Why did BPC-157 scratch closure plateau at day 14?

Sep 11 4190 views 45 posts

I ran a 14-day scratch assay with primary tenocytes, BPC-157 at 10 ng/mL vs vehicle. Days 0-6 closure was faster in treated wells, then both groups flattened. Western blot at day 10 showed PCNA up, but collagen I unchanged. pH 7.2, passage 3, n=4 wells per group. Is this contact inhibition, batch drift, or something about the assay? I expected continued migration. Anyone seen this biphasic pattern?

Contact inhibition is my first guess. Did you re-scratch teh same wells at day 7? Otherwise the closure metric saturates.

Not sure it's contact inihbition. I had a similar plateau in chondrocytes after re-scratch, and media glucose was under 1 g/L by day 8.

👍 1

The PCNA rise without collagen I is the odd part. Did you stain for CD90 or tenomodulin? Fibroblast overgrowth could mask tenocyte response.

Passage 3 can do that. My p3 tenocytes had lower basal collagen I than p1, vehicle included. Did you run p1 side by side?

👍 1 ❤️ 1

What was your FBS lot? We had two lots where migration stalled around day 6-7, not the peptide. New lot, effect shrank.

I ran a 12-day scratch, 8 wells/group, same 10 ng/mL, and saw an early boost. Re-scratch at day 6 kept a small gap, but vehicle also sped up, so I stopped trusting the first closure slope. My hunch is serum carryover plus edge effects. Did you use silicone inserts or pipette scratch? With pipette scratches, the first 24h edge can look cleaner and inflate closure. Also, what was vehicle closure at 24h compared with treated? If vehicle is already 60-70% closed, the early effect may be mostly serum and not the peptide.

Normalize to total protein. PCNA can track loading, not confluence.

Every 12–14 day flattening I've chased in a scratch assay turned out to be evaporation long before it was biology, and the tell is stupidly simple: check the outer ring versus the inner wells separately. We ran a 24-well scratch setup a while back and edge wells stalled around day 12–13 while inner wells kept crawling to day 17, with phenol red drifting orange and maybe 1–2 µL less volume by the end — if your day-10 lysates came off the perimeter wells, that alone looks like a treatment effect quietly dying. Second thing I'd poke at is the mask, not the cells: once the gap drops under roughly 300 µm the front cells start sliding over each other and a manual threshold calls it "closed" while the edge is still two or three layers thick, so reimage the exact same coordinates and ask whether the leading edge is still a single cell thick. And the boring one nobody likes — was your day-10 PCNA blot the same wells you scratched, or a sister plate? Primary tenocytes at P3 can flip toward a more synthetic, less collectively-migratory phenotype pretty fast, and a lot of what people read as contact inhibition is really "these aren't the same cells that started the assay." What do the day-14 edges actually look like up close, rounded and piled or flat and spread? (I mostly lurk here for the GLP side of things, this is just the day job.)

I'd push back on the contact-inhibition read — the PCNA-up/collagen-flat split smells like serum-driven dedifferentiation to me. I had primary tenocytes do exactly that: 10% FBS the whole run, PCNA blazing by day 10, collagen I barely a smudge, and closure flatlined around day 12–14. What unstuck it was dropping to 2% FBS after day 3, adding 50 µg/mL ascorbate-2-phosphate, and re-feeding every 48 h instead of every 72; collagen I finally showed up around day 12 and closure crawled to ~85% by day 18. If you're keeping high serum for the whole assay, the cells are happy to proliferate but not lay down matrix, and the scratch edge piles up enough to look contact-inhibited. What's your feeding schedule and are you scratching a mature monolayer or day-after-seeding? n=4 with primary tenocytes can also hide a lot of well-to-well variation if your day-0 widths aren't within ~10%.

👍 1

If collagen I was flat in lysate but closure stalled, I'd check the insoluble fraction. Tenocytes dump collagen into the matrix and it won't show on a standard RIPA blot. Acid extraction or hydroxyproline from the well might tell you more.

Also, did you re-scratch at day 7? Past that point you're not really measuring migration, you're measuring whatever survived P3 on plastic. They drift.

What coating are you on — plain plastic or collagen/fibronectin? That changes the migration vs proliferation read more than people admit.

👍 6

Scientist-brain aside (I only get lab time between snack runs): did you check mycoplasma? I had a primary-cell scratch plateau around day 12–14, treated wells looked “active” (PCNA up) while matrix markers stalled, and it turned out contaminated. Also worth measuring glucose/lactate at day 6/10 — pH 7.2 at harvest can hide a mid-run swing. What’s your media change schedule? With n=4 primary tenocytes, a small well-to-well nutrient dip can flatten both curves.

👍 5 ❤️ 1

yo, late-night plate checker here. did you log spent-media pH or phenol red hue on the non-imaged wells? i've had scratch assays go flat around d10-14 when local glycolytic waste acidified the monolayer, especially with primary cells at p3. starting pH 7.2 means nothing if it drifts after feeding. also check if the treated wells had more floating debris/cell clumps at the wound edge — area closure can stall while nuclei keep increasing, so DAPI count in the gap might split the difference. what's your media change schedule and did you use HEPES vs bicarbonate? that'd be my next check before blaming batch drift.

👍 2

Not a tenocyte person, but I did a lot of scratch assays before maintenance brain took over. Did you save day-0 images right after the scrape? If treated wells had cleaner edges or less debris, early closure can look better without real migration. At plateau, are cells piling up behind the front instead of filling it? A live-cell track or EdU pulse would separate motility from proliferation. Also, same FBS lot and plate coating throughout? Serum lot/coating changes can flatten tenocyte migration more than treatment does. Just my two cents.

👍 1

Hi, vey new here, so this may be off. Could it be nutrient depletion in the wells? Primary tenocytes can outpace media by ~day 10–14, and migration stalls before actual confluence. Did you measure glucose/lactate in spent media, or change media on a set schedule? I only track things casually, but when I’ve done scratch-type wound tracking, timing of media refresh changed everything. Also, were all wells from the same FBS lot and same passage split? Even small serum lot shifts can flatten the later curve. Did you randomize plate positions to rule out incubator gradients?

👍 5 ❤️ 1

Phase-contrast "open" area can lie. At that density tenocytes lay down provisional matrix, and you get cell-free draped layers that still read as a gap. Fix and stain — crystal violet or DAPI — before trusting the day-14 plateau; I've had wells look 40% open on phase and be fully covered on stain. Also, were you imaging fixed coordinates daily or re-finding the scratch each time? ROI drift alone can fake a plateau.

👍 3

Not tenocyte-specific, but I track assay runs like I track stock: same lot, same day, same tip. Did you use the same FBS lot and collagen-coating batch for both arms? A lot switch can bump PCNA without moving collagen I, then both flatten once cells hit edge confluence. Also, what was your initial gap width in µm, and did you normalize to each well’s own day 0? Uneven starting gaps can make a “plateau” that’s just slow wells pulling the mean. Log media volume per well too—edge evaporation can mimic contact inhibition by day 10–14.

👍 2 ❤️ 1

Tbh two kids here, so my brain is mostly snack-drawer logistics, but I’d check plate position/evaporation first. Fourteen days in a plate, outer wells can dry down and look stalled while inner wells keep going. Were treated and vehicle wells spread evenly across inner/outer wells, or was one group all in one row/column? If so, that’s a confound. Also straight lysate collagen I can miss deposited matrix—did you compare soluble vs insoluble?

👍 1

love the n=4 and day-by-day tracking. One thing I’d check: was the peptide a one-time spike at day 0, or refreshed with media changes? At 37°C in serum-containing medium, peptides can adsorb/degrade, so days 0–6 might jst be the window where active peptide is present; after that both groups look like vehicle. If you refresh, does the treated curve separate again? Also, a short proliferation-arrest or low-serum arm would tell you whether the late plateau is proliferation or just migration running out of steam. What’s your media-change schedule?

👍 2

Not tenocyte-specific, but I’d separate migration from proliferation before blaming contact inhibition. If PCNA is up while collagen I is flat, you may just have more cells piling up, not functional repair. A low-serum or proliferation-arrested scratch arm would tell you whether the day 6–14 flattening is migration stalling. Also, P3 primary tenocytes can drift; did you check tenomodulin/scleraxis vs P0? If they’re down, the plateau may be phenotype loss, not assay saturation. I track recovery like I track lifts—markers matter, but the outcome matters more.

👍 5

Did you add ascorbate/2-phosphate to the media? Primary tenocytes need it for collagen I; PCNA up with flat collagen I is a classic mismatch if it’s missing or oxidized. I’d also check edge wells: in my 24-well scratches, edge wells dry/osmolarity-shift faster and plateau 2–3 days before center wells. I map row/column and normalize each well to its own day-0 gap width, because tip angle can make initial gaps vary by 100+ µm. What’s your edge vs center n? If center wells also plateau, that points more to the assay/readout than contact inhibition.

👍 6 ❤️ 2

Sharp follow-up: did you randomize well position? I’ve been burned by incubator edge effects—my “treated” column dried a hair faster and flattened around day 12–14 while middle wells kept closing. I now map wlls like airport gates and always seed a center/edge control. Also, are you re-drawing the same ROI or letting software auto-threshold each day? Once tenocytes lay down matrix, auto-threshold can freeze the apparent open area even if cells are still moving. Lock exposure/threshold and see if the plateau shifts.

👍 1 ❤️ 1

Not tenocyte-specific, but did all four wells plateau, or is one jagged-edge well dragging the group mean? I’m a broke dorm student who plots per-well spaghetti before trusting averages—one outlier has ruined my life before. Also, what’s confluency at day 6? If treated wells are near 100%, cntact inhibition gets real, and PCNA up + collagen I flat could mean proliferation without much matrix deposition. Maybe try quantifying gap width at fixed coordinates, not just whole-well area, so you can tell migration from monolayer expansion.

👍 8 ❤️ 3

Lab-adjacent here. Did you include ascorbate in the media? Without it, collagen I readout can stay flat even if PCNA rises—tenocytes need it for matrix. Also, were you re-imaging the exact same coordinates? If the scratch was uneven, closure can look plateaued once the remaining gap is wider than the migration front. That would fit contact inhibition less than geometry/media depletion.

👍 3 ❤️ 2

Night-shift lab rat here. If PCNA is up at day 10 but collagen I flat and closure stalls, I'd stop calling it contact inhibition. Check for multilayer piling. Tenocytes can stack once the gap shuts. Also, did you measure gap area or a single width? One axis hides retraction and uneven edges. Stain Ki-67 plus phalloidin: if proliferation is high and actin fronts are static, it's a migration/assay-limit issue, not matrix. I'd also confirm day-0 seeding density per well, since primary tenocytes at P3 can drift fast.

👍 8 ❤️ 2

I don’t run tenocytes, but I’ve had scratch assays look like everyone hits a wall around the same time. The thing that fooled me was comparing raw gap instead of normalizing each well to its own day-0 width. Did you do that? Also, was the day-10 blot from the same wells you imaged, or a separate plate? If separate, I’d lean assay/plate drift before contact inhibition. If same wells, I’d ask whether the outside of the scratch was already confluent at day 6.

👍 5 ❤️ 1

Data-nerd check: did you scribble a grid on the plate lid and re-image the exact same coordinates each day? My scratch assays looked like a real plateau until I compared identical ROIs; most of the flatness was field-to-field variation and edge meniscus. Also, PCNA up without collagen I could mean proliferation is happening back from the wound edge, not at it. What does Ki-67/EdU look like in the first two cell rows at day 7–10? That’d separate contact inhibition from a migration-specific stall.

👍 2

Did you normalize closure to the initial gap area, or just eyeball raw images? If both arms

👍 1

I track my scratch runs in a dumb spreadsheet: initial gap width, daily gap in µm, and media exchange times. The only ones that plateaued like that were runs where the day-6 gap was already <300–400 µm — after that, the leading edges are basically touching, and my ImageJ threshold starts grabbing debris. What was your day-6 gap in µm, and were you re-imaging the exact same marks each day? If not, contact inhibition gets hard to separate from an edge-detection ceiling.

👍 3

Wait, if PCNA was up at day 10 but collagen I flat, I’d want to know whether the extra cells were actually at the wound edge. Did you pretreat with mitomycin-C (or low serum) to separate migration from proliferation? Without that, a day-14 plateau could just be contact inhibition in both groups while treated wells burned through their proliferative burst earlier. And did you normalize collagen I to cell number or just per well? Per well can hide a real shift if density changed. That’s the first thing I’d poke at.

👍 3

was day 6 when wells hit confluence? If yes, contact inhibition is your plateau. PCNA up but collagen I flat = proliferation without matrix. Did you serum-starve before scratch? High serum can erase migration differences once edges get close. I’d check collagen I earlier, day 3–5, not day 10. Also normalize closure to actual day 0 width per well. Night shift brain here, but that’s where I’d look. What was confluency at day 6?

👍 4

Trend-line brain here: did you drop serum after the scratch? Primary tenocytes will happily close via proliferation if

👍 2

Hi, total newbie here (week one), so please ignore if this is obvious. Did you change media or serum lot during the 14 days? I ask because in my own little tracking journal, my scratch photos looked flatter after a media switch. Also, were you imaging the exact same spot each day? If the gap gets narrow, maybe both groups just physically can't close much more. Sorry if silly. Rooting for your n=4!

i’m very new here, so this might be obvious—when you say closure plateaued, are you normalizing to the original scratch area? If the remaining gap gets tiny, the curve can look flat even if cells are still moving, especially if they

👍 3

not a tenocyte person, just a chronic tracker of my own kitchen-wound photos, so take this with salt. The thing that always fools me is imaging threshold: once the well gets a hazy monolayer, software scores “closed” way before I’d call it closed. Did you lock the same day-0 ROI and re-measure every well, or let the scratch boundary redraw each time? Also, PCNA up with collagen I flat makes me wonder if you’re looking at proliferation, not migration. A Ki-67/EdU pulse or mitomycin-C pre-treatment would separate those. Did all four treated wells show the same PCNA bump?

gentle q: did you check tenocyte identity at P3? Primary tenocytes can dedifferentiate and lose collagen I expression after a couple passages, so PCNA up with flat collagen may reflect phenotype drift rather than BPC. Did you run TNMD/SCX or just morphology? Also, were all wells seeded from the same passage/density and same pre-scratch confluence? If one flask was more crowded, that could decouple proliferation from matrix. I’ve been burned by P3 drift in tendon cells.

One thing that bit me in a similar 14-day wound assay (epithelial, not tenocytes): my day-0 images weren’t matched for initial scratch width. By day 6 the faster wells had less room left, so the curve flattened and looked biologic. Did you normalize closure to each well’s own day-0 area, and maybe re-scratch the same wells at day 7? If the treated cells still close faster after re-scratch, your day-14 plateau is probably assay ceiling/saturation, not contact inhibition or batch drift.

Not a tenocyte person, but I’ve run scratch assays. Did you include a mitomycin-C or low-serum arm to separate migration from proliferation? If closure plateaus around day 6 and PCNA is up at day 10, you may just be measuring proliferation in the non-scratched monolayer, not matrix. Also, was the Western done on whole-well lysate or the scratch front? Whole-well dilutes the front and can hide collagen I changes. At P3 primary cells, donor/passage drift can flatten both arms faster than you’d expect. I’d add EdU/Ki-67 and image confluence, not just scratch width.

👍 2

One assay detail I’d probe: are you normalizing closure to each well’s own day-0 scratch width, or comparing raw area? If initial widths differ slightly, treated wells can look faster early and then converge once the slower wells catch up. Also, did you mark the plate bottom and re-image the exact same field each day? A small stage shift can mimic a plateau. I keep weekly averages in my own tracking logs for that reason—baseline-normalized trends usually tell the steadier story than absolute numbers.

👍 2

Cafe owner here, so take this with a grain of salt, but I’m obsessive about batch tracking. Did you image the exact same marked coordinates each day, or just re-find the scratch? By day 6, if treated wells already have a few bridges, area-based closure can look flat while cells are still active. I’d re-analyze width at 3 fixed points per well, and check whether the day 10 collagen lysates came from the same wells as the PCNA blot. That kind of sampling mismatch has burned me in coffee roast curves—same batch, different point, totally different story.

Did you normalize collagen I to DNA or cell count? With PCNA up, total lysate collagen can look flat while per-cell matrix

We hashed this out in an old thread about fibroblast scratch assays: day 14 is long enough that matrix deposition and cell spreading start dominating the readout, not just migration. If PCNA is up but collagen I is flat, I’d wonder whether you’re seeing confluence/contact slowing rather than a real treatment effect. Did you run a mitomycin-C or low-serum arm to tease proliferation from migration? Also, were the plates pre-coated from the same lot? A coating batch change flattened my own curves once, and it looked exactly like a plateau rather than a failed assay.

👍 1

Hi, I’m brand new here (week one, so please excuse if this is obvious). Did you use the exact same pipette tip width and angle to make each scratch? In my own tracking I learned tiny method changes can fake a plateau. Also, were days 6–14 imaged at the same stage coordinates and focus? A slight shift can change thresholding and make closure look flat even if cells are still moving. I’m probably missing something, but that’s what I’d check first. Sending good luck!

I have seen similar plateaus in scratch assays when wells reach near-confluence. Did you record percent confluence alongside gap width? If the treated group hit roughly 90–95% by day 6, contact inhibition would flatten both arms and obscure later differences. Also, PCNA up with collagen I unchanged makes me wonder if you are measuring proliferation more than matrix deposition. A migration-only control or EdU/PCNA co-stain might separate those. I keep weekly averages for my own tracking, and this kind of ceiling effect is easy to miss in daily snapshots.

Tbh hotel-wifi thought: if both arms flatten around day 6, are the wells basically confluent? A narrow scratch can hit the assay ceiling before any migration/proliferation difference shows. Did you image fixed coordinates and measure actual wound area, not just “days to closure”? Also, how do the tenocytes look at day 10—still spindle-shaped or enlarged/flattened? Passage 3 drift could explain PCNA up without collagen I. I’d kill for a scratch that closes on schedule; mine always quits when the hotel breakfast closes.

👍 1