the differentially expressed proteins narrowed to 38 upregulated and 36 downregulated, and the UVB-activated G2/M checkpoint was significantly downregulated.
  • Network pharmacology: 201 photodamage-related genes and 203 iridoid-related genes, with 24 in common; the compound-target network had 31 nodes and 40 edges, with geniposide among the leading linked compounds. Hub targets above the average in the PPI network were IL1B, MMP9, TP53, BCL2 and SIRT1.
  • Molecular docking: the binding energy of geniposide with p53 was -6.0 kcal/mol, with 2 hydrogen bonds to Leu145 and Glu221 (below -5 kcal/mol is usually taken as good binding).

The authors also state clearly in their conclusion that these mechanisms are predictions and leads that still require experimental verification. When writing efficacy claims, treating them as directions for later validation rather than established conclusions is the safer course.

8. How the UCL-3200M and UCL-3500M divide the work

The same series also includes a 302 nm UCL-3500M , and the site already carries an application article on it at a 50 mJ/cm² dose, on retinol and DNA repair (see the article on 302 nm UVB quantitative exposure ). Both can deliver quantitative UVB exposure; the choice depends on sample volume:

Item UCL-3200M UCL-3500M
Lamp configuration 6 × 8W 6 × 15W
Chamber size 32×26×15 cm 44×34×15 cm
Energy range 0.000-9.999 J/cm², 9 presets 0.000-9.999 J/cm², 9 presets
Suited to One plate or a few samples, a wide dose span, fine control at the low-dose end Multiple plates in parallel, large models, batch exposure
Corresponding case This article (10 and 400 mJ/cm²) 50 mJ/cm² skin photodamage and DNA repair

If the project runs mainly on 96-well plates and just one or two plates per batch, the 3200M chamber is a good fit; if a dozen plates or large ex vivo skin samples are exposed at once, the 3500M has more room. The wavelength version system is the same for both: M is 302 nm (UVB), the standard version is 254 nm (UV-C), and L is 365 nm (UV-A), so a version change keeps the same control logic and preset method.

9. Frequently asked questions

The paper does not name the model; how do you tell it is a UCL-3200M?

The method section says "UV crosslinker, Luyor (Shanghai)" plus a 302 nm wavelength, without a model number. The LUYOR series is split by wavelength into three versions: 254 nm is the standard version, 302 nm is M and 365 nm is L. Matching 302 nm gives the UCL-3200M. If your own paper needs a model number, state the wavelength and the dose as well - those two are what make the work reproducible.

At 10 mJ/cm², can energy mode be set accurately enough?

The UCL-3200M energy range is 0.000-9.999 J/cm², so 10 mJ/cm² is 0.010 J/cm², resolved to 1 mJ/cm². The key at the low-dose end is to avoid time mode: the exposure lasts a few seconds and manual timing carries a large relative error. In energy mode the sensor integrates to the set dose and stops, which is what makes the dose stable.

Why does the 3D epidermis model need 400 mJ/cm², 40 times the cell dose?

Because of the added stratum corneum. Two-dimensional HaCaT monolayers have a thin PBS layer over them and UVB reaches the cells almost directly; a reconstructed human epidermis has a stratum corneum and a stratified structure, and UVB is markedly attenuated before it reaches the living cell layer. The dose has to be reset to the barrier structure of the model, not copied.

Should the culture medium be removed during exposure?

It depends on the model. The paper kept a thin PBS layer on the cell model to prevent drying, and submerged the 3D model in culture medium. Either way, the liquid volume must be the same across a batch - liquid thickness directly changes the dose reaching the sample. Replace the medium after exposure, which also clears reactive oxygen species generated by medium photochemistry.

Can the 254 nm UCL-3200 do this kind of UVB work?

Not as a direct substitute. 254 nm is UV-C, with stronger DNA absorption and a different damage mechanism and penetration depth, so results cannot be compared with UVB literature. A UVB photodamage model needs the 302 nm version (UCL-3200M); the 254 nm version suits nucleic acid membrane crosslinking, PCR decontamination and UV mutagenesis, as described in the article on UV mutation breeding.

Related reading

Literature: Zong K, Li X, Zhou F, Dong J, Huang Q, Wu J. Photoprotective Effect and Potential Mechanisms of Gardeniae Fructus Extract in UVB-Irradiated HaCaT Cells. Cosmetics 2025, 12(2), 72. DOI: 10.3390/cosmetics12020072 (CC BY open access). All data in this article are taken from that paper; for specific experimental conditions, refer to the original paper and the instrument manual.

Related sections and models

This article belongs to the laboratory research techniques section.

Other sections on the same product line include UV transilluminators (gel and thin-layer chromatography excited from below, so bands can be read without turning the gel over), UV crosslinkers (dose set in J/cm², the figure that matters for both nucleic acid crosslinking and materials curing), and life science instruments (the entry point for fluorescence imaging, UV crosslinking, UV transmission and cell irradiation).

Models mentioned in this article: UCL-3200, LUV-450, XEPU-1126. Range, accuracy and accessory lists for each model are given on its product page.

  • Official WeChat QR code

    Email: info@xinysun.com

  • Tel: 131-2093-6518

  • Address: 1525 Minqiang Road, Bldg 26, Songjiang, Shanghai

Official WeChat QR code Follow

Official WeChat QR code

Contact

131-2093-6518

Top