EP0563304A1 - Neuronal cholinergic differentiation factor - Google Patents

Abstract

The present invention relates to a neuronal cholinergic differentiation factor (FDCN) derived from a target, as well as to its therapeutic and diagnostic uses. The invention relates to FDCN, its derivatives, analogs and fragments, pharmaceutical compositions containing them as well as antibodies acting against FDCN. The FDCN of the invention is a protein present in extracts from mammalian sweat glands. and which demonstrates thermolability and lability to trypsin as well as an absence of sensitive bond to a heparin-agarose affinity column, and has an isoelectric point (pI) of between approximately 4.8 and 5.2, a non-membranous cellular location and a molecular weight between 16 and 32 kilodaltons. The FDCN protein, as well as its derivatives, analogs and fragments are capable of reducing the expression of tyrosine hydroxylase and of whole catecholamines and of increasing the expression of choline acetyltransferase and of the intestinal vasoactive peptide (PIV) via sympathetic neurons in a cell culture. The FDCN protein and its derivatives, analogs and fragments can be used to induce cholinergic activity in neurons. Such proteins, derivatives, analogs and fragments can be administered for therapeutic purposes to patients suffering from lesions or disorders of the nervous system where it is desirable to maintain the survival and / or cholinergic differentiation of a number of neural types .

Description

NEURONAL CHOLINERGIC DIFFERENTIATION FACTOR

1. INTRODUCTION 5 The present invention relates to a target- derived neuronal cholinergic differentiation factor (NCDF) , and the therapeutic and diagnostic uses thereof. The invention provides NCDF, and derivatives, analogs, and fragments thereof, pharmaceutical _jO compositions of the foregoing, as well as anti-NCDF antibodies.

2. BACKGROUND OF THE INVENTION Most sympathetic neurons are noradrenergic;

15 however, a minority population, including neurons that innervate sweat glands, are cholinergic. The sympathetic neurons that innervate sweat glands are further distinguished in that they contain vasoactive intestinal peptide (VIP) and calcitonin gene-related

20 peptide (CGRP) immunoreactivity (Lundberg et al., 1979, Neuroscience 4, 1539-1559; Landis and Fredieu, 1986, Brain Res. 377, 177-181; Lindh et al., 1989, Cell Tissue Res. 256, 259-273) , whereas many noradrenergic neurons contain neuropeptide Y (NPY) (Lundberg et al.,

25 1982, Acta. Physiol. Scand. 116, 477-480; Lundberg et al., 1983, Neurosci. Lett. 42, 167-172; Jarvi et al., 1986, Neurosci. Lett. 67, 223-227). Although the mature sweat gland innervation is functionally cholinergic, the developing innervation is

30 noradrenergic (Landis and Keefe, 1983, Dev. Biol. 98, 349-372; Leblanc and Landis, 1986, J. Neurosci. 6, 260- 265; Stevens and Landis, 1987, Dev. Biol. 123, 179-190; Landis et al., 1988, Dev Biol. 126, 129-140). When sympathetic axons first innervate the developing sweat

*-* glands, they possess intense catecholamine histofluorescence and immunoreactivity for the catecholamine synthetic enzymes, tyrosine hydroxylase and dopamine β_ hydroxylase. As the gland innervation matures, catecholamine histofluorescence disappears, tyrosine and dopamine β hydroxylase immunoreactivities decrease, and cholinergic and peptidergic properties 5 appear. For example, acetylcholinesterase is detectable at postnatal day 7 (P7) , VIP immunoreactivity at P10, choline acetyltranferase activity at Pll, and cholinergic transmission at P14. Thus, the cholinergic sympathetic innervation of sweat 10 glands undergoes a striking change in neurotransmitter properties during postnatal development.

Several lines of evidence indicate that the change from noradrenergic to cholinergic function in the developing sweat gland innervation is mediated by 15 interactions with the target tissue. First, when the innervation of developing sweat glands is delayed by 7- 10 days, there is a corresponding delay in the disappearance of catecholamine histofluorescence and the appearance of cholinergic properties (Stevens and 0 Landis, 1988, Dev. Biol. 130, 703-720) . Second, if the superior cervical ganglion, which contains noradrenergic sympathetic neurons, is transplanted to the anterior chamber of the eye with footpad tissue containing sweat glands, the neurons innervate the 5 glands, reduce their expression of catecholamine histofluorescence and NPY, and develop immunoreactivity for choline acetyltransferase and VIP (Stevens and Landis, 1990, Dev. Biol. 137, 109-124). Finally, cross-innervation experiments provide direct evidence 0 for a target role. When footpad skin is transplanted in place of hairy skin in the thoracic region of early postnatal rats, the transplant is innervated by sympathetic neurons whose normal targets are piloerectors and blood vessels. The sympathetic fibers ^*-^> that innervate hairy skin are noradrenergic and do not normally contain choline acetyltransferase activity, acetylcholinesterase staining, or VIP immunoreactivity (Schotzinger and Landis, 1990, Cell Tissue Res. 260, 575-587) . Several weeks after innervating the 5 transplanted sweat glands, however, the fibers show a marked reduction in catecholamine fluorescence and express properties characteristic of the innervation of their novel target: they exhibit choline acetyltransferase activity, acetylcholinesterase

10 staining and VIP immunoreactivity (Schotzinger and Landis, 1988, Nature 335, 637-639; and unpublished data) . Conversely, if parotid gland, a target of noradrenergic sympathetic neurons, is transplanted to the footpad in place of the sweat glands, it is

15 innervated predominantly by fibers that normally innervate sweat glands and become cholinergic; in this case, the fibers innervating the transplanted parotid fall to acquire cholinergic properties and continue to express catecholaminergic properties typical of the

20 sympathetic innervation of the parotid glands

(Schotzinger and Landis, 1990, Neuron 4, 91-100). Thus, the normal expression of cholinergic properties in the sweat gland innervation depends on the presence of this particular target, and sweat glands are able to

25 induce cholinergic and certain peptidergic properties in sympathetic neurons that would not normally express them. Since sympathetic axons never contact sweat gland cells or the basal lamina that surrounds them directly (Landis and Keefe, 1983, Dev. Biol. 98, 349-

30372; Uno and Montagna, 1975, Cell Tissue Res. 158, 1- 13; Quick et al., 1984, Anat. Rec. 208, 491-499), it seems likely that the target effect is mediated by a soluble factor.

Several proteins that cause a similar noradrenergic to cholinergic switch in sympathetic neurons developing in cell culture have been identified. These represent potential candidates for the differentiation signal produced by sweat glands. The cholinergic differentiation factor (CDF) purified 5 from heart cell conditioned medium (Patterson and Chun, 1977, Dev. Biol. 56, 263-280; Fukada, 1985, Proc. Natl. Acad. Sci. USA 82, 8795-8799) has been shown to be identical to leukemia inhibitory factor (LIF) (Yamamori et al., 1989, Science 246, 1412-1416). Ciliary 10 neurotrophic factor (CNTF) , originally identified as a survival factor for ciliary neurons (Adler et al., 1979, Science 204, 1434-1436; Barbin et al., 1984, J. Neurochem. 43, 1468-1478; Manthorpe et al., 1986, Brain Res. 367, 282-286), and recently cloned (Lin et al., 151989, Science 246, 1023-1025; Stockli et al., 1989, Nature 342, 920-923), induces cholinergic and reduces catecholaminergic function in cultured sympathetic neurons (Saadat et al., 1989, J. Cell Biol. 108, 1807- 1816) . Comparison of the deduced amino acid sequences 0 (Yamamori et al., 1989, Science 246, 1412-1416; Stockli et al., 1989, Nature 342, 920-923) and of the biological and immunological properties of CDF/LIF and CNTF (Rao et al., 1990, Dev. Biol. 139, 65-74) indicates that^{^}they are distinct factors. In addition, 5 a soluble 50 kd cholinergic factor has been obtained from brain by heparin affinity chromatography (Kessler et al., 1986, Proc. Natl. Acad. Sci. USA 83, 3528- 3532) , and a membrane-associated neurotransmitter- stimulating factor (MANS) has been solubiiized and 0 partially purified from rat spinal cord. The latter activity is associated with a 29 kd protein (Wong and Kessler, 1987, Proc. Natl. Acad. Sci. USA 84, 8726- 8729; Adler et al., 1989, Proc. Natl. Acad. Sci. USA 86, 1080-1083) . It is as yet unclear, however, whether ⁵ the cholinergic-inducing ability of these factors represents their primary, or even a relevant, function in normal development. Several of these factors have been shown in cell culture systems to have additional functions. For example, CDF/LIF inhibits proliferation 5 and induces macrophage differentiation in the Ml myeloid cell line (Hilton et al., 1988, Anal. Biochem. 173, 359-367) and maintains the developmental potential of embyronic stem cells (Smith et al., 1988, Nature 336, 688-690; Williams et al., 1988, Nature 336, 684- 0687) ; CNTF has trophic activity for ciliary neurons (Barbin et al., 1984, J. Neurochem. 43, 1468-1478; Manthorpe et al., 1986, Brain Res. 367, 282-286) and induces astrocytic properties in 0-2A progenitor cells (Hughes et al., 1988, Nature 335, 50-73).

15 Studies of cholinergic induction in cultured sympathetic neurons by heart and skeletal muscle cell conditioned medium have shown that neurons, once induced to acquire cholinergic function, maintain it for a period even if the inducing factor is removed 0 from the culture medium (Patterson and Chun 1977, Dev. Biol. 56, 263-280; Vidal et al., 1987, Development 101, 617-625) . Peptidergic induction is observed in cultured sympathetic neurons with CDF/LIF (Nawa et al., 1990, Neuron 4, 269-277); withdrawal of CDF/LIF results 5 in the return of substance P content to control levels. Cross-innervation experiments involving adult sensory nerves have shown that peptide phenotype can be altered (McMahon and Gibson, 1987, Neurosci. Lett. 73, 9-15) . When a muscle nerve is cross-anastomosed to a 0 cutaneous nerve and induced to innervate targets in the skin, the regenerated muscle nerves appear to acquire immunoreactivity for substance P, and conversely, when the cutaneous nerve was cross-anastomosed to a muscle nerve, substance P immunoreactivity is decreased in the cutaneous nerve. 3. SUMMARY OF THE INVENTION The present invention is directed to a target-derived neuronal cholinergic differentiation factor (NCDF) , and the therapeutic and diagnostic uses thereof. The invention provides NCDF, and derivatives, analogs, and fragments thereof, pharmaceutical compositions containing the foregoing, as well as anti- NCDF antibodies.

The NCDF of the invention is a protein present in extracts of mammalian sweat glands, which exhibits heat and trypsin lability, lack of substantial binding to a heparin-agarose affinity column, an isoelectric point (pi) in the range of approximately 4.8 to 5.2, a non-membrane cellular localization, and an approximate molecular weight in the range of 16 to 32 kilodaltons. The NCDF protein, its derivatives, analogs, and fragments are able to reduce the expression of tyrosine hydroxylase and of total catecholamines, and increase the expression of choline acetyltransferase and vasoactive intestinal peptide (VIP) , by sympathetic neurons in cell culture (in vitro) .

The NCDF protein, its derivatives, analogs, and fragments,^' can be used to induce cholinergic activity in neurons. Such proteins, derivatives, analogs and fragments can be administered therapeutically to patients with nervous system damage or diseases where it is desirable to support survival and/or cholinergic differentiation of a number of neuronal types.

4. DESCRIPTION OF THE FIGURES Figure 1. Soluble protein extracted from sweat glands, hairy skin, parotid gland, liver, or sciatic nerve of adult rats was added to cultures of dissociated sympathetic neurons. Seven days after the addition of extracts, neurons were homogenized and aliquots were assayed for levels of choline acetyltransferase (ChAT) activity by the method of Fonnum (1969, Biσchem. J. 115, 465-472). Samples were run in triplicate. In (a) , 250 μg of protein extracted from the indicated tissues was added. The data are expressed as the fold induction of ChAT activity compared with that present in control cultures grown without added extract, in (b) 250 μg of protein extracted from sciatic nerve or sweat gland was added. The data are expressed as the fold induction of specific activity per mg of protein added. Figure 2. (a) Increasing concentrations of sweat gland extracts cause increased induction of choline acetyltransferase activity. Increasing concentrations of soluble protein extracted from sweat glands of adult rats were added to sympathetic neuron cultures. Seven days after the addition of extract, neurons were homogenized and aliquots were assayed for choline acetyltransferase activity by the method of Fonnum (1969, Bioche . J., 115, 465-472). Samples were run in triplicate. Data are expressed as pmol of activity per min per well ± SD.

(b) Time course of induction of choline acetyltransferase activity. Soluble protein (100 μg/ml) extracted from adult sweat glands was added to sympathetic neuron cultures. Duplicate samples were homogenized at appropriate intervals after the addition of extract and assayed for choline acetyltransferase activity. Data are expressed as pmol of activity per min per well ± SD.

Figure 3. Sweat gland extracts reduce tyrosine hydroxylase. Sympathetic neurons were grown in medium without sweat gland extract (a) or with 100 μg/ml sweat gland extract (b) . Samples were pooled from several wells, homogenized in sample buffer, electrophoresed, and blotted onto nitrocellulose. The ₅ blots were probed with a monoclonal antibody to tyrosine hydroxylase (inset) . The laser densitometer scan (absorbance of 600 nm) of the staining intensity of the bands from control and treated cultures is shown. 10 Figure 4.

(a) Sweat gland extracts modulate the expression of VIP. Serial dilutions of the soluble protein extracted from adult rat sweat glands were added to sympathetic neuron cultures. Cultures were

15 harvested on the eighth day after the addition of extract. Sister wells were assayed either for VIP levels using radioimmunoassay or for choline acetyltransferase activity. All samples were run in duplicate. The data are expressed as pg of VIP per 0 well ± SD or as pmol of choline acatyltransferase activity per well ± SD.

(b) Sweat gland extracts reduce the levels of NPY and elevate the levels of VIP. Sweat gland extracts (100 μg/ml) were added to sympathetic neuron 5 cultures. Cultures were harvested on the eighth day after the addition of extract. Sister wells were assayed for VIP or for NPY by radioimmunoassay. All samples were run in triplicate. Data are expressed as pg of VIP or NPY per well ± SD. 0 Figure 5. Appearance of cholinergic differentiation activity in sweat gland extracts. Sweat gland extracts were prepared from animals at the indicated ages. Approximately equal protein concentrations (100 μg/ml) were added to sympathetic ^£-* neuron cultures. Seven days after the addition of extract, neurons were harvested, and aliquots were assayed for choline acetyltransferase activity. At least three different preparations at each age were tested. Data are expressed as fold induction of 5 choline acetyltransferase per mg of extract protein ± SD.

Figure 6. The cholinergic-inducing activity present in sweat gland extracts is not immunopreci itated with antibodies to CDF/LIF.

10 (a) Sweat gland extracts (DEAE fraction) were incubated with protein A-Sepharose (A) , affinity- purified antibodies to the N-terminal sequence of CDF (B) , or affinity-purified antibodies preincubated with the peptide antigen (C) . After immunoprecipitation,

15 supernatants were added to sympathetic neuron cultures. Ten days after the addition of extract, the cultures were assayed for choline acetyltransferase activity by the method of Fonnum (1969, Biochem. J. , 115, 465-472) . The results are expressed as the fold induction of

20 choline acetyltransferase compared with the activity observed in neurons grown in medium without extract. All samples were run in duplicate.

(b) ¹²⁵I-labeled recombinant CDF/LIF (20,000 cpm) was incubated with affinity-purified antibodies to

25 the N-terminal sequence of CDF (lanes 1 and 3) or affinity-purified antibodies preincubated with the peptide antigen (lanes 2 and 4) in buffer (lanes 1 and 2) or with 10 μg of soluble protein extracted from adult rat sweat gland (lanes 3 and 4) . Following

30 immunoprecipitation by affinity purified antibodies to the N-terminal sequence of CDF, the labeled proteins were extracted by boiling in SDS sample buffer and subjected to SDS-PAGE in a 10% gel. The labeled LIF

(arrow) was localized on X-ray films developed after a

^*v_ 7 day exposure. The higher molecular weight band is labeled bovine- serum albumin immunoprecipitated by protein A-Sepharose.

FIG. 7. CNTF is not detectable in sweat gland extracts. In panel a, 10 ng of recombinant CNTF 5 was blotted onto nitrocellulose. In panels b and c, 60 μg of soluble protein (DEAE fractions) from sciatic nerve extract (lane 1) , from hairy skin extract of adult rat (lane 2) or from sweat gland extract of adult (lane 3) or 21 day (lane 4) animals (panels b and c) 10 were blotted onto nitrocellulose. In panels a and b, the blots were probed with a polyclonal antiserum raised against recombinant rat CNTF, while in panel c the blot was probed with antibody preincubated with 10 μm recombinant CNTF. Panel a documents that the 15 antiserum recognizes CNTF (arrowhead) . As expected, the antiserum recognizes a 24 kilodalton (kd) band present in sciatic nerve extracts (lane 1 b,c) , but no specific bands were evident in hairy skin extracts (lane 2) or in sweat gland extracts from 21 day (lane 203) or adult (lane 4) animals. Arrowheads in b and c indicate 92, 30 and 22.5 kd standards.

Figure 8. CNTF message is not detectable in sweat gland extracts. 30 μg of total RNA from adult sciatic nerve^"(a) , sweat gland (b) , liver (c) and optic 5 nerve (d) was electrophoresed and transferred onto nylon membrane. The membrane was then probed with an oligonucleotide probe to rat CNTF. Arrow shows a positive 1.3 kb band in lane a containing sciatic nerve RNA and a fainter band in the same position in optic 0 nerve (d) . No specific signal is detected in lanes b and c containing sweat gland and liver RNA, respectively.

Figure 9. In Situ Hybridization. Sections of sciatic nerve were probed with an oligonucleotide ⁵ probe to rat CNTF. Panel a shows specific hybridization to Schwann cells in sciatic nerve sections (with an antisense probe) . Panel b shows the same tissue section stained with ethidium bromide. No binding is seen with the sense (control) probe in Panel c, which shows a random distribution of grains. Panel d shows the same tissue section stained with ethidium bromide.

Figure 10. In Situ Hybridization. Sections of sweat gland were probed with an oligonucleotide probe to rat CNTF, as used in Fig. 9. No specific binding is seen in sections of sweat glands (Panel a) . No binding is seen with the sense (control) probe (Panel c) . Panels b and d represent ethidium bromide stained sections. Figure 11. Anion exchange chromatography.

After homogenization and centrifugation as described in Section 6.3.3., the sweat gland extract supernatant was applied to a DEAE ion exchange column, and assayed for choline acetyltransferase (ChAT) inducing activity in sympathetic neurons. Closed squares: ChAT induction. Closed diamonds: NaCl gradient.

Figure 12. Chromatofocussing. The DEAE eluate was chromatographed on a MONO P column, and 0.5 ml extracts we^'re collected and assayed for cholinergic activity (ChAT inducing activity in sympathetic neurons) . Closed squares: ChAT induction. Closed diamonds: pH.

Figure 13. (a) SDS gel fractions betwen 22- 26 kd and 26-32 kd were eluted and added to cultures of dissociated sympathetic neurons. Seven days after the addition of extracts, neurons were homogenized and aliquots were assayed for levels of choline acetyltransferase (ChAT) activity by the method of Fonnum. Samples were run in triplicate. The data are expressed as the fold induction of ChAT activity compared to that present in control cultures grown without added extract.

In b, aliquots of the eluted protein were rerun on an SDS gel and stained with Coomassie blue. Lane a shows the 22-26 kd (lower arrow) fraction and lane b the 26-32 (upper arrow) kd fraction.

5. DETAILED DESCRIPTION OF THE INVENTION The present invention is directed to a 10 target-derived neuronal cholinergic differentiation factor (NCDF) , and the therapeutic and diagnostic uses thereof. The invention provides NCDF, and derivatives, analogs, and fragments thereof, pharmaceutical compositions containing the foregoing, as well as anti- 15 NCDF antibodies.

The NCDF of the invention is a protein present in extracts of mammalian sweat glands, which exhibits heat and trypsin lability, lack of substantial binding to a heparin-agarose affinity column, an 0 isoelectric point (pi) in the range of approximately 4.8 to 5.2, a non-membrane cellular localization, and an approximate molecular weight in the range of 16 to 32 kilodaltons. The NCDF protein, its derivatives, analogs, and fragments are able to reduce the 5 expression of tyrosine hydroxylase and of total catecholamines, and increase the expression of choline acetyltransferase and vasoactive intestinal peptide (VIP) , by sympathetic neurons in cell culture (in vitro) . 0 The NCDF protein, its derivatives, analogs, and fragments, can be used to induce cholinergic activity in neurons. Such proteins, derivatives, analogs and fragments can be administered therapeutically to patients with nervous system damage ^**** or diseases where it is desirable to support survival and/or cholinergic differentiation of a number of neuronal types.

In a specific embodiment of the invention, the NCDF protein is that found in extracts of human sweat glands. In another embodiment, the NCDF protein is that found in extracts of sweat glands from rats.

In a further embodiment of the invention, the NCDF protein, its derivatives, analogs, and fragments may induce the expression of additional peptides such as enkephalin, somatostatin, and substance P.

As detailed in the examples sections, infra, we assayed the effects of sweat gland extracts on the transmitter properties of cultured sympathetic neurons. We found that there exists a soluble factor present in sweat glands (which we term NCDF) , which reduces the expression of catecholamines and tyrosine hydroxylase and induces the expression of choline acetyltransferase and VIP.

5.1. THE NCDF PROTEIN, DERIVATIVES,

ANALOGS AND FRAGMENTS

Human, rat, porcine and other species NCDF, or their functional equivalents, can be used in accordance with the invention. Additionally, the invention relates to NCDF proteins isolated from ovine, bovine, feline, avian, equine, or canine, as well as primate sources and any other species in which NCDF activity exists. The invention also provides for NCDF proteins, fragments and derivatives thereof or their functional equivalents. The invention also provides fragments or derivatives of NCDF proteins which comprise antigenic determinant(s) or which are functionally active. As used herein, functionally active shall mean having positive activity in assays for known NCDF function, e.g. the ability to increase the expression^' of choline acetyltransferase by sympathetic neurons in vitro.

The NCDF derivatives, analogs, or fragments of the invention include, but are not limited to, those containing all or part of the primary amino acid sequence contained in the full-length NCDF protein as purified from sweat gland extracts, including altered sequences in which functionally equivalent amino acid residues are substituted for residues within the sequence resulting in a silent change. For example, one or more amino acid residues within the sequence can be substituted by another amino acid of a similar polarity which acts as a functional equivalent, resulting in a silent alteration. Substitutes for an amino acid within the sequence may be selected from other members of the class to which the amino acid belongs. For example, the nonpolar (hydrophobic) amino acids include alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan and methionine. The polar neutral amino acids include glycine, serine, threonine, cysteine, tyrosine, asparagine, and glutamine. The positively charged (basic) amino acids include arginine, lysine and histidine. The negatively charged (acidic) amino acids include aspartic acid and glutamic acid. Also included within the scope of the invention are NCDF proteins, fragments, analogs or derivatives thereof which are modified, e.g. _f by proteolytic cleavage, linkage to an antibody molecule or other cellular ligand, acetylation, formylation, oxidation, reduction, etc.

5.2. PURIFICATION OF NEURONAL CHOLINERGIC DIFFERENTIATION FACTOR NCDF may be purified from any available source of mammalian sweat glands using techniques known in the art. Such techniques include but are not limited to chro atography (e.g.. ion exchange, affinity, and sizing column chromatography) , centrifugation, differential solubility, or by any 5 other standard technique for the purification of proteins.

For example, and not by way of limitation, it is envisioned that NCDF may be isolated from sweat gland extracts according to the following method. 0 Sweat gland extracts may be prepared according to the method set forth in Section 6.3.3. After homogenization and centrifugation as set forth therein, the supernatant may be collected and applied to an anion exchange column (e.g. DEAE, Whatman DE52 5 cellulose equilibrated in phosphate buffer) , and collected therefrom by methods known in the art. Purified extract may then be subjected to sucrose gradient centrifugation by known methods, with the appropriate fraction concentrated by ultra filtration. 0 The purified NCDF may then be subjected to analytic or preparative polyacrylamide gel electrophoresis. If desired, following elution from a polyacrylamide gel, NCDF may be further purified and freed from certain buffer components by use of a HPLC reverse phase 5 column.

As an example of gel electrophoresis, purified NCDF may be analyzed using a slab SDS- polyacrylamide gel. Purified NCDF or molecular weight standards may be electrophoresed and the gel cut out 0 and processed as follows: the polypeptides may be visualized without fixation by precipitating the protein-associated SDS during an incubation of the gel in 0.25 M KC1 and recording the positions of the standards and NCDF bands. Lanes may then be fixed and ^ stained with Coomassie blue. Other lanes may then be cut into slices, and eluted, e.g. by electroelution or by incubation with Triton X-100, and then the eluates may be assayed for NCDF activity.

5.3. NCDF BIOASSAYS

NCDF activity may be evaluated using any NCDF-sensitive In vivo or in vitro systems. For example, assays including including but not limited to those described in Section 6.3., infra, may be used, e.g., those assaying the ability to increase the expression of choline acetyltransferase or increase the expression of vasoactive intestinal peptide, or reduce the expression of tyrosine hydroxylase, or reduce the expression of total catecholamines, by sympathetic neurons in cell culture.

Alternatively, as but another example, it is envisioned that NCDF activity may be measured by quantitating 24-hour survival of embryonic (E8) chick ciliary ganglion (CG) neurons in monolayer cultures. For example, ciliary ganglia may be collected from E8 chick embryos, dissociated (yielding approximately 20,000 cells per ganglion) and then diluted in HEBM medium contain —iVng 20 percent horse serum as described in Varon et ai. (1979, Brain Res. 173, 29-45). About fifty microliters of cell suspension containing 1,000 neurons (2,000 cells) may then be seeded into icrotitre dishes and then putative NCDF activity may be added. Culture plates may then be maintained at 37°C in 5% C0₂ for 24 hours, after which the cultures may be fixed by the addition of 200 μl 2 per cent glutaraldehyde in HEBM medium, and the number of surviving neurons may be determined visually by direct count under phase contrast microscopy.

5.4. SEQUENCING NCDF The NCDF protein may be sequenced directly or initially cleaved by any protease or other compound known in the art, including, but not limited to, Staphylococcus aureus V8, trypsin, and cyanogen bromide. Peptides may be sequenced by automated Edman degradation on a gas phase microsequencer according to the method of Hewick et al. (1981, J. Biol. Chem. 256, 7990-7997) and Hunkapiller et al. (1983, Methods Enzymol. 91, 227-236) . Detection of phenylthiohydantoin amino acids may then be performed according to Lottspeich (1985, Chromatography 326, 321- 327) . Overlapping fragments of amino acid sequence may be determined and used to deduce longer stretches of contiguous sequence.

5.5. GENERATION OF ANTI-NCDF ANTIBODIES

According to the invention, NCDF protein, or fragments or derivatives thereof, may be used as im unogen to generate anti-NCDF antibodies. Various procedures known in the art may be used for the production of polyclonal antibodies to epitopes of NCDF. For the production of antibody, various host animals can be immunized by injection with NCDF protein,^" or a fragment or derivative thereof, including but not limited to rabbits, mice, rats, etc. Various adjuvants may be used to increase the immunological response, depending on the host species, and including but not limited to Freund's (complete and incomplete) , mineral gels such as aluminum hydroxide, surface active substances such as lysolecithin, pluronic polyols, polyanions, peptides, oil emulsions, keyhole limpet hemocyanins, dinitrophenol, and potentially useful human adjuvants such as BCG (Bacille

Calmette-Guerin) and, Corynebacterium parvum. If desired, to further improve the likelihood of producing an anti-NCDF immune response, once obtained, the amino acid sequence of NCDF may be analyzed in order to identify portions of the molecule 5 which may be associated with increased immunogenicity. For example, the amino acid sequence may be subjected to computer analysis to identify surface epitopes, according to the method of Hopp and Woods (1981, Proc. Natl. Acad. Sci. U.S.A. 78, 3824-3828). 10 For preparation of monoclonal antibodies directed toward NCDF, any technique which provides for the production of antibody molecules by continuous cell lines in culture may be used. For example, the hybridoma technique originally developed by Kohler and 15 Milstein (1975, Nature 256, 495-497), as well as the trioma technique, the human B-cell hybridoma technique (Kozbor et al., 1983, Immunology Today 4, 72), and the EBV-hybridoma technique to produce human monoclonal antibodies (Cole et al., 1985, in "Monoclonal 20 Antibodies and Cancer Therapy," Alan R. Liss, Inc. pp. 77-96) and the like are within the scope of the present invention.

The monoclonal antibodies for therapeutic use may be human monoclonal antibodies or chimeric hu an- 25 mouse (or other species) monoclonal antibodies. Human monoclonal antibodies may be made by any of numerous techniques known in the art (e.g.. Teng et al., 1983, Proc. Natl. Acad. Sci. U.S.A. 80, 7308-7312; Kozbor et al., 1983, Immunology Today 4, 72-79; Olsson et al., 301982, Meth. Enzymol. 92, 3-16). Chimeric antibody molecules may be prepared containing a mouse antigen- binding domain with human constant regions (Morrison et al., 1984, Proc. Natl. Acad. Sci. U.S.A. 81, 6851,

Takeda et al. , 1985, Nature 314, 452. 5 A molecular clone of an antibody to a NCDF epitope can be prepared by known techniques. Recombinant DNA methodology (see e.g., Maniatis et al., 1982, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York) may be used to construct nucleic acid sequences which encode a monoclonal antibody molecule, or antigen binding region thereof.

Antibody molecules may be purified by known techniques, e.g.. im unoabsorption or immunoaffinity chromatography, chromatographic methods such as HPLC (high performance liquid chromatography) , or a combination thereof, etc.

Antibody fragments which contain the idiotype of the molecule can be generated by known techniques. For example, such fragments include but are not limited to: the F(ab')₂ fragment which can be produced by pepsin digestion of the antibody molecule; the Fab' fragments which can be generated by reducing the disulfide bridges of the F(ab')₂ fragment, and the 2 Fab or Fab fragments which can be generated by treating the antibody molecule with papain and a reducing agent.

5.6. UTILITY OF THE INVENTION The present invention relates to NCDF and to peptide fragments, analogs, or derivatives produced therefrom. NCDF protein, peptides, and derivatives, and anti-NCDF antibodies, may be utilized.in diagnostic and therapeutic applications. For most purposes, it is preferable to use NCDF from the same species for diagnostic or therapeutic purposes, although cross- species utility or NCDF may be useful in specific embodiments of the invention. 5.6.1. DIAGNOSTIC APPLICATIONS The present invention, which relates to NCDF protein, peptide fragments, or analogs or derivatives produced therefrom, as well as antibodies directed against NCDF protein, peptides, or derivatives, may be utilized to diagnose diseases and disorders of the nervous system which may be associated with alterations in the pattern of NCDF expression.

Assays can be used to detect, prognose, diagnose, or monitor conditions, disorders, or disease states associated with changes in NCDF expression, including, in particular, conditions resulting in damage and degeneration of neurons which may respond to NCDF, such as parasympathetic neurons, cholinergic neurons, spinal cord neurons, neuroblastoma cells and cells of the adrenal medulla. Such diseases and conditions may include but are not limited to trauma, infarction, infection, degenerative nerve disease, malignancy, or post-operative changes including but not limited to Alzheimer's Disease, Parkinson's Disease, Huntington's Chorea, and amyotrophic lateral sclerosis.

In alternate embodiments of the invention, antibodies directed toward NCDF protein, peptide fragments, analogs or derivatives can be used to diagnose diseases and disorders of the nervous system, including, in particular, those neuronal populations and clinical disorders and diseases listed supra. Antibodies directed toward NCDF proteins of the invention can be used, for example, in in situ hybridization techniques using tissue samples obtained from a patient in need of such evaluation. In a further example, the antibodies of the invention can be used in ELISA procedures to detect and/or measure amounts of NCDF present in tissue or fluid samples; similarly, the antibodies of the invention can be used in Western blot analysis to detect and/or measure NCDF present in tissue or fluid samples.

The immunoassays which can be used to detect or measure NCDF protein, its analogs, derivatives or fragments, include but are not limited to competitive and non-competitive assay systems using techniques such as radioimmunoassays, ELISA (enzyme linked immunosorbent assay) , "sandwich" immunoassays, preσipitin reactions, gel diffusion precipitin reactions, immunodiffusion assays, agglutination assays, complement-fixation assays, immunoradiometric assays, fluorescent immunoassays, protein A immunoassays, and immunoelectrophoresis assays, to name but a few. In further embodiments of the invention, NCDF protein, peptide fragments or derivatives can be used to diagnose diseases and disorders of the nervous system. In a particular embodiment and not by way of limitation, labeled NCDF protein or peptide fragments can be used to identify tissues or cells which express the NCDF receptor in order to identify aberrancies of NCDF receptor expression and consequently, potential abnormalities in the tissue or cellular response to NCDF.

5.6.2. THERAPEUTIC APPLICATIONS The present invention, which relates to NCDF protein, peptide fragments, analogs or derivatives produced therefrom, as well as to antibodies directed against NCDF protein, peptides, analogs or derivatives, may be utilized to treat diseases and disorders of the nervous system which may be associated with alterations in the pattern of NCDF expression or which may benefit from exposure to NCDF or anti-NCDF antibodies. NCDF, and its derivatives, fragments, and analogs, can be used to support the survival and cholinergic differentiation of a number of neuronal types, including spinal motor neurons, parasympathetic neurons of the ciliary ganglion, etc. The NCDF products of the present invention may have utility in supporting jj\ vivo the survival and differentiation of certain cell populations, including but not limited to, spinal motor neurons, parasympathetic neurons (including ciliary ganglion neurons which innervate the iris, heart, gastrointestinal tract and other visceral structures) . Thus, in specific embodiments of the present invention, a pharmaceutical preparation containing NCDF or its active derivatives, fragments or analogs, can be administered to patients in whom the central nervous system is damaged. In another embodiment of the invention, a pharmaceutical preparation containing NCDF or its active derivatives, fragments, or analogs, alone or in combination with another neurotrophic factor (e.g. CNTF, NGF, BDNF (brain-derived neurotrophic factor) , or NT-3 (neurotrophin-3)) can be administered to patients suffering from pathological conditions resulting from damage to or fifiysiological imbalance, overactivity or underactivity of the autonomic nervous system, or conditions which might be aggravated by such autonomic nervous system damage or imbalance. Such disorders might include, but are not limited to: chronic anhidrosis and hyperhidrosis, cardiac arhythmias, chronic constipation, neurogenic bladder dysfunction and ejaculatory disturbances.

In various embodiments of the invention, NCDF protein, peptide fragments or derivatives can be administered to patients in whom the nervous system has been damaged by trauma, surgery, ischemia, infection (e.g. polio or- A.I.D.S.) , metabolic disease, nutritional deficiency, malignancy, or toxic agents. The invention in particular can be used to treat conditions in which damage has occurred to neurons, by administering effective therapeutic amounts of NCDF protein or peptide fragments or derivatives or analogs. In various specific embodiments of the invention, NCDF can be administered to spinal cord neurons which have been damaged, for example, by trauma, infarction, infection, degenerative disease or surgical lesion. It may be desirable to administer the NCDF-related peptides or NCDF protein by adsorption onto a membrane, e.g. a silastic membrane, that could be implanted in the proximity of the damaged nerve. The present invention can also be used for example in hastening the recovery of patients suffering from diabetic neuropathies, e.g. mononeuropathy multiplex or impotence. In further embodiments of the invention, NCDF protein or peptide fragments or derivatives derived therefrom, can be used to treat congenital conditions or neurodegenerative disorders, including, but not limited to, Alzheimer's disease, ageing, peripheral neuropathies, Parkinson's disease, Huntington's c orea and diseases and disorders of motorneurons; in particular, the invention can be used to treat congenital or neurodegenerative disorders associated with cholinergic neuron dysfunction.

In a specific embodiment of the invention, it is envisioned that administration of NCDF protein, or peptide fragments or derivatives derived therefrom, can be used in conjunction with surgical implantation of tissue or other sustained release compositions in the treatment of Alzheimer's disease, amyotrophic lateral sclerosis and other motorneuron diseases (including, for example, Werdnig-Hoffman disease), and Parkinson's disease. NCDF^*may also be useful in the treatment of a variety of dementias as well as congenital learning disorders.

In further embodiments of the invention, NCDF protein, fragments or derivatives can be used in conjunction with other cytokines to achieve a desired neurotrophic effect. For example, and not by way of limitation, according to the invention NCDF can be used together with NGF to achieve a stimulatory effect on growth and survival of neurons. It is envisioned that NCDF may function synergistically with other CNS- derived peptide factors yet to be fully characterized, in the growth, development, and survival of a wide array of neuronal subpopulations in the central and peripheral nervous system.

In still further embodiments of the invention, antibodies directed toward NCDF protein, or peptide fragments or derivatives thereof, can be administered to patients suffering from a variety of neurologic disorders and diseases and who are in need of such treatment. For example, patients who suffer from excessive production of NCDF may be in need of such treatment. Anti-NCDF antibodies can be used in prevention of ^'aberrant regeneration of sensory neurons (e.g. post-operatively) , or in the treatment of chronic pain syndromes.

5.6.3. PHARMACEUTICAL COMPOSITIONS The active compositions of the invention, which may comprise all or portions of the NCDF protein, peptide fragments or analogs or derivatives produced therefrom, or antibodies (or antibody fragments) directed toward NCDF protein, peptide fragments, or derivatives, or a combination of NCDF and a second agent (such as NGF) may be administered in any sterile biocompatible pharmaceutical carrier, including, but not limited to, saline, buffered saline, dextrose, and water.

The amount of NCDF protein, peptide fragment, derivative, or antibody which will be effective in the treatment of a particular disorder or condition will depend on the nature of the disorder or condition, and can be determined by standard clinical techniques. Where possible, it is desirable to determine the dose- response curve first in vitro, e.g. in the NCDF bioassay systems described supra. and then in useful animal model systems prior to testing in humans.

Methods of introduction include but are not limited to intradermal, intramuscular, intraperitoneal, intravenous, subcutaneous, oral, and intranasal. In addition, it may be desirable to introduce the pharmaceutical compositions of the invention into the central nervous system by any suitable route, including intraventricular and intrathecal injection; intraventricular injection may be facilitated by an intraventricular catheter, for example, attached to a reservoir, such as an Ommaya reservoir.

Further, it may be desirable to administer the pharmaceutical compositions of the invention locally to the area in need of treatment; this may be achieved by, for example, and not by way of limitation, local infusion during surgery, by injection, by means of a catheter, or by means of an implant, said implant being of a porous, non-porous, or gelatinous material, including membranes, such as sialastic membranes, or fibers.

The invention also provides for pharmaceutical compositions comprising NCDF proteins, peptide fragments, analogs, or derivatives administered via liposomes, microparticles, or microcapsules. In various embodiments of the invention, it may be useful to use such compositions to achieve sustained release of NCDF and NCDF-related products.

6. CHARACTERIZATION OF A TARGET-DERIVED

NEURONAL CHOLINERGIC DIFFERENTIATION FACTOR

The sympathetic innervation of rat sweat glands undergoes a target-induced switch from a noradrenergic to a cholinergic and peptidergic phenotype during development. Treatment of cultured sympathetic neurons with sweat gland extracts mimics many of the charges seen in vivo. Extracts induce choline acetyltransferase activity and vasoactive intestinal peptide expression in the neurons in a dose- dependent fashion while reducing catecholaminergic properties and neuropeptide Y. The cholinergic differentiation activity appears in developing glands of postnatal day 5 rats and is maintained in adult glands. It is a heat-labile, trypsin-sensitive, acidic protein that does not bind to heparin-agarose. Immunoprecipitation experiments with an antiserum directed against an N-terminal peptide of a cholinergic differentiation factor (CDF/LIF) from heart cells suggest that the sweat gland differentiation factor is not CDF/LIF. The sweat gland activity is a likely candidate for mediating the target-directed change in sympathetic neurotransmitter function observed in vivo. To identify the target factor responsible for the adrenergic to cholinergic switch observed in the developing sweat gland innervation and to explore the possible in vivo role of the candidate cholinergic factors identified in cell culture, we assayed the effects of sweat gland extracts on the transmitter properties of cultured sympathetic neurons. We found that sweat glands contain a soluble factor(s) with the appropriate spectrum of activities: it reduces the expression of catecholamines and tyrosine hydroxylase and induces the expression of choline acetyltransferase and VIP. This activity is present when the phenotype of the sweat gland innervation is changing. Our 5 initial characterization of the sweat gland-derived choline acetyltransferase-inducing activity permits a comparison with the cholinergic factors previously identified in cell culture.

10 6.1. RESULTS

6.1.1. SWEAT GLAND EXTRACTS INDUCE CHOLINE ACETYL- TRANSFERASE ACTIVITY IN SYMPATHETIC NEURONS

To examine the effects of soluble factors present in sweat glands on neurotransmitter status, extracts from the footpads of adult rats were added to

15 sympathetic neuron cultures at a concentration of 250 μg of extract protein per ml. Neurons in sister wells were grown either in medium without added tissue extract or in medium containing an equal protein

__ concentration of liver, hairy skin, or parotid gland extracts prepared in the same manner as the sweat gland extracts. The addition of sweat gland extract caused a 15-fold induction in choline acetyltransferase activity compared with^"'neurons grown in medium alone or with

₂₅ extracts of liver, hairy skin or parotid gland (Figure la) .

One potential source of the cholinergic- inducing activity in the sweat gland extracts is the possible presence of CNTF in the peripheral nerve

30 plexus of the footpad tissue. Comparison of the cholinergic-inducing activity in sciatic nerve extracts and sweat gland extracts (Figure lb) , however, indicated that the level of induction per mg of extract protein was similar despite the fact that the

35 peripheral nerve plexus constitutes only a small proportion of the gland tissue. This observation and the finding that extracts of hairy skin did not cause choline acetyltransferase induction in cultured sympathetic neurons even though the hairy skin contains a plexus of sympathetic and sensory nerve fibers and endings comparable to that in sweat gland-containing skin make it unlikely that the ability of sweat gland extracts to induce cholinergic function is due to CNTF potentially derived from Schwann cells (Stockli et al., 1989, Nature 342, 920-923).

TABLE I

The Cholinergic-Inducing Effect of Sweat Gland Extracts Is Independent of Serum

Choline Acetyltransferase Activity

Medium SG Extract

Defined medium 8.02 ± 1.64 109.83 ±

5.57

L15-C0₂ + serum 18. 05 ± 5.57 126. 06 ±

5.58

Sympathetic neurons were cultured in L15-C0₂ either lacking serum or containing 5% rat serum with 300 μg/ml sweat gland extract. Cells were harvested 7 days after the addition of extracts and aliquots were tested for choline acetyltransferase activity by the method of Fonnum (1969, Biochem. J. 115, 465-472). Samples were run in triplicate. Data are expressed as pmol of choline acetyltransferase per min per well ± SEM.

The-;.induction of choline acetyltransferase activity by the sweat gland extract was due to a direct effect on sympathetic neurons. Since the neurons were grown in the continuous presence of 10 μM cytosine arabinoside, nonneuronal cells were virtually absent; thus, it is unlikely that the sweat gland<^*extracts exerted their influence indirectly via nonneuronal cells. In addition, when sympathetic neuron cultures were maintained in serum-free medium that yielded cultures free of nonneuronal cells, cholinergic induction was seen following treatment with sweat gland extracts (Table I) . This observation also makes it very unlikely that the sweat gland extract potentiated the cholinergic-inducing effects of the rat serum present in normal growth medium (Wolinsky et al., 1985, J. Neurosci. 5, 1497-1508; Wolinsky and Patterson, 1983, J. Neurosci. 3, 1495-1500).

Sweat gland extracts were tested for their ability to support the survival of cultured sympathetic neurons. Table II shows that neurons plated in medium lacking nerve growth factor (NGF) but containing sweat gland extracts at a dose of 1 mg/ml did not survive more than 3 days in culture. Furthermore, there was no significant difference in neuron number in cultures grown with or without sweat gland extract even at extract doses as high as 1 mg/ml in the presence of 50 ng/ml of NGF. Since the levels of choline acetyltransferase activity and acetylcholine synthesis are initially very low in dissociated sympathetic neuron cultures: (Johnson et al., 1976, Nature 262, 308-310; Johnson, 1980, J. Cell Biol. 84, 630-691; Patterson and Chun, 1977, Dev. Biol. 60, 473-481) and there was no significant change in cell number with a 50 to 100-fold induction of choline acetyltransferase activity, it is extremely unlikely that the induction of choline acetyltransferase observed in the presence of sweat glan ^"extract is due to the selective survival of preexisting cholinergic neurons.

TABLE II

Effect of NGF and Sweat Gland Extracts on Survival of

Sympathetic Neurons

Cell Number

-NGF -NGF+ Ext +NGF +NGF +

Ext Day 2 3027 ± 35 4734 ± 164 5027 ± 231 5324 ± 186

Day 5 10 + 12 12 ± 13 4624 ± 112 4867 ± 128

Sympathetic neurons were cultured in 56 well plates in L15-C0₂ with NGF (50 ng/ml) for 2 days. On the second day, the culture medium was replaced with medium containing no NGF (-NGF) , no NGF but with 1 mg/ml sweat gland extract

(-NGF + Ext) , 50 ng/ml NGF (+ NGF) , or 50 ng/ml NGF plus 1 mg/ml sweat gland extract (+NGF + Ext) . Cells were counted after an additional 3 days in culture. Samples were run in triplicate. Data are expressed as the number of cells surviving per well ± SEM.

6.1.2. CHOLINE ACETYLTRANSFERASE

INDUCTION IS DOSE DEPENDENT

Serial dilutions of sweat gland extracts were added to cultured sympathetic neurons, and the wells were assayed for: choline acetyltransferase activity 7 days later. Induction was seen with doses as low as 10 μg/ml and increased with the addition of increasing amounts of extract to the maximum dose tested (Figure 2a) . When extract concentrations much higher than 1 mg/ml were used, some toxicity was evident in the cultures: neuron number was reduced, and cell bodies were smaller in size. Toxicity may be due to high concentrations of the cholinergic-inducing activity in the extract, since similar effects of high doses of other cholinergic-inducing factors have been described (Fukada, 1985, Proc. Natl. Acad. Sci. USA 82, 8795- 8799; Saadat et al., 1989, J. Cell Biol. 108, 1007- 1016) . Alternatively, it may be due to other compounds in the preparation.

The time course of induction was also determined. Elevated choline acetyltransferase activity was detected as early as the second day in culture and continued to increase through day 14, the last time point assayed (Figure 2b) . This time course of cholinergic induction in sympathetic neuron cultures is similar to that reported for heart and muscle cell conditioned medium factors, presumably CDF/LIF (Patterson and Chun, 1977, Dev. Biol. 60, 473-481; Raynaud et al., 1987, Dev. Biol. 121, 548-558), and for CNTF (Saadat et al., 1989, J. Cell Biol. 108, 1807- 1816) . In contrast, increased choline acetyltransferase activity is seen significantly sooner following treatment of sympathetic neurons with MANS (Adler et al., 1989, Proc. Natl. Acad. Sci. USA 86, 1080-1083) or treatment of spinal cord cultures with a soluble factor isolated from skeletal muscle (McManaman et al., 1988, J. Biol. Chem. 263, 5890-5897).

6.1.3. SWEAT GLAND EXTRACTS CAUSE A REDUCTION IN THE EXPRESSION OF NORADRENERGIC PROPERTIES Not only does choline acetyltransferase activity appear during the normal development of the sweat gland innervation, but there is also a concomitant reduction in tyrosine hydroxylase immunoreactivity and catecholamine histofluorescence (Landis and Keefe, 1983, Dev. Biol. 98, 349-372; Landis et al, 1988, Dev. Biol. 126, 129-140). If the sweat gland extracts contained a factor(s) that played a role in altering neurotransmitter phenotype, one would predict that it would decrease the expression of noradrenergic properties in cultured sympathetic neurons. To assay the effect of sweat gland extracts on tyrosine hydroxylase levels, equal protein aliquots of neurons grown with and without extract were subjected to SDS-PAGE (sodium dodecyl sulfate- polyacryla ide gel electrophoresis) , blotted onto nitrocellulose, and probed with a monoclonal antibody to tyrosine hydroxylase (Rohrer et al., 1986, J. Neurosci. 6, 2616-2624; the kind gift of Dr. A. Acheson, University of Edmonton) . A single band was evident at 62 kd, the expected molecular mass (Lamouroux et al., 1979, Proc. Natl. Acad. Sci. USA 79, 3881-3885) . Visual inspection of the immunoblots suggested that the amount of immunoreactivity was significantly reduced in cultures grown with sweat gland extracts (Figure 3) . When the color intensity was read with a laser densitometer, cultures grown with 100 μg/ml sweat gland extract exhibited a 2.5-fold reduction in the level of tyrosine hydroxylase- detected. In contrast, levels of immunoreactivity for a cell surface adhesion molecule, LI (Rathjen and Schachner, 1984, EMBO. J. 3, 1-10), revealed with a polyclonal antiserum (the kind gift of Dr. U. Rutishauser, Case Western Reserve University) were not reduced when assayed in a similar manner.

To determine whether the change in the level of tyrosine hydroxylase was associated with a corresponding change in the level of catecholamines, the catecholamine content was determined in cultures of sympathetic neurons grown with and without sweat gland extract. The total catecholamine content of wells incubated with^•sweat gland extracts was reduced compared with that of control cultures (Table III) .

Sympathetic neurons were grown with sweat gland extracts (100 μg/ml, 250 μg/ml and 1 mg/ml) . Seven days after the addition of extracts, the cultures were harvested and assayed for catecholamine content by high-performance liquid chromatography. Samples were run in triplicate. Data are expressed as mean pmol of catecholamines per dish + SEM. The figures in brackets are the mean fold choline acetyltransferse induction assayed in sister wells by the method of Fonnum (1969, Biochem. J. , 115, 465-472).

An inverse correlation was observed between choline acetyltransferase activity and catecholamine content; as the induction of choline acetyltransferase increased, the catecholamine content decreased. This relationship has been observed previously in studies with heart and skeletal muscle cell conditioned medium (Patterson and Chun. 1977, Dev. Biol. 56, 263-280; Raynaud et al., 1987, Dev. Biol. 121, 548-558). 6.1.4. SWEAT GLAND EXTRACTS ALTER NEUROPEPTIDE EXPRESSION

Changes in neuropeptide expression are observed in the developing sweat gland innervation. VIP immunoreactivity is initially absent but becomes detectable by PIO; the immunoreactivity increases in extent and intensity with subsequent development. Since the sympathetic innervation of foot pads transplanted to the thorax acquires VIP immunoreactivity, sweat glands are able to induce VIP expression in addition to choline acetyltransferase activity (Schotzinger and Landis, 1988, Nature 335, 637-639; unpublished data). We therefore assayed cultures of sympathetic neurons treated with sweat gland extract by radioimmunoassay to determine whether extracts increased VIP levels. Sympathetic neurons grown in control medium contain relatively little VIP immunoreactivity. Sweat gland extracts significantly increased VIP (Figure 4a) ; a dose of 100 μg/ml causes an induction of 80 pg per well of VIP, a more than 4- fold increase over the levels present in control cultures. The induction of VIP expression increased with increasing concentrations of sweat gland extracts (Figure 4a) . -,.r

The effect of sweat gland extract on NPY content was examined because previous studies have shown that while many noradrenergic sympathetic neurons contain NPY immunoreactivity, cholinergic sympathetic neurons, including those that innervate sweat glands, do not (Landis, et al., 1988, Dev. Biol. 126, 129-140; Lindh et al., 1989, Cell Tissue Res. 296, 259-273). The content of NPY-like immunoreactivity was high in control cultures, as observed in previous studies (Marek and Mains, 1989, J. Neurochem. 52, 1807-1816; Nawa & Sah, 1990, Neuron 4, 279-287) . Growth in the presence of sweat gland extract led to a reduction in NPY content (Figure 4b) . This reduction is in marked contrast to the elevation of VIP content and indicated that sweat gland extracts regulate the levels of the two peptides differentially. The decreased expression of NPY in sympathetic neurons grown with sweat gland extract is consistent with results of a previous study of target effect on peptide expression in vivo: following transplantation of the superior cervical ganglion from newborn rats to the anterior chamber of the eye, NPY-ΪR was absent when the ganglion was

10 cotransplanted with sweat glands, but present when the ganglion was cotransplanted with the pineal gland (Stevens and Landis, 1990, Dev. Biol. 737, 109-124).

6.1.5. AGE DEPENDENCE OF CHOLINE ACETYL-

15 TRANSFERASE-INDUCING ACTIVITY IN EXTRACTS OF SWEAT GLANDS

The change in neurotransmitter properties in the developing sweat gland innervation occurs postnatally and is esentially complete by P21. To

20 determine the earliest age at which detectable cholinergic-inducing activity was present in developing glands, extracts were prepared from footpads of animals ranging in age from 2 to 21 days and were assayed for their ability to induce choline acetyltransferase

25 activity (Figure 5) . Increased levels of choline acetyltransferase were detected in cultures treated with extracts from P5 glands, and choline acetyltransferase-inducing activity was present at all subsequent ages. Less than a 2-fold difference was

30 evident in the specific cholinergic-inducing activity present in footpads of P5 and adult animals when the inducing activity was expressed as the amount of choline acetyltransferase activity detected per mg of extract protein. The amount of protein extracted from

^ 20 footpads, however, varied almost 15-fold from the youngest to the oldest animals. Thus, the absolute amount of choline acetyltransferase-inducing activity increased approximately 30-fold during development. These results indicate that cholinergic differentiation activity is present in developing glands when the properties of the innervation change. In addition, extracts from sweat glands of P9, P14 and P21 rats were found to increase the expression of VIP and decrease the levels of tyrosine hydroxylase, as well as increase choline acetyltransferase (data not shown) .

6.1.6. INITIAL CHARACTERIZATION OF THE

FACTOR(S) RESPONSIBLE FOR CHOLINE ACETYLTRANSFERASE INDUCTION

Preliminary characterization of the choline acetyltransferase-inducing activity is summarized in Table IV.

TABLE IV

Physicochemical Characterization of Cholinergic Differentiation Activity

% Activity Retained

Thermal stability

-20°C/-70°C storage 90

Freeze-thawing 50 Boiling (100^βC for 5 min) 0

Protease treatment Trypsin 0

Trypsin + inhibitor 27

Heparin-agarose chromatography

Flow-through 55 eluate 8

Centricon retention

10 kd cutoff 95 50 kd cutoff 50

DEAE chromatography

Flow-through 2

0.25 M elute 50

Aliquots of sweat gland extracts (100 μg/ml) were incubated as described below or in Experimental Procedures. To examine the effects of protease treatment, aliquots were incubated for 1 hr with trypsin (1 mg/ml) or with trypsin and trypsin inhibitor (3 mg/ml) . For Centricon separation, samples were spun in a SS34 rotor until the retentate volume was 25 ml. The retentate was diluted to 1 ml and spun again. After three such spins, the flow-through was collected and concentrated. Choline acetyltransferase activity was determined in cultures 7 days after the addition of treated extracts. A value of 100% represents activity evident in cultures exposed to untreated extract and 0% represents activity in cultures grown without the addition of extract.

The activity was heat and trypsin labile and retained by a Centricon filter with a 10 kd cutoff, indicating that the activity is a protein. The activity was only partially retained by a Centricon filter with a 30 kd cutoff, suggesting that a low molecular mass protein is responsible for the induction of choline acetyltransferase. The cholinergic-inducing activity was relatively stable; little activity was lost with storage at -20°C and on repeated freeze-thawing. Since none of the activity bound to a heparin-agarose column, the sweat gland cholinergic factor does not appear to be a heparin binding protein like the 50 kd soluble cholinergic factor from brain (Kessler et al., 1986, Proc. Natl. Acad. Sci. USA 83, 3528-3532). Almost all choline acetyltransferase-inducing activity and 35% of the protein were recovered in the 0.25 M eluate from a DEAE column, indicating that the differentiation activity is an acidic protein(s) and that this can be used as an initial purification step. The 0.25 DEAE eluate not only induced choline acetyltransferase activity, but also increased levels of VIP and reduced levels of tyrosine hydroxylase (data not shown). Thus, the several effects of the sweat gland extract on neurotransmitter properties of cultured sympathetic neurons are not readily separated.

6.1.7. FAILURE TO IMMUNOPRECIPITATE BIOLOGICAL ACTIVITY WITH ANTIBODIES TO CDF/LIF

One candidate for the cholinergic-inducing activity in sweat gland extracts is CDF/LIF, since it has many of the same effects on the neurotransmitter properties of cultured sympathetic neurons. Antisera generated against a synthetic peptide whose sequence corresponds to the N-terminal peptide sequence of CDF/LIF can immunoprecipitate the cholinergic-inducing activity from a partially purified DEAE fraction from heart cell conditioned medium (Yamamori et al., 1989, Science 246, 1412-1416; Rao et al., 1990, Dev. Biol. 139, 65-74). When the DEAE fraction of the sweat gland extract was treated with affinity-purified antibodies to CDF/LIF, there was no detectable decrease in the ability of the extract to induce choline acetyltransferase activity (Figure 6a) . Since the DEAE fraction is relatively crude, it was possible that inhibitory proteins or proteases, present in the sweat gland extract, were responsible for the inability to immunoprecipitate activity. In parallel experiments, however, the same antibodies added to the DEAE fraction of sweat gland extracts were able to precipitate iodinated recombinant CDF/LIF (the kind gift of Dr. T. Yamamori, California Institute of Technology) as the appropriate 20 kd band (Figure 6b) . Thus, the principal cholinergic-inducing activity present in the sweat gland extracts is not likely to be CDF/LIF.

6.1.8. NCDF IS DISTINCT FROM CNTF Another candidate for the cholinergic inducing activity in sweat glands was ciliary neurotrophic factor (CNTF) . To examine this possibility, equal amounts of cholinergic inducing activity from sciatic nerve extracts and sweat gland extracts were loaded on an SDS-PAGE gel, electrophoresed and probed with a polyclonal antiserum generated against recombinant rat CNTF (a kind gift of Dr. Mark Furth, Regeneron Pharmaceuticals) . The antiserum recognized recombinant CNTF (Fig. 7a) and a 24 kd band present in the sciatic nerve extracts (Fig. 7b) . Binding was completely blocked by preincubating the antibody with lOμ recombinant CNTF. ' No specific band was evident in the lanes containing either hairy skin extracts or sweat gland extracts even though the lanes containing the sweat gland and sciatic nerve extracts contained the same amount of cholinergic inducing activity for cultured sympathetic neurons (see also Fig. lb) . Loading ten-fold more cholinergic inducing activity and overstaining the blots failed to reveal any specific binding in lanes containing sweat gland extracts. Thus, the cholinergic inducing activity present in sweat gland extracts does not appear to be identical to CNTF. 5

6.1.9. NORTHERNS AND IN SITU HYBRIDIZATION WITH AN OLIGONUCLEOTIDE PROBE AGAINST RAT CNTF

To determine if message for CNTF could be detected in sweat glands and to identify the cells that

¹⁰ may be producing the CNTF/CNTF-like molecule, we prepared RNA from the adult sweat gland and probed the Northern blots for message with a probe against rat CNTF (Fig. 8). A band of 1.3 kb, the expected size of CNTF message, was detected in sciatic nerves. In

'^■*•* contrast, no specific binding was detected in lanes containing RNA from liver or sweat glands.

The same probe was also used to probe sections of sciatic nerve and sweat gland by in situ hybridization, as a more sensitive assay of the cells 0 type that may be making CNTF/CNTF-like molecule.

Schwann cells in the sciatic nerve showed a specific hybridization signal with the antisense probe as compared to the sense control (Fig. 9) . No specific signal however, could be detected in sweat gland tissue 5 (Fig. 10) .

6.1.10. ANIONIC EXCHANGE CHROMATOGRAPHY

The 0.25 M DEAE eluate containeci the ChAT and VIP inducing, and tyrosine hydroxylase reducing, 0 activity. Almost all activity and 35% of the protein was recovered in the 0.25 M eluate of a DEAE column, suggesting that this can be used as an initial purification step (Fig. 11) . 5 6.1.11. ISOELECTRIC FOCUSING The DEAE eluate was chromatographed on a MONO P column and 0.5 ml fractions were collected and assayed for cholinergic activity. Fig. 12 shows that the ChAT activity was eluted at between pH 4.8 to 5.2 with a peak of activity at pH 5.0, indicating that the pi of the active protein was in this range. This is similar to the value reported for CNTF purified from sciatic nerve extracts and differs from LIF which is a strongly basic protein.

6.1.12. SIZE FRACTIONATION To determine the molecular weight of the cholinergic inducing activity, partially purified fractions were chromatographed on a sizing column. Fractions were tested for activity on sympathetic neuron cultures. Almost all the activity eluted in a peak between 16 kd and 32 kd protein markers, indicating that NCDF is a low molecular weight protein.

To more accurately determine the molecular weight of NDCF, fractions of the sweat gland extract purified on a chromatofocussing column (Fig. 12) were run on a SDS PAGE gel, and bands of the appropriate molecular weight were cut out and the proteins eluted using an electroeluter. Aliquots of the extracted proteins were tested for activity (Fig. 13) . The activity had a molecular weight between 22-26 kd (Fig. 13) .

The same fraction which had cholinergic inducing activity was also tested for its ability to modulate levels of tyrosine hydroxylase. The same SDS PAGE eluted fraction had tyrosine hydroxylase reducing activity. 6.2. DISCUSSION

We prepared low salt extracts of sweat gland tissue and tested the ability of these extracts to modify the neurotransmitter phenotype of cultured sympathetic neurons. Extracts of sweat glands but not of liver, hairy skin, or parotid gland increase levels of choline acetyltransferase activity and of VIP-like immunoreactivity in a dose-dependent fashion. As the levels of choline acetyltransferase activity increase in the cultured neurons, there is a concomitant 10 decrease in the catecholamine content and tyrosine hydroxylase. Thus, extracts of soluble protein from sweat glands cause many of the changes in cultured sympathetic neurons that characterize the developing

,_ sweat gland innervation in vivo and that are induced by the glands in cross-innervation experiments.

The ability to alter neurotransmitter properties is present in sweat gland extracts of animals between P5 and P21, when the fibers innervating ₀ the sweat glands are changing from noradrenergic to cholinergic (Landis and Keefe, 1983, Dev. Biol. 98, 349-372; Leblanc and Landis, 1986, J. Neurosci. 6, 260- 265; Landis et al., 1988, Dev. Biol. 126, 129-140). Extracts from P5 animals increase choline acetyltransferase activity, and when extracts of glands from animals between P9 and P21 are tested, they alter all three transmitter properties examined: they increase choline aceyltransferase and VIP-like immunoreactivity and reduce tyrosine hydroxylase. 0 Furthermore, since elevated levels of choline acetyltransferase activity are detectable after 2 days of treatment in culture, the extract is able to induce changes with a time course consistent with in vivo studies. Establishing a more precise temporal 5 correlation is difficult, since the changes observed in the neurotransmitter properties of the terminal plexus in the sweat glands in situ presumably reflect retrograde transport of a target-derived signal, altered expression of transmitter synthetic enzymes and peptides, and anterograde transport of these molecules to the terminals.

Sweat glands of adult as well as developing animals contain NCDF activity.

Although comparison of the levels of biological activity observed in vitro with those present in vivo is difficult, it is of interest to estimate whether the glands contain enough cholinergic- inducing activity to mediate the switch. Retrograde tracing studies suggest that at least 200 neurons innervate the six palmar pads (Siegel and Landis, unpublished data) . Our extraction procedure yields about 10 mg of soluble protein per gram of footpad tissue from 21-day-old animals, or approximately 80 μg per single pad. Since choline acetyltransferase activity is induced at concentrations as low as 10 μg/ml in cultures containing several thousand neurons, it appears that they do contain sufficient quantities of cholinergic-inducing activity. The concentration of cholinergic-inducing activity present in sweat gland extracts is greater than that in spinal cord extracts (Wong and Kessler, 1987, Proc. Natl. Acad. Sci. USA 84, 8726-8729; Adler et al. 1989, Proc. Natl. Acad. Sci. USA 86, 1059-1083) and at least as high as that in sciatic nerve extracts (Sendtner et al. , 1989, Soc. Neurosci. Abs. 15, 710; Rao et al., 1990, Dev. Biol. 139, 65-74).

Since CDF/LIF (Fukada, 1985, Proc. Natl. Acad. Sci. USA 82, 8795-8799; Yamamori et al. , 1989, Science 246, 1412-1416; Nawa and Patterson, 1990, Neuron 4, 269-277) and CNTF (Sendtner et al., 1989, Soc. Neurosci. Abs. 15, 170; Ernsberger et al., 1989, Neuron 2, 1275-1284) cause the induction of cholingeric function, the reduction of catecholaminergic function, and an increase in VIP expression in sympathetic neuron cultures, it is clear that a single molecule can affect changes in all these properties. Two observations from the present studies are consistent with the notion that one molecule in the extracts is responsible. Extracts prepared from animals of different ages influence the several properties assayed, and more importantly, the several effects were not resolved into distinct activities in the preliminary characterization that we have performed. Thus, a single molecule seems likely; however, the possibility that several factors are involved cannot be formally excluded.

It is of interest to compare the properties of the activity in sweat gland extract with the several factors that have been previously described to induce cholinergic function in cultured sympathetic neurons. Since the cholinergic-inducing activity present in the sweat gland extracts is easily extracted in low salt solutions and no detectable activity is associated with membranes (unpμblished data) , it is not likely to be related to the membrane-associated factors that induce choline acetyltransferase (Wong and Kessler, 1987,

Proc. Natl. Acad. Sci. USA 84, 8726-8729; Adler et al.,

1989, Proc. Natl. Acad. Sci. USA 86, 1059-1083) and reduce levels of tyrosine hydroxylase (Rap et al.,

1990, Dev. Biol. 139, 65-74; Lee et al., 1990, Exp. Neur. 108, 109-113) in cultured sympathetic neurons.

In addition, there is a difference in the time course of induction of choline acetyltransferase activity; cultures treated with sweat gland extracts exhibit a small increase of activity after 2 days, whereas cultures treated with a membrane-associated cholinergic-inducing activity show high levels of activity in the same time period (Adler et al., 1989,

Proc. Natl. Acad. Sci. U.S.A. 86, 1080-1083). Two soluble factors, CDF/LIF and CNTF, are similar in their effects on sympathetic neurons; they increase choline acetyltransferase and VIP expression and reduce tyrosine hydroxylase and catecholamine content (Fukada,

1985, Proc. Natl. Acad. Sci. USA 82, 8795-8799;

Yamamori et al., 1989, Science 246, 1412-1416; Sendtner et al., 1989, Soc. Neurosci. Abs. 15, 710; Ernsberger et al., 1989, Neuron 2, 1275-1284; Nawa and Patterson,

1990, Neuron 4, 269-277). In addition, like sweat gland extract, both CDF/LIF (Nawa and Patterson, 1990,

Neuron 4, 269-277) and extracts of sciatic nerve containing CNTF (unpublished data) decrease NPY expression.

CDF/LIF was an attractive candidate; it has a consensus signal sequence, it is glycosylated, and it is secreted by heart cells (Patterson and Chun, 1977, Dev. Biol 56, 263-280; Yamamori et al., 1989, Science

246, 1412-1416) . CDF/LIF, however, does not bind to a

DEAE column (Fukada, 1985, Proc. Natl. Acad. Sci. USA

82, 8795-8799), unlike the cholinergic-inducing activity in sweat glands. Furthermore, affinity- purified antibodies raised against the N-terminal region of CDF/LIF can immunoprecipitate the cholinergic-inducing activity from the DEAE or Sephadex fractions of heart cell conditioned medium (Yamamori et al., 1989, Science 246, 1412-1416; Rao et al., 1990, Dev. Biol. 139, 65-74), but these antibodies do not immunoprecipitate the cholinergic-inducing activity from sweat gland extracts. Thus, it is unlikely that the major cholinergic factor in the extract is identical to CDF/LIF. CNTF was another likely candidate. It is an acidic protein and can be eluted from a DEAE column by 0.25 M NaCl (Manthorpe et al., 1980, Neurochem. 34, 69- 75; Manthorpe et al., 1986, Brain Res. 367, 282-286; Barbin et al., 1984, J. Neurochem 43, 1468-1478), much like the cholinergic-inducing activity present in the sweat gland extracts. However, Northern blot analysis of adult glabrous skin containing sweat glands failed to demonstrate detectable message for CNTF, whereas abundant message was present in sciatic nerves (Stockli et al., 1989, Nature 342, 920-923; Sendtner et al., 1989, Soc. Neurosci. Abs. 15, 170) even though the two tissues contain similar levels of cholinergic-inducing activity. In addition, immunoblot experiments with a polyclonal antiserum generated against and recognizing recombinant CNTF failed to reveal any CNTF-like immunoreactivity in sweat gland extracts. Furthermore, Northern blot and in situ hybridization assays failed to reveal any specific hybridization in samples from sweat glands with a rat CNTF probe. These data, together with the fact that CNTF appears to be a cytosolic protein (Stockli et al., 1989, Nature 342, 920-923; Lin et al., 1989, Science 246, 1023-1025); while a candidate sweat gland differentiation molecule is likely to be secreted to exert an effect on the innervation, suggest that CNTF is an unlikely candidate for the sweat gland-derived cholinergic factor.

In summary, we have shown that a cholinergic sympathetic target tissue, sweat glands, contains cholinergic differentiating activity that mimics in culture the effects of the target on sympathetic neurons in vivo. Our preliminary purification and analysis suggest that the cholinergic-inducing activity present in the sweat gland extracts is not identical to either CDF/LIF or CNTF and that it is a novel factor. Thus, the cholinergic-inducing activity present in sweat gland extracts represents an excellent candidate for mediating the target-induced phenotypic changes in the cholinergic sympathetic neurons that innervate sweat glands.

6.3. EXPERIMENTAL PROCEDURES 6.3.1. MATERIALS Cell culture reagents were obtained from GIBCO (Grand Island, NY) and culture plates were from Corning (Corning, NY) . The Centricon filters were purchased from Amicon (Danvers, MA) . [³H]-acetyl-CoA and Bolton Hunter reagent were purchased from New England Nuclear (Wilmington, DE) . Dispase was obtained from Boehringer Mannheim (Indianapolis, IN) , and collagenase was from Worthington Biochemicals (Freehold, NJ) . VIP radioimmunoassay kits were obtained from Incstar (Stillwater, MN) , and NPY radioimmunoassay reagents were from Amersham. NGF (the kind gift of Dr. K. Neat, Case Western Reserve

University) was prepared from male mouse submaxillary glands as described by Bocchini and Angeletti (1969, Proc. Natl. Acad. Sci. USA 64, 787-794). Pierce protein assay^" it was obtained from Pierce (Rockford, IL) . ITS Promix was from Collaborative Research

(Bedford, MA) and reagents for SDA-PAGE were from Bio- Rad (Richmond, CA) . Avidin-conjugated alkaline phosphatase was obtained from Cappel (Westchester, PA) , and goat anti-mouse and anti-rabbit secondary antibodies were from Jackson Immunologicals (Westgrove, PA) . Other chemicals were purchased from Sigma^' (St. Louis, MO) .

6.3.2. CELL CULTURE Cultures of rat sympathetic neurons were prepared as described by Hawrot and Patterson (1979, Meth. Enzymthol. 58, 574-583). Neurons from the superior cervical ganglia of newborn rats were 5 dissociated enzymatically with Dispase (5 mg/ml) and collagenase (1 mg/ml and plated in 96-well plates coated sequentially with polylysine (0.1 mg/ml) and laminin (10 μg/15ml) About 2000-3000 neurons were plated per well except where indicated. The neurons 0 were grown in Leibovitz's L15-C0₂ medium with NGF (100 ng/ml) , 100 U of penicillin, 100 μg of streptomycin, 10 μM cytosine arabinoside, and 5% rat serum, and the medium changed every third or fourth day. In some experiments, cells were grown without rat serum in L15- 5 C0₂ supplemented with transferrin, selenium, bovine serum albumin, insulin and fatty acids.

The tissue extracts were diluted in growth medium, sterilized by passage through a 0.2 μm filter and added to the neurons from the third day of culture 0 on. Neurons were harvested for assay between days 9 and 14 of culture.

6.3.3. TISSUE EXTRACTS To prepare sweat gland extracts, footpads 5 were extracted from rats of various postnatal ages and weighed. Tissue from 20 animals was generally processed at one time. The weight of footpads from 20 rats varied from 0.5 to 5 grams, depending upon the age of the animal. The tissue was homogenized for 5 sec in 0 ιo vol of 10 mM phosphate butter (pH 7.0) with a

Polytron. The extract was then centrifuged at 100,000 x g for 1 hr. The supernatant was collected, filtered through a 0.2 μm filter, and concentrated using a Centricon filter with a 10 kd cutoff. The protein ^ concentration was determined with a Pierce protein assay kit. Extracts of liver, sciatic nerves and parotid glands were prepared in a similar manner. To prepare hairy skin extracts, the skin over the thoracic region was shaved and dissected free from the underlying panniculosis carnosus muscle and weighed. The skin was then cut into smaller pieces before being homogenized and processed as described above.

6.3.4. ASSAYS 10 The induction of cholinergic function was determined by assaying choline acetyltransferase activity in homogenates essentially according to the method of Fonnum (1969, Biochem. J. 115, 465-472). To increase the sensitivity of the assay, an incubation 15 period of 1 hr was used. All activity was inhibitable by 500 μM napthylvinyl pyridine, a specific inhibitor of choline acetyltransferase activity. Protein concentration was assayed by the method of Lowry using bovine serum albumin as a standard.

20 Catecholamine content was assayed by high performance liquid chromatography (Rittenhouse et al., 1988, Neurosci. 25, 207-215) on a 5 μm pore reverse- phase C-18 column (Altex Ultrasphere-IP; Beckman, Berkeley, CA) Using a colorometric detector (5100A,

25 ESA, Bedford, MA) . Three electrodes were set in series at +0.36, +0.03, and -0.38 V relative to a reference electrode. Standards at known dilutions (5 pmol) were run at the same time to estimate the concentration. The total catecholamine content of a well was obtained

30 by summing the levels of norepinephrine, dopamine, and DOPAC, a metabolite present in each extract. Neither epinephrine nor DOPA was detected.

The amount of tyrosine hydroxylase present in the cultured neurons was determined by semiquantitative

^ analysis of immunoblots. Cell cultures were homogenized in sample buffer (50 mM Tris (pH 6.8) with 2% SDS, 10% glycerol, 0.004% bromophenol blue, and 3% /3-mercaptoethanol) , aliquots of the extract were run on a 10% SDS-PAGE gel, and the proteins were blotted onto nitrocellulose. The nitrocellulose blots were blocked in blocking buffer (5% defatted milk in Tris-buffered saline [pH 7.2]) and then incubated with a monoclonal antibody against tyrosine hydroxylase (the kind gift of Dr. Ann Acheson, University of Alberta, Edmonton) overnight. The blots were then sequentially incubated with a biotinylated secondary antibody and avidin conjugated to alkaline phosphatase. The reaction product was developed with Nitro Blue Tetrazolium and 5-bromo-4-chloro-3-indoyl phosphate in 10 mM bicarbonate butter (pH 9.5). After optimal color development, the reaction was stopped by rinsing in distilled water. The blots were allowed to dry, and the color intensity was read on a scanning laser densitometer (Shimadzu) . Comparisons were made between samples run in parallel lanes and treated identically. Neuropeptide levels were determined by radioimmunoassay. Cultures were rinsed once with PBS and then homogenized in 100 μl of 2 M acetic acid. After boiling^"for 5 min, samples were centrifuged for 1 min in an Eppendorf microfuge. The supernatants were dried under vacuum and stored at -70°C for subsequent assays. VIP was assayed using a kit obtained from INCSTAR with primary antibodies previously demonstrated to show minimal cross-reactivity with other peptides. TO assay NPY by radioimmunoassay, antibodies, standards, and labeled tracer were obtained from

A ersham and peptide content was determined by the delayed tracer method. Since the antibody shows only

64% cross-reactivity with rat NPY, standards were also run with rat NPY (Peninsula Laboratories) and sample values were read from the standard curve.

6.3.5. DEAE CHROMATOGRAPHY Aliquots of the soluble extract were diluted

5-fold with 10 mM phosphate buffer (pH 7.0) and applied to 0.9 X 10 cm DEAE column (Whatman/Bioprobe) at a flow rate of 10 ml/hr. The column was washed with an equal volume of phosphate buffer. The wash and flow-through were pooled and concentrated using a Centricon filter with a 10 kd cutoff. The bound protein was eluted with 10 ml of 0.25 M NaCl and concentrated in a similar manner.

6.3.6. HEPARIN A CHROMATOGRAPHY

Aliquots of the soluble protein were diluted with buffer (10 mM phosphate, 150 mM NaCl (pH 7.2) and applied to a heparin-agarose column. Bound protein was eluted with 5 M NaCl. The efficacy of column binding was tested using ¹²⁵I-labeled basic fibroblast growth factor (a kind gift of Dr. J. Unnerstall, Case Western Reserve University) .

6.3;^'" . IMMUNOPRECIPITATION EXPERIMENTS WITH CDF/LIF ANTIBODIES

For immunoprecipitation experiments in which the biological activity of the factors was tested, aliquots of tissue extract sufficient for a cell culture assay were added to buffer (PBS [pH 7.3] with 2% bovine serum albumin, 0.2% Triton X-100, and 0.02% PEG 6000) . Affinity-purified antibodies raised*against a synthetic peptide corresponding to the N-terminal region of CDF (Rao et al., 1990, Dev. Biol. 139, 65-74) were added to each vial to a final concentration of 10 μM. After an overnight incubation, the antigen- antibody complex was adsorbed to 10 μl of protein A- Sepharose for an additional 2 hr at room temperature. The bound complexes were separated by centrifugation, and the supernatant was diluted into L15-C0₂ medium and used for cell culture assays. Two controls were performed to ensure that the loss of activity consequent to absorption was due to a specific effect of the antibody. Aliquots of extract were incubated without the antibody and treated as described above, and other aliquots were treated with antibody that had been previously absorbed with 50 μM synthetic peptide originally used as antigen.

In other experiments, CDF/LIF (a kind gift of Dr. Yamamori, California Institute of Technology) was iodinated with Bolton Hunter reagent as described previously (Rao et al., 1990, Dev. Biol. 139, 65-74). About 20,000 cpm were added to buffer or an equal volume of the DEAE fraction of the sweat gland extract and immunoprecipitated with the N-terminal antibody as described above. The counts that were immunoprecipitated were extracted and analyzed by SDS- PAGE.

6.3.8. WESTERN BLOT EXPERIMENTS WITH CNTF ANTISERUM Aliquots of extracts (60 μg/lane) were run on a 15% SDS PAGE minigel (Biorad) and the proteins were blotted onto nitrocellulose. The nitrocellulose blots were blocked in blocking buffer (5% defatted milk in Tris buffered saline, pH 7.2) and then incubated for two hours with a polyclonal antiserum raised against recombinant rat CNTF (1:1000 dilution; the kind_fgift of

Dr. Donna Morrissey, Regeneron Pharmaceuticals) or with the antibody preincubated with 10 μM CNTF (Regeneron

Pharmaceuticals) . The blots were then sequentially incubated with a biotinylated secondary antibody for an hour and then with avidin-conjugated alkaline phosphatase for 30 min. The bound enzyme was detected with Nitro Blue Tetrazolium and 5-bromo-4-chloro-3 indoyl phosphate in 10 mM bicarbonate buffer, pH 9.5. After optimal color development, the reaction was 5 stopped by rinsing in distilled water.

6.3.9. NORTHERN BLOTS AND IN SITU HYBRIDIZATION

For Northern blot hybridzation, total RNA was prepared from liver, sweat gland and sciatic nerve 10 using the single step guanidinium-isothioσyanate method (Chomczynski and Sacchi, 1987, Analytical Biochem. 162, 156-159) . 30 μg of total RNA was loaded per lane and transferred to a Genescreen nylon membrane. Blots were probed with a 45 base pair oligonucleotide probe 15 (region 99-144) against rat CNTF radiolabelled with ³²P. Blots were sequentially washed and then examined by autoradiography.

The same probe used in the Northern blot experiments (except that it was labelled with {α 20³⁵S}dATP) was used for the in-situ hybridization experiments as described (Siegel, 1989, Methods in Neuroscience, ed. Conn P.M., Academic Press 1, 136- 150) . In brief, fresh frozen sections of tissue were fixed in 4% formaldehyde for ten minutes, and rinsed 5 twice in PBS slides, air dried and processed for hybridization. Hybridization was performed at 42°C for fifteen hours in a humidified container using 10⁷ cpm/ml of probe. After washing the sections were dipped in Kodak NTB-3 emulsion and exposed for 6 weeks. Sections ⁰ were developed, fixed and counterstained with ethidium bromide.

Various modifications of the invention in addition to those shown and described herein will 5 become apparent to those skilled in the art from the foregoing description and accompanying drawings. Such modifications are intended to fall within the scope of the appended claims.

Various references are cited herein, the disclosures of which are incorporated by reference herein in their entireties.

Claims

WHAT IS CLAIMED IS:

1. A substantially purified protein characterized by the following properties: (a) capable of being isolated from extracts of mammalian sweat glands;

(b) heat and trypsin lability;

(c) lack of substantial binding to a heparin agarose affinity column;

(d) an isoelectric point in the range of approximately 4.8 to 5.2;

(e) a non-membrane cellular localization; (f) ability to increase the expression of choline acetyltransferase by sympathetic neurons in vitro; and (g) an approximate molecular weight in the range of 16-32 kilodaltons.

2. The protein of claim 1 which is further characterized by the ability to reduce the expression of tyrosine hydroxylase by sympathetic neurons in vitro.

3. The protein of claim 1 which is further characterized by the ability to increase the expression of vasoactive intestinal peptide by sympathetic neurons in vitro.

4. The protein of claim 1 which is further characterized by the ability to reduce the expression of total catecholamines by sympathetic neurons in vitro.

5. The protein of claim 2 which is further characterized by the ability to reduce the expression of total catecholamines by sympathetic neurons in vitro.

6. The protein of claim 1 in which the sweat glands are those of an adult mammal.

7. The protein of claim 1 in which the sweat glands are those of a human.

8. The protein of claim 1 in which the sweat glands are those of a rat.

9. A method of inducing cholinergic activity in neurons comprising exposing the neurons to an effective amount of the protein of claim 1.

10. A method of inducing cholinergic activity in neurons comprising exposing the neurons to an effective amount of the protein of claim 2.

11. A method of inducing cholinergic activity in neurons comprising exposing the neurons to an effective amount of the protein of claim 3.

12. A method of inducing cholinergic activity in neurons comprising exposing the neurons to an effective amount of the protein of claim 4.

13. A method of inducing cholinergic activity in neurons comprising exposing the neurons to an effective amount of the protein of claim 5.

14. A method of inducing cholinergic activity in neurons comprising exposing the neurons to an effective amount of the protein of claim 6.

15. A method of inducing cholinergic activity in neurons comprising exposing the neurons to an effective amount of the protein of claim 7.

16. A method of inducing cholinergic activity in neurons comprising exposing the neurons to an effective amount of the protein of claim 8.

17. An analog, derivative or fragment of the protein of claim 1 which has the ability to reduce the expression of tyrosine hydroxylase by sympathetic neurons in vitro.

18. An analog, derivative or fragment of the protein of claim 2 which has the ability to reduce the expression of tyrosine hydroxylase by sympathetic neurons in vitro.

19. An analog, derivative or fragment of the protein of claim 3 which has the ability to reduce the expression of tyrosine hydroxylase by sympathetic neurons in vitro.

20. An analog, derivative or fragment of the protein of claim 4 which has the ability to reduce the expression of tyrosine hydroxylase by sympathetic neurons in vitro.

21. An analog, derivative or fragment of the protein of claim 1 which has the ability to increase the expression of choline acetyltransferase by sympathetic neurons jLn vitro.

22. An analog, derivative or fragment of the protein of claim 2 which has the ability to increase the expression of choline acetyltransferase by sympathetic neurons in vitro.

23. An analog, derivative or fragment of the protein of claim 3 which has the ability to increase the expression of choline acetyltransferase by sympathetic neurons in vitro.

24. An analog, derivative or fragment of the protein of claim 4 which has the ability to increase the expression of choline acetyltransferase by sympathetic neurons in vitro.

25. An analog, derivative or fragment of the protein of claim 1 which has the ability to increase the expression of vasoactive intestinal peptide by sympathetic neurons in vitro.

26. An analog, derivative or fragment of the protein of claim 2 which has the ability to increase the expression of vasoactive intestinal peptide by sympathetic neurons in vitro.

27. An analog, derivative or fragment of the protein of claim 3 which has the ability to increase the expression of vasoactive intestinal peptide by sympathetic neurons in vitro.

28. An analog, derivative or fragment of the protein of claim 4 which has the ability to increase the expression of vasoactive intestinal peptide by sympathetic neurons in vitro.

29. An analog, derivative or fragment of the protein of claim 1 which has the ability to reduce the expression of total catecholamines by sympathetic neurons in vitro.

30. An analog, derivative or fragment of the protein of claim 2 which has the ability to reduce the expression of total catecholamines by sympathetic neurons in vitro.

31. An analog, derivative or fragment of the protein of claim 3 which has the ability to reduce the expression of total catecholamines by sympathetic neurons in vitro.

32. An analog, derivative or fragment of the protein of claim 4 which has the ability to reduce the expression of total catecholamines by sympathetic neurons in vitro.

33. An antibody capable of specifically binding the protein of claim 1.

34. An antibody capable of specifically binding the protein of claim 2.

35. An antibody capable of specifically binding the protein of claim 3.

36. An antibody capable of specifically binding the protein of claim 4.

37. An antibody capable of specifically binding the protein of claim 5.

38. An antibody capable of specifically binding the protein of claim 6.

39. An antibody capable of specifically binding the protein of claim 7.