FAQ

Q: We injected anti-DBH-SAP (Cat. #IT-03) into the hypothalamus of Sprague-Dawley rats and sacrificed them 2 weeks later. We did not see any reduction in the DBH fiber staining.

When the drug arrived, we aliquoted it in 1-µl snap-cap tubes on ice, and stored them at -80°C. For injections, a 1-µl aliquot was diluted to a little over 10 µl so that we had a final concentration of 1 µg/10 µl. We administered two injections of 100 nl on each side (with 10 ng of anti-DBH-SAP) using a 0.5-µl Hamilton syringe attached to a stereotax. The needle was a 33-gauge with a blunt tip. I tried previously to use glass micropipette tips attached to a Hamilton syringe with the line filled with mineral oil, but found that the actual volume displacement was too unreliable.

A: We have not had any problems related to the stability of anti-DBH-SAP. In our work, failure to lesion is nearly always associated with a misplaced injection. From the information conveyed, I would suggest the following:

(1) It is possible that no drug was actually delivered to the brain. Two things could be done to ensure drug delivery. The first would be to add a tracer to the saporin solution that could be identified histologically. The second would be to visually monitor drug delivery using a calibrated tip. Air bubbles, pressure leaks and compression of the liquid can interfere with accurate delivery.

(2) It is possible that the anti-DBH-SAP was not delivered to the correct site, so that the expected uptake into the targeted terminals did not occur. Again, marking the site so it is clear where the injection was would help evaluate your accuracy. Establishing a reliable set of stereotaxic coordinates that work in your lab, in your rats and with your equipment and then using a dye to estimate the diffusion radius of your selected injection volume are always good ways to start. However, that being said, it should not be difficult to locate the injection site with such a large injector (33 g) – so #1 seems more likely to be the problem in the case you describe. Also, I would add that the larger the injector, the more nonspecific damage there will be. Glass capillary micropipettes are by far preferable to stainless steel cannulas in providing more reliable delivery of small volumes and in producing less nonspecific damage. Chronically implanted cannulas should be avoided, in my opinion, because gliosis at the cannula tip is apt to occur and this may alter the diffusion pattern of the injected substance, as well as interfering with lesion analysis.

(3) Try a different anesthetic. We have not tested a lot of anesthetics, but we have had problems getting a good lesion that we think are attributable to use of a ketamine/xylazine/acepromazine anesthetic cocktail. So we routinely avoid that one.

(4) I assume you are looking at fibers in the area of the injection. If not, it would be important to make sure the fibers being evaluated are associated with the same neurons innervating the terminal field at the injection site. Secondly, the 2-week wait mentioned between toxin injection and histology is critical for evaluating the lesion to assure that immunoreactive products are no longer present. Making sure that tissue processing controls are stringently adhered to so that controls and lesioned animals are run together in the same batch is also important.

(5) You might try injecting only one side and comparing terminal staining with the non-injected side in the same animal. This would not be a good idea, however, if the injection site is too close to the midline, so that both sides might be damaged from a unilateral injection.

Related: Anti-DBH-SAP (Cat. #IT-03)

Q: I have a question about Retrograde Transport. There is a comment in the FAQ “In fact, we recommend that you wait two weeks at least to see immunohistological evidence of a toxic effect after injection of a saporin toxin in vivo.” Are there data to support this recommendation? As a researcher who utilizes your toxin products I often get asked about the time course of toxin action. It’s difficult to answer because the literature is currently limited with regard to in vivo toxin application. Any citations, advice, or comments would be greatly appreciated.

A: Actually there’s quite a bit of in vivo use. Or is it just that I think the glass is half full? If you search PubMed for the use of the immunotoxin 192-IgG-SAP using the terms ‘192’ and ‘saporin,’ you’ll get 223 hits, and all of these describe in vivo use.

As far as the two-week idea, you’re right, it’s a bit more challenging to pin that down in the literature. In our book, Suicide Transport and Immunolesioning [1], Ron Wiley discusses at several points the microglial infiltration that occurs and subsides by 14 days. That is what cleans out the antigens that you probably would use to demonstrate cell death – that is, they aren’t there any more. You might want to see if it’s in your library; it’s a good basic source of info. You can also view more References on our website.

As far as the process, Waite et al. [2] show the appearance of behavioral effects associated with neuronal loss at day four and plateauing at day 7. This coincides with the time course seen in vitro [3]. At this point, microglia will infiltrate; this is nicely described in Seeger et al. [4]. However, they stop at 7 days, which is probably the peak day for infiltration. Once there is complete removal of the detritus, microglia down-regulate and at 14 days, you don’t see them, or the antigens that belonged to the cells that were eliminated. So that’s the idea behind waiting.

Q: Do you have a product which can be used to produce retrograde lesions WITHOUT killing cells at the site of injection? What I’d like to do is to kill neurons that project to an efferent nucleus without damaging neurons in the efferent nucleus itself.

A: Making a selective retrograde neural lesion based only on the criterion that the cells to be lesioned are afferent to a particular nucleus or population of neurons, is a formidable challenge at present. Conceptually, this would seem to require a targeted toxin that was taken up only by afferent terminals and not by dendrites, cell bodies and/or axonal membranes of the neurons that are to be de-afferented.

There are some instances in which you can avoid local killing, but only in the case in which there are no receptors in that area, except from projections. So for instance, cholinergic neurons will project to the cortex. You can inject 192-IgG-SAP (Cat. #IT-01) there; it will be taken up and eliminate basal forebrain neurons with little harm to other cortical neurons. Or you can inject anti-DBH-SAP (Cat. #IT-03) into the spinal cord; it will eliminate brainstem neurons that project to there. Currently, this task is best suited to immunotoxins since there is little data on using neuropeptide toxin conjugates to produce retrograde lesions. If armed to kill, a growth factor such as NGF might also work, if toxin conjugation did not damage binding and intracellular trafficking of the NGF. But these are special cases (which you can find in our reference lists for these products).

Targeting presynaptic antigens that are common to all types of axon terminals would seem a dubious undertaking since success with an immunotoxin requires the target molecule be displayed on the external surface of the terminal and not be present at all on cell bodies or dendrites. I do not know of a suitable target molecule for this purpose.

See: Targeted Toxins

References

1. Wiley RG et al. Suicide Transport and Immunolesioning. , 1994. R.G. Landes, Houston
2. Waite JJ et al. Time course of cholinergic and monoaminergic changes in rat brain after immunolesioning with 192 IgG-saporin. Neurosci Lett 169:154-158, 1994.
3. Mantyh PW et al. Inhibition of hyperalgesia by ablation of lamina I spinal neurons expressing the substance P receptor. Science 278:275-279, 1997.
4. Seeger G et al. Electron microscopic evidence for microglial phagocytotic activity and cholinergic cell death after administration of the immunotoxin 192IgG-saporin in rat. J Neurosci Res 48:465-476, 1997.

Q: I spoke with someone from your technical service over the phone and got the impression that your product dermorphin-SAP (Cat. #IT-12) is not a retrograde and will only affect the terminals or the cells that express mu opioid receptors in the injection site in the brain. I have three questions:

1) Do you have any written document on this issue?

A: That the peptide-toxins don’t undergo retrograde transport is an example of negative data, so people haven’t really been publishing too much on that. But two articles deal specifically with it: Lappi and Wiley[1] and Bugarith et al.[2] The latter, in particular, presents solid data on the inability of the peptide ligand toxin NPY-SAP (Cat. #IT-28) to undergo retrograde transport. I don’t think we have a single example of a peptide-ligand toxin that undergoes retrograde transport. In order for a peptide-toxin to kill cells, the cell body must have the receptor and the toxin must be injected within reach of the cell body. We’ve made a mistake in not putting that in the data sheets, and will begin to change that.

2) Will dermorphin-SAP also kill terminals in the injection site or just cell bodies?

A: Let me cite for you: Tokuno et al., Efferent projections from the striatal patch compartment: anterograde degeneration after selective ablation of neurons expressing mu-opioid receptor in rats.[3] As the title implies, they address the issue of elimination of processes following cell body destruction.

3) If it also kills terminals, will it affect their remote cell bodies?

A: I’m not sure I understand this question, but that won’t stop me from trying to answer it: The situation is the contrary, because the destruction of processes comes from the action taking place in the cell body. Our experience is that once the cell body is gone, it’s just a matter of time for the process to go away. This makes these toxins a little different than others. In fact, we recommend that you wait two weeks at least to see immunohistological evidence of a toxic effect after injection of a saporin toxin in vivo. That’s how long it takes the removal process to get rid of all the antigens that you might want to use for evidence of cell loss.

Q: Can I inject NPY-SAP to destroy projections through retrograde transport?

A: Regarding NPY-SAP, a peptide-toxin, see previous response. The antibody-toxins such as 192-IgG-SAP (Cat. #IT-01) or anti-DBH-SAP (Cat. #IT-03) will undergo retrograde transport from terminals to cell bodies. Thus, you can put 192-IgG-SAP into the cortex and it will destroy neurons in the basal forebrain, because the saporin (probably the whole conjugate) is transported from the projection to the cell body. Likewise, anti-DBH-SAP in the spinal cord destroyed hindbrain catecholaminergic neurons by retrograde transport.[4] All the antibody-toxins appear to undergo retrograde transport. 

Finally, the lectin-toxins, CTB-SAP (Cat. #IT-14) and IB4-SAP (Cat. #IT-10) undergo retrograde transport, just like the native lectins do. CTB-SAP is well-described in Llewellyn-Smith et al.[5] and several others. For IB4-SAP, Vulchanova et al.[6] describe use, along with several other articles on our reference page. In addition, detailed discussions are available in the book Molecular Neurosurgery with Targeted Toxins,[7] available from Humana Press.

See: Targeted Toxins

References

1. Lappi DA et al. Entering through the doors of perception: characterization of a highly selective Substance P receptor-targeted toxin. Neuropeptides 34(5):323-328, 2000.
2. Bugarith K et al. Basomedial hypothalamic injections of neuropeptide Y conjugated to saporin selectively disrupt hypothalamic controls of food intake. Endocrinology 146(3):1179-1191, 2005.
3. Tokuno H et al. Efferent projections from the striatal patch compartment: anterograde degeneration after selective ablation of neurons expressing mu-opioid receptor in rats. Neurosci Lett 332(1):5-8, 2002.
4. Ritter S et al. Immunotoxic destruction of distinct catecholamine subgroups produces selective impairment of glucoregulatory responses and neuronal activation. J Comp Neurol 432(2):197-216, 2001.
5. Llewellyn-Smith IJ et al. Retrogradely transported CTB-saporin kills sympathetic preganglionic neurons. Neuroreport 10(2):307-312, 1999.
6. Vulchanova L et al. Cytotoxic targeting of isolectin IB4-binding sensory neurons. Neuroscience 108(1):143-155, 2001.
7. Wiley RG et al. Molecular neurosurgery with targeted toxins. , 2005. Humana Press, Totowa, New Jersey

Q: In a recent experiment using a saporin-antibody conjugate injected systemically we saw changes in dendritic cells that could be consistent with an LPS effect. Does ATS test for LPS and has this ever been identified as a problem before?

A: Yes, this can happen, but we here at ATS will swear innocence. One of our collaborators just reported the same thing (the first comment like that in several years), so I’ll tell you the story.

Generally our materials have a very low endotoxin content, on the order of less than 1 EU/mg protein. We check this occasionally because most of our customers provide an easy assay. That is, these things are often injected into the brains of rats, and they’ll die within a few minutes if there is an LPS (lipopolysaccharide) content such as you described. That would be disastrous for us, so we do pay attention to this issue. The recent situation was with an immunotoxin that was also injected systemically, and gave a response similar to what you’re stating. The material had been thawed and used in a set of experiments. Then, because of concerns about the effects of freezing and thawing, it was left in the refrigerator for a considerable period of time. It was then used in the experiments that gave the LPS-consistent result. We believe that this material was no longer sterile and that during the time between uses it grew bacteria. We went back and assayed the original material and it gave the usual less than 1 EU/mg value.

The bottom line is that once the sterility is broken, the material is a decent growth medium for bacteria (a protein in PBS). We filter sterilize all of our targeted toxins and package them in a sterile manner, but we do not add preservative. The thinking there is that in sensitive situations, a preservative can cause its own biological response/effect.

This situation causes us to print rather stringent use instructions: aliquot and store frozen at -20°C.

See: Targeted Toxins

Q: We just completed surgeries where we implanted third ventricular cannulas and temporary bilatera cannulas directed into the nucleus tractus solitarius in the brainstem of animals. We injected either the Blank-SAP control toxin (Cat. #IT-21) or the experimental material Oxytocin-SAP into the bilateral NTS cannulae over a 30-second period. However, within the next week — two weeks post-surgery, we lost 13 of the 19 animals treated; they appeared not to be able to groom properly and lost over 20% of their body weight. This was apparent in both the Blank-SAP and the Oxytocin-SAP groups. We gave a dose of 40 ng/300 nl for each of the reagents. This dose was determined based on a published article using another of ATS’s targeted toxins. I’m very surprised by my results. Can you offer any explanation/advice?

A: This is a particularly disturbing result; it appears that a dose was chosen by comparison to one used with another targeted toxin. Although this can be a good approximating tool to begin a dose-ranging study, it usually doesn’t take into account the tissue, system, target molecule — so many parameters that are important to determining the proper dosage. The literature is quite extensive on targeted toxins, and so there may be a comparable starting dose that has been published. Let’s use, for example, 4 mg. Reduce that amount by 20% quantities (4, 3.2, 2.4) and test in a small number of animals to determine a value that is safe and effective. If no trouble is seen at the highest dose, and the effect is minimal, that would indicate a higher dose may be acceptable. You can then test doses in 20% increased increments (4.8, 5.6, 6.4). The effects you see in your animals should only be reflective of the particular cell type you are eliminating. In the case of control reagents, such as Blank-SAP, no cell type is being targeted, so if you are seeing any kind of result, then you are certainly over-dosing.

Q: Is there some kind of formula that one can use that will help determine a starting point for establishing a range of doses to test in animals prior to initiating a study? For example, if the targeted toxin is administered intravenously, does it take more or less material than when administered directly into tissue?

A: Start with a few animals and do dose-ranging as discussed in the previous question. The various modes of application are really too wide to discuss in any detail here, but I, a biochemist by training, always like the approach of thinking about what sort of concentration will be needed to have a cytotoxic effect. Generally, these molecules have an ED50 in the nanomolar to picomolar range. Obviously if you inject systemically, the material from the first becomes greatly diluted, relative to an injection directly into tissue, and so you’ll need a lot more. If you inject directly into tissue the local concentration can be quite high.

See: Targeted Toxins, Control Conjugates

Q: You have stated previously that it was unlikely that saporin compounds or constituents would be excreted in urine or feces. However, you acknowledge that experimental data is lacking. Have there been any tests of animal urine or feces for saporin content? My animal care staff are concerned.

A: One of the reasons that no studies have been done on excretion of saporin is that there isn’t much on the theoretical side to cause concern. The primary issue is that the quantity used in mice (and even rabbits) is so small that when looked at in human terms (i.e., an animal 10 to 100-times larger), the dosage becomes insignificant. The LD50 for saporin in mice is 4-8 mg/kg;[1] that would translate in humans to more than you’ll ever use! The immunotoxins, which contain only about 20% saporin by weight, really do not contain all that much saporin.

Looking at it another way, you need a concentration of about 100 nM to see even a vague hint of toxicity of saporin to cells. In human blood, that would correspond to 24 mg injected systemically into a person. It would be really expensive for anyone to get close to that number.

As far as urine and feces go, the same calculations are appropriate, but there will be considerable degradation – the protein content in urine and feces is quite low and the probability is that you will be dealing with only saporin. Remember, saporin is a plant protein that is related to proteins in foods that we eat (cucumbers, for example).

Q: Are there any studies which indicate what doses of saporin (by itself or compounded with an antibody) would be hazardous if ingested or injected (i.e. systemic dose level resulting in death or organ dysfunction).

A: When there is an antibody that does recognize a human epitope (the human p75-saporin immunotoxin that is used in rabbits, for example), at about 1 pM one sees the slightest bit of toxicity to cells. That translates, if injected by error into a human blood supply, to about 170 micrograms. That also is a gigantic dose. I am using very conservative numbers here, and the bottom line is that you cannot accidentally reach such dangerous levels under normal handling situations.

Having said all this, we still recommend that our customers take excellent care of themselves and we state clearly that precautions should be taken by people handling these materials, just as they should use precautions with all laboratory chemicals. Please refer to the data sheets provided with our products for safety instructions.

See: Saporin (Cat. #PR-01)

References

1. Stirpe F et al. Hepatotoxicity of immunotoxins made with saporin, a ribosome-inactivating protein from Saponaria officinalis. Virchows Arch B Cell Pathol Incl Mol Pathol 53(5):259-271, 1987.

Q: We are re-examining some data collected using an immunotoxin not prepared by your company (VChAT-sap). Our results in vivo indicated that there were non-specific effects although the creators claimed it was specific. We ran a Western Blot and determined that about half the saporin was not bound to the antibody. This may have been the problem but I want to confirm the cytotoxicity of unbound saporin. Can you confirm that? Further, the antibody in question never bound selectively in ferret tissue. Does this suggest a problem as well? Can you give me information on the best way to design and test an immunotoxin.

A: Yes, saporin at higher concentrations can be cytotoxic. Without specific binding, you will only see non-specific cytotoxicity. The sequence of the immunogen for that antibody is from the rat protein, so I’m not sure if it would target the ferret protein (there usually is good sequence homology among transporters). We worked a lot with this antibody and with an immunotoxin (made by us), but never got any sort of results that would indicate that it was working. We also were greatly concerned that the epitope is an intracellular epitope, and so we have difficulty understanding, from a theoretical standpoint, how it even could work. Because of many concerns, we never commercialized it, and we believe all the effects in the literature were non-specific, but “credible” because of the unusual experimental system that was used.

The best way to design and test an immunotoxin is to talk to us. If your antibody has been tested and shown to be internalized by the cell you are targeting, there should be no problem with the activity. The conjugate will only work as well as your antibody does. We recommend that you try a second immunotoxin before having a custom conjugation performed. This allows you to use a secondary agent conjugated to saporin that “piggybacks” on your antibody and makes a second immunotoxin for use in vitro to test specificity and internalization of your antibody. You can check out the publication that talks about one of the secondary conjugates, Mab-ZAP (Cat. #IT-04). We also have many other secondary conjugates or you can biotinylate your material and use Streptavidin-ZAP (Cat. #IT-27). 

See: ZAP Conjugates, Custom Conjugates

References

1. Kohls MD et al. Mab-ZAP: A tool for evaluating antibody efficacy for use in an immunotoxin. BioTechniques 28(1):162-165, 2000.

Q: Why do your directions for SSP-SAP (Cat. #IT-11) state that it is to be used within hours after dissolution? To my knowledge, both proteins and peptides are stable in clean solution.

A: In fact, the two components of SSP-SAP (Stable Substance P and Saporin) are quite stable. However, we have found that many things happen in laboratories and some of them can impact stability. Probably the most severe is the loss of sterility. In that case, over time at room temperature or at 4°C, bacteria can grow on this rather excellent “medium.” This would cause inactivation. Because many laboratories, due to molecular biology work, have high levels of resident bacteria, we prefer to emphasize playing it safe.

Even if saporin is a stable protein, it is a protein and can suffer denaturation. This occurs more rapidly at room temperature than at 4°C, and hardly at all in the frozen state (really, it is stable for years when stored at -80°C). The maintenance of precise activity is of extreme importance to our customers who use these materials in vivo (their assays are very sensitive), and so we choose to advise the most conservative course.

Q: I understand that theoretically only one molecule of Saporin taken up by a cell is enough to induce cell death. I have been looking for literature on this topic but have not come across anything.

A: Definitely theoretical. The only article that we know of that states anything close to that is: Yamaizumi et al (1978) One molecule of diphtheria toxin fragment A introduced into a cell can kill the cell. Cell 15(1): 245-250. As you can see, this article speaks to the enzymatic chain of diphtheria toxin, which has a slightly different mechanism of action for shutting down protein synthesis, but otherwise is similar to saporin. In fact, we test all sorts of toxins against cells in controlled conditions, and we have only one candidate that is in this range; all the rest are orders of magnitude away. It takes more than thousands per cell. Another question would be: how many actually get in?

In another FAQ, you addressed the question of one molecule of saporin killing a cell. Your response overlooked the data on ricin, abrin and modeccin (Eiklid, Olsnes and Pihl, Exp Cell Res, 126:321-326, 1980). In that paper, they showed that these RIP toxins applied to cells in culture produce all-or-none lethality. They used radioactive amino acid uptake and incorporation (as memory serves) and found only two types of cells, those with absolutely no uptake of label or those that were entirely normal – nothing in between. Also, if the data on ricin-induced apoptosis is correct (numerous authors), and I believe it is, then at low doses, the cells die from triggering apoptosis which seem possible with a single molecule of RIP free in the cytoplasm. To further complete your answer, someone (I haven’t found the article yet) showed that it took, on average, about 10,000 molecules of ricin/cell to kill cells in culture. This gives a hint at the efficiency of internalization and translocation in that cell type. I am not aware anyone else has looked at these issues with saporin conjugates.

A: Overlooking literature is actually a favorite sport of mine, but in this case I would respectfully point out conflicting information. There is a study of something that is quite between an all-or-none phenomenon: Barbieri et al. FEBS Lett, 2003 Mar 13;538(1-3):178-82. These authors document that ribosome-inactivating proteins have transforming activity on the classic FDA assay cell line: NIH3T3 cells. This would be a non-toxic activity that one presumes is due to internalization, and is somewhat on the none side of all or none, but hey, it’s an activity nonetheless. 

See: Targeted Toxins

Q: What dosage of 192-Saporin (192-IgG-SAP, Cat. #IT-01) should be used in the lateral ventricle to eliminate cholinergic neurons in the basal forebrain, including substantia innominata (SI)? I read that Calza et al [1] used 2 or 3 micrograms/4.5 µl and found this was highly effective.

A: It has been our experience that two- or three-micrograms into the lateral ventricle is necessary to obtain a maximum cholinergic basal forebrain (CBF) lesion. However, these doses typically kill some cerebellar Purkinje cells. Another issue is that some cholinergic neurons in the NBM region are never killed by 192-Saporin.

Q: Should we expect to be able to kill all or almost all ChAT SI neurons?

A: Mesulam’s lab has some data [2,3] to suggest that these neurons innervate the amygdala and adjacent cortex. Generally lesions of the septum and diagonal band are complete, but when you get more caudal, i.e. SI region, there will be some cholinergic neurons left. When you do ChAT or AChE stains, the amygdala and adjacent cortex are not denervated whereas the hippocampus, olfactory system and all the rest of the cortex are devoid of cholinergic terminals.

Q: Is there another toxin that will eliminate the remaining ChAT SI neurons?

A: There may be other targeted conjugates that could clean out the residual cells in the SI region if we knew what markers they co-express. For example, our SSP-saporin (Cat. #IT-11) conjugate is very good at removing cells that express the NK-1 receptor such as striatal cholinergic interneurons.

See: 192-IgG-SAP (Cat. #IT-01), SSP-SAP (Cat. #IT-11), Targeted Toxins

References

1. Calza L et al. Neural stem cells and cholinergic neurons: Regulation by immunolesion and treatment with mitogens, retinoic acid, and nerve growth factor. Proc Natl Acad Sci U S A 100(12):7325-7330, 2003.
2. Heckers S et al. Two types of cholinergic projections to the rat amygdala. Neuroscience 60:383-397, 1994.
3. Heckers S et al. Complete and selective cholinergic denervation of rat neocortex and hippocampus but not amygdala by an immunotoxin against the p75 NGF receptor. J Neurosci 14:1271-1289, 1994.

Q: I recently tried to order avidinylated-SAP (Cat. #IT-09) and was told that this product has been replaced with a new product, Streptavidin-ZAP (Cat. #IT-27). Why did you replace avidinylated-SAP?

A: We initially had good results with avidinylated-SAP. It combined well with biotinylated antibody to produce extremely potent cytotoxic materials, and had low toxicity itself. However, the weaknesses of avidin are well-documented. Probably the most severe is its high isoelectric point that has been suggested to cause nonspecific binding. As we produced more batches of avidinylated-SAP and completed comparative studies, we in fact, found this to be the case.

Q: I use avidinylated-SAP to demonstrate that my antibody internalizes. It worked quite well for me.

A: A couple of months ago we received two reports from customers that they were seeing that, even in batches that had performed well in quality control testing, there was a non-specific cytotoxicity with some cells and/or cell lines. Since a major use of this material is to demonstrate internalization of the biotinylated targeting agent, this was an unacceptable situation. We changed to streptavidin to overcome these specificity issues. As shown in the figure, streptavidin-SAP has an excellent capacity to transform a biotinylated reagent into a potent cytotoxic targeting vehicle, while streptavidin-SAP alone has no detectable cytotoxicity.

See: Streptavidin-ZAP (Cat. #IT-27)