I'm curious how much damage is potentially inflicted by shear stress by pipetting. I know that syringes used for stem cell injection can cause a lot of damage. However, to what extent does this happen with P20 and P200 pipette tips? Understandably the shear modulus of bacterial cells is significantly different from that of cancer cells, which will be different from that of stem cells.

  • $\begingroup$ It might be possible to compare P200 with P20 on the cheap (anybody up for a crowd-sourced experiment on HeLa or E. coli?), but I suspect you actual question asks for absolute numbers, and is therefore almost impossible to answer. It is hard to envisage any non-pipette method to deliver 100% viable cells, and even harder to think of alternatives (measure free DNA after pipeting? how can that be converted to cell counts?). User137 suggests something decent, but still short of absolute. In the absence if evidence, all the answers you will receive are bound to be speculative "expert opinion". $\endgroup$
    – nvja
    Sep 1, 2014 at 1:36

7 Answers 7


This is an excellent question, I have been training people to culture cells for about 12 years and students have a hard time grasping this and appreciating the importance etc.

What cells are usually experiencing during pipetting is analogous to a crowd of people trying to fit through the doorway of a building. Shearing during pipetting is certainly a legitimate concern in cell culture. You will notice its negative effect on viability most explicitly when pipetting cells in freezing medium (containing DMSO) following a thaw. Until their DMSO concentration drops their membranes are weaker & more fluid. That's why you pipette the frozen cells drop-wise to the fresh medium, to be especially gentle at this point.

Prokaryotic cells such as the above mentioned TOP10 cells are treated with Calcium chloride and glycerol which has essentially the same effect on their walls and membranes. Hence pipetting should be delicately performed after thawing these cells as well.

With that being said, if you compare the sensitivity of prokaryotic and eukaryotic cells to shearing, eukaryotic cells are profoundly more sensitive.

For individuals using competent bacteria for sub-cloning and other routine uses, killing a small percentage of of your cells is not that important. However if you are using the transformation to generate cDNA libraries it's very important that you have a titer sufficiently representing the transcriptome the library is composed of.

In these instances using a more premium competent cell, adhering to correct temperatures and minimizing pipetting is essential. This is why many are taught to swirl the DNA the with the competent cells, rather than the harsher alternative: pipetting up-and-down.

The four most important factors that contribute to cellular shearing are, and in order form most to least contributory:

  1. Diameter of the bore in the pipette tip, the smaller the more shearing.
  2. Speed at which the suspension is passing through the opening of the tip. The faster the more damaging.
  3. Size and rigidity of the cell. Larger cells are more prone to damage. Cells with a murein wall are less prone to damage.
  4. Concentration of cells. Cultures that are more concentrated are more prone to damage.

Because the make up if the cell itself is so influential on the amount of damage, and because of the enormous variety of cells; designing a representative experiment to assess damage would be very difficult and laborious.

Finally, the variety of tips and seriologicals is also enormous. However if one were to attempt to assess this it could be done:

Choose a representative variety of cell lines, choose a representative variety of pipettes. One would probably want to use a robotic pipette to be able to evaluate incrementally assigned speeds and pressures. Look at different culture concentrations, phases of the growth curve, time between pipetting and analysis, distance between the exiting fluid and the culture etc, etc.

I think a very successful analysis method would be using propidium iodide and FACS.

After your experiment which will cost a lot of time and $, I think you will find common sense rules: keep pipetting to a minimum and use wider tips whenever possible.

  • 1
    $\begingroup$ This is a good explanation but does not actually give a numerical representation of percent death to no harm caused by the palleting. $\endgroup$
    – user1357
    Aug 31, 2014 at 2:56
  • $\begingroup$ Good point Here I will rephrase as it seems to be getting quite a number votes $\endgroup$
    – rhill45
    Sep 2, 2014 at 1:04
  • $\begingroup$ Won't the material with which the pipette is made affect the cell too? I would imagine cell membranes having different affinity to different materials $\endgroup$
    – One Face
    Jan 25, 2015 at 13:57
  • 1
    $\begingroup$ @CRags certainly yes but really minimal. $\endgroup$
    – rhill45
    Jan 27, 2015 at 11:32

It's an easy experiment to do. Take your cells aliquot them into 10 microfuge tubes, and pipette each suspension increasing amount of times, stain with trypan blue and count.

The most important factors will be which pipette-type you use; I would expect a p1000 to cause more damage then a p200 then a p20 due to velocity of the fluid. Also the most important factor will be the skill of the scientist, if you pipette slowly it should decrease stress as opposed to pipetting quickly.

In my experience it depends on the pipette-type and the skill of the operator. The only way to answer this for you is to try it empirically.

  • $\begingroup$ Well, then that will heavily be based on operator error. Anecdotally, I've heard that it is the other way; a P1000 would causes less damage than a P20. $\endgroup$
    – bobthejoe
    Jun 13, 2012 at 6:42
  • $\begingroup$ Artem is right: the only way to say is empirically. The P1000 will influence the cell damage likely because of the velocity. In contrast, the P200 and P20 could give some damage due to the narrow tip hole. $\endgroup$ Jun 13, 2012 at 18:36
  • 1
    $\begingroup$ Give it a try and post the results. I do believe that operator skill is the single most important factor. $\endgroup$
    – Artem
    Jun 18, 2012 at 21:50
  • $\begingroup$ We used to use a peristaltic pump to plate HepG2 cells into 384 and 1536 plates. When we moved to primary hepatocytes the pump killed too many and had to move to hand plating with an 8 channel pipette. So cell type matters a lot. $\endgroup$
    – user137
    Jul 30, 2014 at 23:17
  • $\begingroup$ Naw this isn't going to measure the correct....just naw $\endgroup$
    – user1357
    Aug 31, 2014 at 2:54

Anecdotally I have not observed any cell death upon pipetting of E. coli DH5alpha or TOP10, however as competent cells, mixing by pipetting up and down is discouraged due to the compromised cell wall.

  • $\begingroup$ Good pointIt's pretty hard to kill bacteria mechanically period, the competent cells are more sensitive though and concern comes when building cDNA library's mostly as you need a very representative number of colonies $\endgroup$
    – rhill45
    Aug 30, 2014 at 16:57
  • $\begingroup$ This is a fine observation. $\endgroup$
    – user1357
    Aug 31, 2014 at 2:56

This question would be better served in the physics se or chemistry. I don't think we have a generic engineering where fluid dynamics could be described in a different method than physics but if we did it should also be answered there.

Indeed the osmotic or liquid pressure in and of itself would cause changes possibly damages to the cell. Then other eexothermic and endothermic reactions to the chemicals due to force of impact like a chem or friction burn.

I'm sure the other requirements of conservation of energy which I half-hazardly ignored.

I found this interesting paper which focuses on the shear stress in general.

Pipetting causes shear stress and elevation of phosphorylated stress-activated protein kinase/jun kinase in preimplantation embryos.

Xie Y1, Wang F, Puscheck EE, Rappolee DA.

Shear stress at 1.2 dynes/cm(2) induces stress-activated protein kinase/jun kinase phosphorylation that precedes and causes apoptosis in embryos (Xie et al., 2006b, Biol Reprod). Pipetting embryos is necessary for many protocols, from in vitro fertilization to collecting embryos prior to analyzing gene expression by microarrays. We sought to determine if pipetting upregulates phosphorylated MAPK8/9 (formerly known as stress-activated protein kinase/jun kinase/SAPK/JNK1, 2). We found that phosphorylated MAPK8/9, a marker of MAPK8/9 activation, is upregulated in a dose-dependent manner by pipetting. Whereas embryos with the zona pellucida removed were more sensitive to stress-induced lethality mediated by 1.2 dynes/cm(2) shear force, phosphorylated MAPK8/9 was induced at lower numbers of pipet triturations in hatched embryos at E4.5. E4.5 embryos were more sensitive to induction of MAPK8/9 than unhatched embryos at E2.5 or E3.5. E3.5 embryos also showed a pipetting dose-dependent induction of FOS protein (formerly known as c-fos), a marker of shear stress in many cell types. Phosphorylated MAPK8/9 measured in ex vivo embryos from E1.5 to E4.5 were expressed at low levels. Embryos that had been pipetted sufficiently to induce phosphorylated MAPK8/9 and FOS had the same number of cells as untreated embryos 24 hr later. This suggests that rapid phosphorylation of MAPK8/9 due to transient shear stress does not mediate long-term negative biological outcomes. But, it is possible that techniques requiring multiple handling events would induce MAPK8/9 and cause biological outcomes or that other biological outcomes are affected by low amounts of transient shear stress. This study suggests that embryo handling prior to experimental measurement of signal transduction phosphoproteins, proteins and mRNA should be performed with care. Indeed, it is likely that shear stress may cause rapid transient changes in hundreds of proteins and mRNA.

  • 1
    $\begingroup$ @rhill45 yeah the question was about cell shear stress damage during pipetting..the other answered seemed to buy into the only damage coming from the pal tip..but i wanted to point out that the shear stress from the pipetted liquid is equally likely to damage/destroy the cell. $\endgroup$
    – user1357
    Sep 1, 2014 at 23:21
  • 1
    $\begingroup$ @rhill45 also if your going to learn php you should invest time in asp because many industries see saw between the two. $\endgroup$
    – user1357
    Sep 1, 2014 at 23:43

Shear stress $\tau$ in this small sizes is usually measured in dyne/cm2 or N/m2 = Pa. The equations betweeen them: $1dyn/cm^2 = 10^{-5}N/cm^2 = 0.1N/m^2 = 0.1Pa$.

What kind of damages zygotes can suffer by pipetting?

Using scanning electron microscopy, we found open holes on the surface of lysed eggs, indicating failure of the plasma membrane to reseal after microinjection. No holes were seen in unlysed eggs, but many of them had membrane alterations suggestive of healed punctures.

Even a small 1.2 dyn/cm2 shear stress induces apoptosis by pipetting zygotes. So zygotes have their critical shear stress level by 1.2 dyn/cm2 and pipetting involves greater forces than 1.2 dyn/cm2.

Shear stress at 1.2 dynes/cm2 induces stress-activated protein kinase/jun kinase phosphorylation that precedes and causes apoptosis in embryos (Xie et al., 2006b, Biol Reprod). Pipetting embryos is necessary for many protocols, from in vitro fertilization to collecting embryos prior to analyzing gene expression by microarrays. We sought to determine if pipetting upregulates phosphorylated MAPK8/9 (formerly known as stress-activated protein kinase/jun kinase/SAPK/JNK1, 2). We found that phosphorylated MAPK8/9, a marker of MAPK8/9 activation, is upregulated in a dose-dependent manner by pipetting.

The critical shear stress level is somewhere between 0.01 and 1000 dyn/cm2 by animal cells depending on the cell type and species. (I think the average is somewhere about 50 dyn/cm2, but it is very hard to differentiate between articles mentioning critical shear levels and most lethal shear levels, so the range and the average might be lower.) The death constant (1/h) increases exponentially by increasing the shear stress.

An apparatus for the detailed investigation of the influence of shear stress on adherent BHK cells was developed. Shear forces between 0.0 and 2.5 N m−2 were studied. The influence on cell viability, cell morphology, cell lysis, and cell size was determined. Increasing shear forces as well as increasing exposure duration caused increasing changes in cell morphology and cell death. A “critical shear stress level” was determined.

Shear stress related damage to a mouse hybridoma was examined by Abu-Reesh and Kargi under laminar and turbulent conditions in a coaxial cylinder Searle viscometer. Cells were exposed to 5 to 100 N/m2 shear stress levels for 0.5 to 3.0 h. At a given shear stress and exposure time, turbulent shear was much more damaging than laminar shear as also reported in the past for protozoa and plant cells. Under turbulent conditions, damage occurred when shear stress exceeded 5 N/m2. Respiratory activity of the cells was damaged earlier than the cell membrane, thus implying transmission of the stress signal to the interior of the cell. Cell damage followed first-order kinetics both in laminar and turbulent environments. For turbulent shear stress levels of 5 to 30 N/m2, the death rate constant (kd) increased exponentially with increasing stress level; the kd values varied over 0.1 to 1.0 1/h.

Subconfluent endothelial cultures continuously exposed to 1–5 dynes/cm2 shear proliferate at a rate comparable to that of static cultures and reach the same saturation density (≃ 1.0–1.5 × 105 cells/cm2 ). When exposed to a laminar shear stress of 5–10 dynes/cm2 , confluent monolayers undergo a time-dependent change in cell shape from polygonal to ellipsoidal and become uniformly oriented with flow. Regeneration of linear “wounds” in confluent monolayer appears to be influenced by the direction of the applied force. Preliminary studies indicate that certain endothelial cell functions, including fluid endocytosis, cytoskeletal assembly and nonthrombogenic surface properties, also are sensitive to shear stress. These observations suggest that fluid mechanical forces can directly influence endothelial cell structure and function.

Shear stress above 0.25 dyne/cm(2) resulted in dramatic loss of podocytes but not of proximal tubular epithelial cells (LLC-PK(1) cells) after 20 h.

A series of careful studies has been made on blood damage in a rotational viscometer. Specific attention has been focused on the effects of solid surface interaction, centrifugal force, air interface interaction, mixing of sheared and unsheared layers, cell-cell interaction, and viscous heating. The results show that there is a threshold shear stress, 1500 dynes/cm2, above which extensive cell damage is directly due to shear stress, and the various secondary effects listed above are negligible.

The shear stress threshold of some dinoflagellates (microalgae) is even lower than that of erythrocytes (0.029 N/m2). For example, a continuous laminar shear stress level of only 0.0044 N/m2 (equivalent to a shear rate of 2.2 1/s) has proved lethal to the dinoflagellate Gonyaulax polyedra.

Other cell types are not necessary as sensitive as animal cells and they don't necessary react with apoptosis (about 10 dyn/cm2) to shear stress, so you have to use necrotic (about 5000 dyn/cm2) forces to destroy them :

cell type                       size                shear sensitivity
microbial cells                 1-10μm              low
microbial pellets/clumps        up to 1cm           moderate
plant cells                     100μm               moderate/high
plant cell aggregates           up to 1-2cm         high
animal cells                    20μm                high
animal cells on microcarriers   80-200μm            very high
fungi cells                     2-10μm              moderate/high

Results show that Chinese Hamster Ovaries and Human Embryonic Kidney cells will enter the apoptotic pathway when subjected to low levels of hydrodynamic stress (around 2.0 Pa) in oscillating, extensional flow. In contrast, necrotic death prevails when the cells are exposed to hydrodynamic stresses around 1.0 Pa in simple shear flow or around 500 Pa in extensional flow.

The shear sensitivity is not determined only by cell type and species, there are many other factors involved:

  • type of cell and species
  • composition and thickness of cell wall when present
  • size and morphology of cell
  • the intensity and nature of shear stress, whether turbulent or laminar, or associated with interfaces (e.g. during bubble rise and rupture)
  • growth history, both short-term (e.g. starvation) and long term adaptation
  • growth medium (trace elements, vitamins, carbon and nitrogen sources)
  • growth rate
  • growth stage
  • type and concentration of shear protective agents if present

Cells can be very sensitive to shear stress caused by turbulent flow, while not so sensitive to shear stress caused by laminar flow.

On the basis of laminar flow viscometriy measurements, a critical shear stress level of 80-200 N/m2 has been suggested for Morindata citrifolia cells.

... while for Daucus carota a shear stress level of 50 N/m2 has been associated with cell damage. In other study, carrot cells in a laminar flow Couette viscosimeter lost the ability to grow and divide in the shear stress range of 0.5-100 N/m2. The intracellular enzyme activity was impaired at shear stress levels above 3000N/m2, but significant lysis did not occur until a shear stress level of 10.000 N/m2 applied over a prolonged perioud (>1h).

In contrast to the behavior in laminar flow, the cells were quite sensitive to turbulent impeller agitation. Impeller tip speeds of ~1.1 m/s lysed a significant proportion of the cells within 40min.

The bubble damage is severe (1000 cells by a single 3.5mm size bubble) because of the cell adherence to the interface of the bubble and the strong forces involved (>1000 dyn/cm2 by stirred bioreactors). The adhesion and so the damage can be reduced with surfactants.

It is proposed that when cells are either attached to, or very near, a rupturing bubble, the hydrodynamic forces associated with the rupture are sufficient to kill the cells.

All experiments were conducted with Spodoptera frugiperda (SF-9) insect cells, in TNM-FH and SFML medium, with and without Pluronic F-68. Experiments indicate that approximately 1050 cells are killed per single, 3.5-mm bubble rupture in TNM-FH medium and approximately the same number of dead cells are present in the upward jet. It was also observed that the concentration of cells in this upward jet is higher than the cell suspension in TNM-FH medium without Pluronic F-68 by a factor of two. It is believed that this higher concentration is the result of cells adhering to the bubble interface. These cells are swept up into the upward jet during the bubble rupture process. Finally, it is suggested that a thin layer around the bubble containing these absorbed cells is the “hypothetical killing volume” presented by other researchers.

For a hybridoma line, reported that exposure to laminar shear stress (208 N/m2) in unaerated flow in a cone and plate viscometer led to substantial loss in cell count and viability within 20 min. At a constant 180 s exposure, increasing shear stress over 100-350 N/m2 linearly enhanced cell disruption, with >90% of the cells being destroyed at 350 N/m2 stress level. Shear stres levels associated with bubble rupture at the surface of a bioreactor may range over 100-300 N/m2. These values are remarkably consistent with shear rates that damaged hybridomas in unaerated laminar flow experiments.

Smaller hole pipettes cause more damage.

We also examined aspects of the gene transfer procedure that might influence survival such as the size of injection pipettes and their taper relative to zygote diameter, possible toxicity of the injection medium, the timing of injection, and immediate vs. delayed pipette withdrawal. The only factors that significantly affected cell viability were pipette size and taper, and timing of injection in relation to first cleavage. This suggests that zygote viability correlates inversely with the size of the hole produced by the injection pipette and that damage to the membrane is less successfully repaired as the fertilized egg readies itself for division.

It is hard to find anything about the level of shear stress by pipetting. It can be certainly more than 1 dyn/cm2. It has a short duration (at most a few seconds). I think the following factors can influence the shear stress levels by pipetting:

  • pipette type
  • flow speed (faster flow can be more likely turbulent)
  • bubble formation

Probably more factors are involved but I am not a pipetting expert. ;-) I agree with the others, it surely depends on the personal skills e.g. an amateur can create huge bubbles by pipetting, which can kill a lot of cells by formation and disruption...

I agree with Artem that this is an experiment to do especially if the result is important for you. What you need to create a model about pipette damage, are the shear stress levels by pipetting and the critical shear stress levels of the cells. I think it is hard to design and experiment in which you can measure the shear stress levels in your pipettes and there is no flow model for pipetting as far as I know, so it can be a good topic for a thesis or a diploma work.


I know this is a two-year-old thread, but I though I would post this anyway in case someone else is searching for this answer like I was tonight. This article cited below doesn't discuss pipette tips vs needles specifically, but it does discuss the difference in effects of tapering vs cylindrical needles on cell damage. Unfortunately, the math in the paper is a bit beyond me (I'm a doctor, not an engineer, and they didn't cover this stuff in medical school), but they found that with cylindrical needles result in approximately 5x the amount of cell death compared to a tapered needle at any given flow rate, and approximately 6-8x the amount of cells damaged. This was likely due in part to the need for higher pressures to increase flow rates in cylindrical needles compared to tapered needles and variations in shear due to the geometries. While a tapered needle is not identical to a plastic pipette tip, the geometries are comparable.

Biotechnol Prog. 2011 Nov-Dec;27(6):1777-84. Effect of needle geometry on flow rate and cell damage in the dispensing-based biofabrication process. Li M1, Tian X, Schreyer DJ, Chen X.



I have a real life biological example. simplified experiment setup: ex vivo purified B cells undergo 3 washes in PBS (spin and resuspend), then they are incubated in vitro for 24 hrs and analyzed by flow cytometry using Acqua Zombie and PI for dead cell exclusion. Experiment 1: pipeting after washes using 1ml tip Experiment 2: pipeting after washes using 10ml pipette

Cell survival as analyzed by flow cytometry: Experiment 1: 9.96% Experiment 2: 43.9%


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.