Yes, the number of cells is a key to get your xenograft. usually 3-10 million cells. So this is the reason, when you study cancer stem cells you need to inject as less as possible cells to demonstrate you isolated cancer stem cells. If not study stem cells. your target is to get xenograft so inject more cells better. However, if some cancer cell lines are not able to grow tumors in mice, you will not get xenograft even if you inject 20 million cells. For difficult cell lines for tumorigenesis, I would suggest to use xenograft tumor for making animal model as soon as you get one mouse growing tumor then you can sub-implant it to make more tumor xenograft for experiment. This is Draft and may have some typo or grammar error. Fengzhi Li

I agree with Fengzhi Li

Generally, you can inject subcutaneously three cell densities/100 µL mouse (3, 5 and 10 million of tumor cells, 5 mice for each cell density) with and without matrigel.

There are some tumor hystotypes which have a latency period of 10-20 days before that tumor starts to grow. You will choose the cellular density which appears to be homogeneous with a 100% of tumor engraftment.

It is possible that the tumor cell line is not tumorigenic if you don't obtain a tumor growth curve, in this case, if it engrafts in a mouse, you can collect the tumor lesion and re-inject the tumor cells at three densities, because you could select a clone tumorigenic.

I agree with Loredana. A titration curve with a different number of cells is the first step. The right number of cells have to chose in bases to parameters important for the study (those that Loredana said and the time to development of tumour mass is important for high throughtput too. About the drug responsiveness, the number of cells can modify the response intensity but in any case you have to performe a control condition. Any response should be normalized vs the control condition

To increase your success rate, beside using a big number of cells, I would also consider using a more immunocompromised mice.

Good

M.

I have seen an example of 1000 melanoma cells producing mouse tumors.

A larger number of cells die just after inoculation. My opinion is that after certain (cell type specific) threshold, cell number is not critical for succesfull experiment but of course all animals within an experiment should have similar inoculation. Also other things than cell number have an effect to tumorigenesis, such as mouse strain: nude vs. scid, inbred vs. outbred nude. If tumors grow too large and become negrotic, they are impossible to treat.

I would recommend using a cell number based on the growth rate of your cell line and the mouse strain you are using. If you are using balb/c normal (non-immunocompromised) than I would recommend using a higher cell number. if you cancer cell line in vitro grows very rapidly where its PD is one day I would recommend a lower range ~5 million or so in balb/c. This is to ensure that you do not get exponentially growing tumors. If you get exponentially growing tumors your drug would not even have a chance to have any effect before you have you euthanize your mice based on your IACUC policy. It might be wise to obtain some mice and just inject them with varying doses of cells and determine which is a suitable growth rate and use that growth rate to do your actual efficacy study. Best of luck!

I agree with all, but I suggest 1*106 cells, which comes 1000000/ mouse

I recommend doing a quick lit search and check out ATCC for the cell line that you are planning to utilize as your cancer model. These numbers are likely available. In my experience, I would utilize a few mice as Loredana described to do a titration curve of cell numbers. As others have suggested, cell lines can produce tumors in athymic nude mice with as little as 1000 cells. I would typically inject 5 x 10^5, 1 x 10^6 and 2 x 10^6 in 100 ul of cell media or PBS per injection site using a 25 gauge needle on a 1 ml syringe for breast cancer cell lines. As a quick note - use one 50 ml conical tube for each concentration. It is really hard to get a needle into a 15 ml tube and 1.5 ml flip top tubes will spill easily. I would inject all three concentrations into on mouse, one dose in each flank and one behind the head. The second issue is when to deliver your tumor treatment. I suggest that as soon as tumors are appoximately 0.5 cm, start your treatment method. Make sure to use the appropriate controls, because sometimes tumors will spontaneously regress depending on the cell line or potential cell line contamination ie mycoplasma. This is a suggestion for subcutaneous injection. Innjections intravenously, intracardially, or orthotopically will have a different ideal cell number to establish tumor growth.

It is also very important, that your cells are not too confluent, when you harvest them (max. 40-50% confluency)

All the answers are correct- but if you provide information what cell lines you are using you could probably get much more concrete answer.

There is a limiting factor when talking about the number of cells to inject. a good starting point is 2.5 million, the Confluence requirement is actually cell line specific, some lines grow with a high confluence and some with ~50%, the best way to approach a new cell line in your lab is to do growth evaluation studies, if it is a commercially available line start why the suppliers recommendations, if it is a human derived cell line all bets are off and you must run growth evaluations. If you are using a specific cell line for the target(s) it typically expresses, you must confirm the target expression both in vitro and ex vivo, every time you conduct a new passage, as the target expression WILL change through different passages, if you are using stromal targets also note that a percentage of the stroma changes from human to a mouse in as few as 2 passages.

Hope this helps

It varies by xenograft model/cell line. You will need to determine this empirically using your model.

We use small fragments of tumor containing approximately 5-10K cells to start tumors from patients, and maintain lines by serial transplant.

Some tumors will not grow as dissociated cells in our hands.

Good luck

Depending on the cell line, you can use some matrigel mixed with your cells. Matrigel is a mixture of extracellular matrix proteins which will support the growth of some cancer human cells in immunosupressed mice. I have used it with the prostate cancer cell line LNCaP at 1 x 10^6 cells per inoculation. It is tricky to use though, because matrigel is liquid at 4C but solidifies at room temp. So the mixture matrigel + cells has to stay at 4C until inoculation, meaning you have to be quick between the moment you load the syringe and the moment you inoculate.

Best,

Tumor growth depends on how aggressive the cells are! Also depends what kind of study you are going to do.

However, I found minimum 5000 of cells (HCT116-colon cancer cells) form subcutaneous tumor and took about 30 days to grow . 2-8 million cells form tumor very fast. If you want to test the drug effects, it is good to use moderate # of cells (after testing with different numbers of cells to establish the growth rate of the particular cells you want to use) to grow the tumor slow and steady. If the growth rate is very fast your drug may not work.

Thanks

Growth rates and cell numbers required highly depend on the mouse strain, the cell line and the conditions. I use limiting dilution transplantation to determine the number of cells required to forma tumor. You will also see the difference between high and low numbers in such a setting right away. I would recommend. 10 mio to 1000 cells in 10fold steps to start. That will give you 3x5 = 15 mice transplanted total and it should give you an answer to your question.

I am using balb/c mice to reject DFTD tumors while I use NOD/SCID to engraft DFTD tumours. I have recently did an experiment comparing 10^6 , 10^5, 10^4 cells. 10^ 6 cells produces the quickest response but 10^4 cells is taking about 2 or 3 times longer.

Some people say the size of the syringe needle affect the results as well. If size the needle is too small, cell may die when passing through the needle. Is that true?

Sure the number of the injected cells can make a difference. You can increase the number of the cells. This depends on the site of injection and the type of cells you are using

In certain models you can mix your cells with Matrigel, and this helps also

Thank you every one for your valuable suggestions and guidelines.. I will plan for a titration curve for optimizing the cell dosage, with proper reference.... Once again thank you all...

Is there any special care to take while we inject the cell suspension subcutaneously into the mice. I wanted to inject it in the flanks of mice.

Also, if you inject too many cells, the tumor may grow quickly, but without sufficient vascularization. It may necrose from the middle, or, if you're doing drug studies, the drug will not distribute to the tumor as much as the rest of the body, resulting in a false negative.

Considering following two points you can optimize the cell concentration.

a. Depend on type of cancerous cell line to be used for induction.

b. We used cell number for xenograft model is 2.5 to 5X10 to the power 6/animal

For leukemias (primary leukemia cells from patients with ALL), 1-5 million intravenously for unirradiated NOD/SCID mice or CB17 mice that are irradiated with 2 Gy gamma rays.

Yes you will get a tumor with a large number of cells. The only difference that occurs with higher cell number is a more rapid growth rate of tumor and faster appearance of tumor masses. It is best of inject several mice with various doses of cells (usually 1e6 to 5e6 is a good number for most cancer cell lines) and determine a favorable growth rate. Use 5e6 for a slower growing cell line (change media required after 2-3days, passage every 2-3 days) and 1e6 for rapidly growing one (change media/passage required every day like A375 melanoma).

Hello Pramod,

for better response Can you explain about cell lines and mouse strain you are using? .

Cell numbers for tumor induction in nude or SCID mice is not same for all cell lines. It totally depends on the following factors:

1) cell morphology and doubling time,

2) % Viability of cells during inoculation,Tumor take up and growth rate,

3) followed protocol,

4) cell and matrigel ratio(if you are using matrigel),

5) SCID/NUDE Mice Strain(Balb/c,CD1,Nu/Nu,Hairless SCID...etc),

6) cancer cell line origin site,

7) Angiogenicity og Cancerlous cells,

8) Site of inoculation,

Apart from these factors it also depends on Threshold number of cell lines.

As per my experience what i have tried in our lab, range is 1*10 ^6 to 10 *10^6

cells for different cell lines.Above 10 million cells will be very thick so you can face difficulties while inocultion.So maximum times we are using 5 million cells.

For optimization of numbers & validation first you do some review or start with 5 * 10 ^ 6 with 2 more diffrent cell numbers.

I hope these informations will be helpful for you.

Best luck,,

milind

It does depend on cell line correct?

In my opinion for any cell type first we should perform a pretest with varying cell numbers, say 0.5 x10^6, 1 X10^6, 5 X10^6 cells & sometimes upto 10 million (i hope this should be enough what ever the cell line it may).

1. While doing pretest, we should consider conditions like injecting cells with matrigel and without matrigel. Based on the outcome, one should decide whether matrigel is actually required or not for the tumor development of that particular cell type being used.

2. In case, if both conditions (with and without matrigel does not work, consideration should be on increasing the matrigel concentration) or there is huge amount of tumor development while using matrigel, then one should consider the difference between tumor volume in positive and negative control. Sometimes it may not show any difference. This happens when the matrigel with high growth factor is used. For example with respect to HEK293 cells it happens. So then the best option could be using matrigel with reduced growth factor where the positive and negative control will show actual tumor development potential. Note: Matrigel has types based on concentration of growth factors in it. 

3. with respect to nature of  cell type being used, one should carefully review literatures whether to use matrigel or not. Some cell lines may not need matrigel. For example NIH3T3. If matrigel is mixed and injected, Irrespective of your  hypothesis it becomes tumorigenic. 

Finally one the best way to optimise the cell number required for xenograft injection is reviewing the literature which already used the same cell line in same animal models. This will help a lot to save time from doing pre-test to optimize cell number. if no such informations, it is best to perform pre-test to optimize the cell number. Tumor growth is usually observed until the tumor develops necrotic cell death in the surfaces which denotes its threshold,  which it will not grow any more. This too depends upon cell type being used.

Thank you,

If you are doing drug studies after the tumor grows, I like to inject a lower number of cells, so that the tumor has time for angiogenesis.  This allows for better pharmacokinetics.  Also, without angiogenesis, the tumor will necrose in the center, making it hard to judge if the cell death is due to your treatment or the center running out of nutrients.

Hello everyone (and especially Prof. Paul) I am really interested to have a good vascularized tumor to treat with intravenous injection of nanoparticles bearing a molecular targeting.

How much you normally reduce to have better vascularized tumors? The half?

Normally we inject 2 million PC3 cells into NMRI-nu mice and we have to wait more then 3 weeks the tumors to develop while sometimes it's not even developing.

Now I am thinking to inject 1 million PC3 cells/flank in NMRI-nu mice but I am scared that reducing the number leads to no tumor formation.

Have you some advice to avoid that?

Thanks a lot 

The number of cells needs to be inoculated in the mice varies between the cell lines and tumor types as well. Prostate cancer cells like VCaP and LNCaP develop palpable size tumor at 10 million cells in 3 weeks in immunodeficient mice whereas ovarian and endometrial cell lines require 1 million cells for successful engraftment.

Does any one experienced BALL1 leukemia (blood cancer) cell line xenograft in mice? How much cells per injection?

Start with 1 million cells

Different xenografts do require different amounts of cells for the starting amount (although 1 million is a pretty stable amount). You can see the differences based on tissue type and other characteristics here: http://altogenlabs.com/xenograft-models/. The route of administration can also affect the necessary number of cells. In general, to prevent varying growth from affecting the progression of a study, drugs/therapeutics are administered once tumors have reached a certain volume (e.g. 100 mm^3).

Can anyone provide information on a publication that discusses the effect of starting cell inoculation number and rate of initial tumor growth on vascularization, or any aspect of tumor development?