Comments on
Biochemical Purification of a Mammalian Slit Protein as a Positive Regulator of Sensory Axon Elongation and Branching Cell, 96, 771-784
Wang, Brose, Arnott, Kidd, Goodman, Henzel, Tessier-Lavigne

 Mammalian neurons often sprout additional branches which allow the cell to interact with several of its neighboring neurons rather than just the two to either pole.  These chaps, Wang et. al., claim theyíve isolated a protein which positively regulates this growth.  There has been work done previously with growth factors in neurons, but much remains to be discovered about growth in these long-lived cells. This paper focused on cells in the dorsal root ganglia (DRG), as these tend to exhibit growth of collateral branches. (Figure 1C-F ) Also, someone else has grown them before and the DRG cells cultured well.  The rat was the chosen species.
    Their chosen neurons were cultured in a thick (3-D to the cells) layer of collagen gel in light enough concentration to easily distinguish individual cells.  (Figure 1G&H)  There was some difficulty in maintaining a control­the cells in only the growth medium did not survive 24h.  Embryonic rat DRG cells from embryos 14 and 17 days old were grown to determine age differences.  Nerve growth factor­necessary in the medium­did not differentially promote branching with cells of these ages (Figure 1G, 1H).  The E17 cells did branch more, and there were indications that a protein was involved in this acceleration of growth.  The researchers added concentrates of the cell fraction to their cultures, and noted that the cell membrane fraction had the most effect.  This effect could be reduced by washing the fraction in 1M NaCl; this suggested that their desired protein may be peripherally associated with the neuron cell membrane.
    They tried what appropriate proteins they could guess (Table 1). That was unsuccessful, so they tested concentrated extracellular salt extracts from E13 and E17 spinal cord, extracts from other species, and found that calf brain extract worked admirably (Figure 2 D, G, H, I).  Calf brain extract was fractioned with centrifugation and chromatography and an activator of 140 kDa was isolated. A microsequence of the minute amount of isolate was revealed as  (benefiting from the universal sequencing projects) being similar to a Slit protein.  The 140 kDa protein showed the most homology with the Slit2 protein.
    The isolated 140 kDa protein and independently isolated Slit2 were tested on the neurons in media and were found to stimulate branching in to a similar degree as the calf brain extract.  To show that the Slit2 could be the in vivo activator, and not just an in vitro stimulator, spinal cords of E14 and E17 were mapped for mRNAs of Slit2, Slit1 and Slit3.  Also mapped were mRNAs for Robo proteins, proteins known to bind to Slit.  The Slit genes were being transcribed in the embryonic spine at the appropriate time, but this did not eliminate the possibility that Robo was the activity gene.

Figure 1.  This figure is primarily illustration to inform the reader unfamiliar with embryonic neuron development in mammals.  Significant is the panels that demonstrate the viability of both E14 and E17 cells (G&H) as potential study targets.

Figure 2.  Figure two records in photographic form the effects of some of the tried stimuli on the development of E14 cells. High salt extract from E13 spinal cord (B) and E17 spinal cord (C) were not as effective as the high salt calf brain extract (D, G&H).  This is a well controlled experiment (A, E, F).  The chart (I) quantifies the increase then leveling off of branching activity under higher concentration conditions of the calf brain extract.

Table 1.  The table includes all of the proteins they had on hand that would be reasonable candidates for the activity protein after they found the effects of the calf brain extract.  They would feel awfully silly having gone through the pain of isolating their target to find that they had 0.1 kg of it on a shelf.  The people in charge of budgets for the Howard Hughes wouldnít be amused in that circumstance either.

Figure 3.  These photographs illustrate their testing of the fractioning of the calf brain extract to localize the activity protein.  A is a schematic diagram of their chromatography process. B is control and C-E show the effects of these chromatography fractions, then the fractions together. The FT+eluate containing the activity protein suggests that the protein may be fragmenting.

Figure 4.  These are four chromatography set ups from the schematic diagram in 3A. The significance in these gels is the band at 140 kDa in each of them.  A semiquantitative (they couldnít use radioisotopes, so they tested uptake of some kind) examination of these gels found the 140 kDa band to be the most active in each chromatography design.

Figure 5.  This is the sequence of amino acids found in the 140 kDa band by their microsequencer.  This showed some homology with the Slit2-C protein.

Figure 6.  This controlled gel (the negative control shows that nothing else in the Cos cell extract carries the histidine marker) identifies the 140 kDa band as a fragment of the Slit2 protein and determines that the fragment and the whole of the protein may be isolated.

Figure 7.  This set of photographs shows the functioning of the purified amino terminus fragment of Slit2 and the whole Slit2 protein.  The fragment of Slit2 stimulated branching (G is the prettiest picture of neuron arborealization given in this paper) and the whole protein also did to a degree.

Figure 8.  This is a quantification of the effect of the isolated Slit2 protein.  The amino terminus fragment (the diamond) and the combination of that plus the eluate (the triangle) stimulate branching to a significant degree better than the controls.  The whole Slit2 protein (the asterisk) stimulates branching, but not, to my eye, significantly better than the controls.  These stimulation patterns are similar in both the neurite length and the amount of branching.  That the increased growth rate these two indicators are not directly
correlated indicates that branching is being stimulated, and is not just an outgrowth of increased size.

Figure 9.  This set of photographs illustrates the expression of Slit and Robo genes in E14 and E17 spinal cord.  The colorant has been in situ hybridized with the mRNAs of the various Slit and Robo proteins.  This establishes the expression of these genes at the proper (whatís the antonym for ectopic?) point in development, though it does not determine if they are translated

    I like this paper.  They solve their question early: a high salt extract of calf brain stimulates arborealization in E14 and E17 rat DRG cells.  They then spend quite some time isolating the activity agent and demonstrating that the Slit2 protein is the correct one.  Their experiments are all well designed, controlled, gain positive results for their arguments, demonstrate function, and are thorough.  There are more experiments I would like them to do, but these explore fine details or new questions.  I give it to them that Slit2 stimulates branching in developing DRG neurons.

    They suggest that Robo may also be the activity protein.  This they leave open to be explored in a future paper of theirs; it would be an obvious step to test the collagen grown cells with a Robo isolate, but more on that later.  They also make the suggestion that this may be a factor in repairing spinal trauma, should this have the same effect of arborealization stimulation in adult spinal tissue.  I would like to see more.

    A curious project that should be well within the scope of this technology would be to overstimulate the branching process E14-E17 in vivo and observing the surviving individuals.  Studies on neuron activity and interaction can be done with super branched individuals as compared to wildtype.  I know of plenty of genetic diseases by which deficient neural stimulation has been studied, but cannot recall any which, with healthy cells, creates super stimulation in an orderly fashion.  (I think that some involve uncontrolled neural activity, but this is a different matter.)
    Other neurons also branch (brain, the spinal neurons these DRG neurons are branching to reach, brain areas stimulated by practiceómusicianís finger control) and these have embryonic branching ages.  The collagen in vitro assay seems to work well; other types of neurons can be tested with Slit2 amino terminus fragments at the embryonic branching ages.  Also, neurons of any age can be tested.  Adult neurons must branch under Slit2 stimulus before ideas such as trauma treatments can be discussed.  If the neurons only branch at the prescribed windows in embryonic development, then there is no point in considering remedies to mad cow disease and other degenerative ailments.  Should the adult neural branching activity be heightened by the Slit2, it would lead to great use, as most permanent spinal trauma damage is from the atrophy of the muscles which remain useless after the natural cycle of branching bridges the gap. The prospect of abridging this time is tempting enough for this egg to be counted before it hatches.
    Slit2 fragmented in the chromatography columns and the amino terminus was more potent than the whole.  This should be furthered, until necessary and sufficient fragments are isolated. This would have many benefits.  If the interaction between Slit and Robo were well documented, the necessary portions of the Slit2 may answer whether Robo is involved as some sort of hybrid, or is necessary in an interaction. The smaller chunks of necessary Slit2 would also be key in identifying the complete pathway and the specific physical interaction between the Slit2 and the neuron.

To find out more about the department I signed away two good years of my life go to:

Davidson College Biology Department Home Page

To find out more about this particular wonderful class this assignment is for, go to:
                                                                        Biology 304
© Copyright 2000 Davidson, NC 28036
Send comments, questions, and suggestions to: