US20090111744A1

US20090111744A1 - Human Growth Gene and Short Stature Gene Region

Info

Publication number: US20090111744A1
Application number: US11/748,769
Authority: US
Inventors: Gudrun Rappold-Hoerbrand; Ercole Rao
Original assignee: Individual
Current assignee: Individual
Priority date: 1996-10-01
Filing date: 2007-05-15
Publication date: 2009-04-30
Also published as: CN1232499A; NZ334970A; ATE230026T1; CA2647169A1; EP0946721B1; EP0946721A1; WO1998014568A1; HUP9904175A3; CA2267097A1; DE69718052D1; DE69718052T2; DK0946721T3; SI0946721T1; IL129015A0; AU744188B2; CZ297640B6; AU744188C; HUP9904175A2; US7252974B2; AU4625297A

Abstract

Subject of the present invention is an isolated human nucleic acid molecule encoding polypeptides containing a homeobox domain of sixty amino acids having the amino acid sequence of SEQ ID NO: 1 and having regulating activity on human growth. Three novel genes residing within the about 500 kb short stature critical region on the X and Y chromosome were identified. At least one of these genes is responsible for the short stature phenotype. The cDNA corresponding to this gene may be used in diagnostic tools, and to further characterize the molecular basis for the short stature-phenotype. In addition, the identification of the gene product of the gene provides new means and methods for the development of superior therapies for short stature.

Description

This application is a divisional of U.S. Ser. No. 10/158,160 filed May 31, 2002, which is a continuation of U.S. Ser. No. 09/147,699, filed Jun. 24, 1999, now abandoned, which is a 371 of PCT/EP97/05355, filed on Sep. 29, 1997, which claims the benefit of U.S. provisional application 60/027,633, filed on Oct. 1, 1996.
The present invention relates to the isolation, identification and characterization of newly identified human genes responsible for disorders relating to human growth, especially for short stature or Turner syndrome, as well as the diagnosis and therapy of such disorders.
The isolated genomic DNA or fragments thereof can be used for pharmaceutical purposes or as diagnostic tools or reagents for identification or characterization of the genetic defect involved in such disorders. Subject of the present invention are further human growth proteins (transcription factors A, B and C) which are expressed after transcription of said DNA into RNA or mRNA and which can be used in the therapeutic treatment of disorders related to mutations in said genes. The invention further relates to appropriate cDNA sequences which can be used for the preparation of recombinant proteins suitable for the treatment of such disorders. Subject of the invention are further plasmid vectors for the expression of the DNA of these genes and appropriate cells containing such DNAs. It is a further subject of the present invention to provide means and methods for the genetic treatment of such disorders in the area of molecular medicine using an expression plasmid prepared by incorporating the DNA of this invention downstream from an expression promotor which effects expression in a mammalian host cell.
Growth is one of the fundamental aspects in the development of an organism, regulated by a highly organised and complex system. Height is a multifactorial trait, influenced by both environmental and genetic factors. Developmental malformations concerning body height are common phenomena among humans of all races. With an incidence of 3 in 100, growth retardation resulting in short stature account for the large majority of inborn deficiencies seen in humans.
With an incidence of 1:2500 life-born phenotypic females, Turner syndrome is a common chromosomal disorder (Rosenfeld et al., 1996). It has been estimated that 1-2% of all human conceptions are 45,X and that as many as 99% of such fetuses do not come to term (Hall and Gilchrist, 1990; Robins, 1990). Significant clinical variability exists in the phenotype of persons with Turner syndrome (or Ullrich-Turner syndrome) (Ullrich, 1930; Turner, 1938). Short stature, however, is a consistent finding and together with gonadal dysgenesis considered as the lead symptoms of this disorder. Turner syndrome is a true multifactorial disorder. Both the embryonic lethality, the short stature, gonadal dysgenesis and the characteristic somatic features are thought to be due to monosomy of genes common to the X and Y chromosomes. The diploid dosis of those X-Y homologous genes are suggested to be requested for normal human development. Turner genes (or anti-Turner genes) are expected to be expressed in females from both the active and inactive X chromosomes or Y chromosome to ensure correct dosage of gene product. Haploinsufficiency (deficiency due to only one active copy), consequently would be the suggested genetic mechanism underlying the disease.
A variety of mechanisms underlying short stature have been elucidated so far. Growth hormone and growth hormone receptor deficiencies as well as skeletal disorders have been described as causes for the short stature phenotype (Martial et al., 1979; Phillips et al., 1981; Leung et al., 1987; Goddard et al., 1995). Recently, mutations in three human fibroblast growth factor receptor-encoding genes (FGFR 1-3) were identified as the cause of various skeletal disorders, including the most common form of dwarfism, achondroplasia (Shiang et al., 1994; Rousseau et al., 1994; Muenke and Schell, 1995). A well-known and frequent (1:2500 females) chromosomal disorder, Turner Syndrome (45,X), is also consistently associated with short stature. Taken together, however, all these different known causes account for only a small fraction of all short patients, leaving the vast majority of short stature cases unexplained to date.
The sex chromosomes X and Y are believed to harbor genes influencing height (Ogata and Matsuo, 1993). This could be deduced from genotype-phenotype correlations in patients with sex chromosome abnormalities. Cytogenetic studies have provided evidence that terminal deletions of the short arms of either the X or the Y chromosome consistently lead to short stature in the respective individuals (Zuffardi et al., 1982; Curry et al., 1984). More than 20 chromosomal rearrangements associated with terminal deletions of chromosome Xp and Yp have been reported that localize the gene(s) responsible for short stature to the pseudoautosomal region (PAR1) (Ballabio et al., 1989, Schaefer et al., 1993). This localisation has been narrowed down to the most distal 700 kb of DNA of the PAR1 region, with DXYS 15 as the flanking marker (Ogata et al., 1992; 1995).
Mammalian growth regulation is organized as a complex system. It is conceivable that multiple growth promoting genes (proteins) interact with one another in a highly organized way. One of those genes controlling height has tentatively been mapped to the pseudoautosomal region PAR1 (Ballabio et al., 1989), a region known to be freely exchanged between the X and Y chromosomes (for a review see Rappold, 1993). The entire PAR1 region is approximately 2,700 kb.
The critical region for short stature has been defined with deletion patients. Short stature is the consequence when an entire 700 kb region is deleted or when a specific gene within this critical region is present in haploid state, is interrupted or mutated (as is the case with idiotypic short stature or Turner sydrome). The frequency of Turner's syndrome is 1 in 2500 females worldwide; the frequency of this kind of idiopathic short stature can be estimated to be 1 in 4.000-5.000 persons. Turner females and some short stature individuals usually receive an unspecific treatment with growth hormone (GH) for many years to over a decade although it is well known that they have normal GH levels and GH deficiency is not the problem. The treatment of such patients is very expensive (estimated costs approximately 30.000 USD p.a.). Therefore, the problem existed to provide a method and means for distinguishing short stature patients on the one side who have a genetic defect in the respective gene and on the other side patients who do not have any genetic defect in this gene. Patients with a genetic defect in the respective gene—either a complete gene deletion (as in Turner syndrome) or a point mutation (as in idiopathic short stature)—should be susceptible for an alternative treatment without human GH, which now can be devised.
Genotype/phenotype correlations have supported the existence of a growth gene in the proximal part of Yq and in the distal part of Yp. Short stature is also consistently found in individuals with terminal deletions of Xp. Recently, an extensive search for male and female patients with partial monosomies of the pseudoautosomal region has been undertaken. On the basis of genotype-phenotype correlations, a minimal common region of deletion of 700 kb DNA adjacent to the telomere was determined (Ogata et al., 1992; Ogata et al., 1995). The region of interest was shown to lie between genetic markers DXYS20 (3cosPP) and DXYS15 (113D) and all candidate genes for growth control from within the PAR1 region (e.g., the hemopoietic growth factor receptor a; CSF2RA) (Gough et al., 1990) were excluded based on their physical location (Rappold et al., 1992). That is, the genes were within the 700 kb deletion region of the 2.700 kb PAR1 region.
Deletions of the pseudoautosomal region (PAR1) of the sex chromosomes were recently discovered in individuals with short stature and subsequently a minimal common deletion region of 700 kb within PAR1 was defined. Southern blot analysis on DNA of patients AK and SS using different pseudoautosomal markers has identified an Xp terminal deletion of about 700 kb distal to DXYS 15 (113D) (Ogata et al, 1992; Ogata et al, 1995).
The gene region corresponding to short stature has been identified as a region of approximately 500 kb, preferably approximately 170 kb in the PAR1 region of the X and Y chromosomes. Three genes in this region have been identified as candidates for the short stature gene. These genes were designated SHOX (also referred to as SHOX93 or HOX93), (SHOX=short stature homeobox-containing gene), pET92 and SHOT (SHOX-like homeobox gene on chromosome three). The gene SHOX which has two separate splicing sites resulting in two variations (SHOX a and b) is of particular importance. In preliminary investigations, essential parts of the nucleotide sequence of the short stature gene could be analysed (SEQ ID No. 8). Respective exons or parts thereof could be predicted and identified (e.g. exon I [G310]; exon II [ET93]; exon IV [G108]; pET92). The obtained sequence information could then be used for designing appropriate primers or nucleotide probes which hybridize to parts of the SHOX gene or fragments thereof. By conventional methods, the SHOX gene can then be isolated. By further analysis of the DNA sequence of the genes responsible for short stature, the nucleotide sequence of exons I-V could be refined (v. FIG. 1-3). The gene SHOX contains a homeobox sequence (SEQ ID NO: I) of approximately 180 bp (v. FIG. 2 and FIG. 3), starting from the nucleotide coding for amino acid position 117 (Q) to the nucleotide coding for amino acid position 176 (E), i.e. from CAG (440) to GAG (619). The homeobox sequence is identified as the homeobox-pET93 (SHOX) sequence and two point mutations have been found in individuals with short stature in a German (A1) and a Japanese patient by screening up to date 250 individuals with idiopahtic short stature. Both point mutations were found at the identical position and leading to a protein truncation at amino acid position 195, suggesting that there may exist a hot spot of mutation. Due to the fact that both mutations found, which lead to a protein truncation, are at the identical position, it is possible that a putative hot spot of recombination exisits with exon 4 (G108). Exon specific primers can therefore be used as indicated below, e.g. GCA CAG CCA ACC ACC TAG (for) or TGG AAA GGC ATC ATC CGT AAG (rev).
The above-mentioned novel homeobox-containing gene, SHOX, which is located within the 170 kb interval, is alternatively spliced generating two proteins with diverse function. Mutation analysis and DNA sequencing were used to demonstrate that short stature can be caused by mutations in SHOX.
The identification and cloning of the short stature critical region according to the present invention was performed as follows: Extensive physical mapping studies on 15 individuals with partial monosomy in the pseudoautosomal region (PAR1) were performed. By correlating the height of those individuals with their deletion breakpoints a short stature (SS) critical region of approximately 700 kb was defined. This region was subsequently cloned as an overlapping cosmid contig using yeast artificial chromosomes (YACs) from PAR 1 (Ried et al., 1996) and by cosmid walking. To search for candidate genes for SS within this interval, a variety of techniques were applied to an approximately 600 kb region between the distal end of cosmid 56G10 and the proximal end of 51D11. Using cDNA selection, exon trapping, and CpG island cloning, the two novel genes were identified.
The position of the short stature critical interval could be refined to a smaller interval of 170 kb of DNA by characterizing three further specific individuals (GA, AT and RY), who were consistently short. To precisely localize the rearrangement breakpoints of those individuals, fluorescence in situ hybridization (FISH) on metaphase chromosomes was carried out using cosmids from the contig. Patient GA, with a terminal deletion and normal height, defined the distal boundary of the critical region (with the breakpoint on cosmid 110E3), and patient AT, with an X chromosome inversion and normal height, the proximal boundary (with the breakpoint on cosmid 34F5). The Y-chromosomal breakpoint of patient RY, with a terminal deletion and short stature, was also found to be contained on cosmid 34F5, suggesting that this region contains sequences predisposing to chromosome rearrangements.
The entire region, bounded by the Xp/Yp telomere, has been cloned as a set of overlapping cosmids. Fluorescence in situ hybridization (FISH) with cosmids from this region was used to study six patients with X chromosomal rearrangements, three with normal height and three with short stature. Genotype-phenotype correlations narrowed down the critical short stature interval to 270 kb of DNA or even less as 170 kb, containing the gene or genes with an important role in human growth. A minimal tiling path of six to eight cosmids bridging this interval is now available for interphase and metaphase FISH providing a valuable tool for diagnostic investigations on patients with idiopathic short stature.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a gene map of the SHOX gene including five exons which are identified as follows: exon I: G310, exon II: ET93, exon III: ET45, exon IV: G108 and exons Va and Vb, whereby exons Va and Vb result from two different splicing sites of the SHOX gene. Exon II and III contain the homeobox sequence of 180 nucleotides.

FIGS. 2 and 3 are the nucleotide and predicted amino acid sequences of SHOXa and SHOXb:

- SHOX a: The predicted start of translation begins at nucleotide 92 with the first in-frame stop codon (TGA) at nucleotides 968-970, yielding an open reading frame of 876 bp that encodes a predicted protein of 292 amino acids (designated as transcription factor A or SHOXa protein, respectively). An in-frame, 5′stop codon at nucleotide 4, the start codon and the predicted termination stop codon are in bold. The homeobox is boxed (starting from amino acid position 117 (Q) to 176 (E), i.e. CAG thru GAG in the nucleotide sequence). The locations of introns are indicated with arrows. Two putative polyadenylation signals in the 3′untranslated region are underlined.
- SHOX b: An open reading frame of 876 bp exists from A in the first methionin at nucleotide 92 to the in-frame stop codon at nucleotide 767-769, yielding an open reading frame of 675 bp that encodes a predicted protein of 225 amino acids (transcription factor B or SHOXb protein, respectively). The locations of introns are indicated with arrows. Exons I-IV are identical with SHOXa, exon V is specific for SHOX b. A putative polyadenylation signal in the 3′ untranslated region is underlined.

FIG. 4 are the nucleotide (SEQ ID NO:43) and predicted amino acid (SEQ ID NO: 16) sequence of SHOT. The predicted start of translation begins at nucleotide 43 with the first in-frame stop codon (TGA) at nucleotides 613-615, yielding an open reading frame of 573 bp that encodes a predicted protein of 190 amino acids (designated as transcription factor C or SHOT protein, respectively). The homeobox is boxed (starting from amino acid position 11 (Q) to 70 (E), i.e. CAG thru GAG in the nucleotide sequence). The locations of introns are indicated with arrows. Two putative polyadenylation signals in the 3′untranslated region are underlined

FIG. 5 gives the exon/intron organization of the human SHOX gene and the respective positions in the nucleotide sequence (Intron/Exon sequences (SEQ ID NOS 44-49, respectively in order of appearance) and Exon/Intron sequences (SEQ ID NOS 50-55, respectively in order of appearance)).

BRIEF DESCRIPTION OF THE SEQ ID:

SEQ ID NO.1: translated amino acid sequence of the homeobox domain (180 bp)
SEQ ID NO.2: exon II (ET93) of the SHOX gene
SEQ ID NO. 3: exon I (G310) of the SHOX gene
SEQ ID NO. 4: exon III (ET45) of the SHOX gene
SEQ ID NO. 5: exon IV (G108) of the SHOX gene
SEQ ID NO. 6: exon Va of the SHOX gene
SEQ ID NO. 7: exon Vb of the SHOX gene
SEQ ID NO. 8: preliminary nucleotide sequence of the SHOX gene
SEQ ID NO.9: ET92 gene
SEQ ID NO.10: SHOXa sequence (see also FIG. 2)
SEQ ID NO.1: transcription factor A (see also FIG. 2)
SEQ ID NO. 12: SHOXb sequence (see also FIG. 3)
SEQ ID NO. 13: transcription factor B (see also FIG. 3)
SEQ ID NO. 14: SHOX gene
SEQ ID NO. 15: SHOT sequence
SEQ ID NO. 16: transcription factor C (see also FIG. 4)
Since the target gene leading to disorders in human growth (e.g. short stature region) was unknown prior to the present invention, the biological and clinical association of patients with this deletion could give insights to the function of this gene. In the present study, fluorescence in situ hybridization (FISH) was used to examine metaphase and interphase lymphocyte nuclei of six patients. The aim was to test all cosmids of the overlapping set for their utility as FISH probes and to determine the breakpoint regions in all four cases, thereby determining the minimal critical region for the short stature gene.
Duplication and deletion of genomic DNA can be technically assessed by carefully controlled quantitative PCR or dose estimation on Southern blots or by using RFLPs. However, a particularly reliable method for the accurate distinction between single and double dose of markers is FISH, the clinical application of is presently routine. Whereas in interphase FISH, the pure absence or presence of a molecular marker can be evaluated, FISH on metaphase chromosomes may provide a semi-quantitative measurement of inter-cosmid deletions. The present inventor has determined that deletions of about 10 kb (25% of signal reduction) can still be detected. This is of importance, as practically all disease genes on the human X chromosome have been associated with smaller and larger deletions in the range from a few kilobases to several megabases of DNA (Nelson et al., 1995).
Subject of the present invention are therefore DNA sequences or fragments thereof which are part of the genes responsible for human growth (or for short stature, respectively, in case of genetic defects in these genes). Three genes responsible for human growth were identified: SHOX, pET92 and SHOT. DNA sequences or fragments of these genes, as well as the respective full length DNA sequences of these genes can be transformed in an appropriate vector and transfected into cells. When such vectors are introduced into cells in an appropriate way as they are present in healthy humans, it is devisable to treat diseases involved with short stature, i.e. Turners syndrome, by modern means of gene therapy. For example, short stature can be treated by removing the respective mutated growth genes responsible for short stature. It is also possible to stimulate the respective genes which compensate the action of the genes responsible for short stature, i.e. by inserting DNA sequences before, after or within the growth/short stature genes in order to increase the expression of the healthy allels. By such modifications of the genes, the growth/short stature genes become activated or silent, respectively. This can be accomplished by inserting DNA sequences at appropriate sites within or adjacent to the gene, so that these inserted DNA sequences interfere with the growth/short stature genes and thereby activate or prevent their transcription. It is also devisable to insert a regulatory element (e.g. a promotor sequence) before said growth genes to stimulate the genes to become active. It is further devisable to stimulate the respective promotor sequence in order to overexpress—in the case of Turner syndrome—the healthy functional allele and to compensate for the missing allele. The modification of genes can be generally achieved by inserting exogenous DNA sequences into the growth gene/short stature gene via homologous recombination.
The DNA sequences according to the present invention can also be used for transformation of said sequences into animals, such as mammals, via an appropriate vector system. These transgenic animals can then be used for in vivo investigations for screening or identifying pharamceutical agents which are useful in the treatment of diseases involved with short stature. If the animals positively respond to the administration of a candidate compound or agent, such agent or compound or derivatives thereof would be devisable as pharmaceutical agents. By appropriate means, the DNA sequences of the present invention can also be used in genetic experiments aiming at finding methods in order to compensate for the loss of genes responsible for short stature (knock-out animals).
In a further object of this invention, the DNA sequences can also be used to be transformed into cells. These cells can be used for identifying pharmaceutical agents useful for the treatment of diseases involved with short stature, or for screening of such compounds or library of compounds. In an appropriate test system, variations in the phenotype or in the expression pattern of these cells can be determined, thereby allowing the identification of interesting candidate agents in the development of pharmaceutical drugs.
The DNA sequences of the present invention can also be used for the design of appropriate primers which hybridize with segments of the short stature genes or fragments thereof under stringent conditions. Appropriate primer sequences can be constructed which are useful in the diagnosis of people who have a genetic defect causing short stature. In this respect it is noteworthy that the two mutations found occur at the identical position, suggesting that a mutational hot spot exists.
In general, DNA sequences according to the present invention are understood to embrace also such DNA sequences which are degenerate to the specific sequences shown, based on the degeneracy of the genetic code, or which hybridize under stringent conditions with the specifically shown DNA sequences.
The present invention encompasses especially the following aspects:

- a) An isolated human nucleic acid molecule encoding polypeptides containing a homeobox domain of sixty amino acids having the amino acid sequence of SEQ ID NO: 1 and having regulating activity on human growth.
- b) An isolated DNA molecule comprising the nucleotide sequence essentially as indicated in FIG. 2, FIG. 3 or FIG. 4, and especially as shown in SEQ ID NO: 10, SEQ ID NO: 12 or SEQ ID NO: 15.
- c) DNA molecules capable of hybridizing to the DNA molecules of item b).
- d) DNA molecules of item c) above which are capable of hybridization with the DNA molecules of item 2. under a temperature of 60-70° C. and in the presence of a standard buffer solution.
- e) DNA molecules comprising a nucleotide sequence having a homology of seventy percent or higher with the nucleotide sequence of SEQ ID NO: 10, SEQ ID NO: 12 or SEQ ID NO: 15 and encoding a polypeptide having regulating activity on human growth.
- f) Human growth proteins having the amino acid sequence of SEQ ID NO: 11, 13 or 16 or a functional fragment thereof.
- g) Antibodies obtained from immunization of animals with human growth proteins of item f) or antigenic variants thereof.
- h) Pharmaceutical compositions comprising human growth proteins or functional fragments thereof for treating disorders caused by genetic mutations of the human growth gene.
- i) A method of screening for a substance effective for the treatment of disorders mentioned above under item h) comprising detecting messenger RNA hybridizing to any of the DNA molecules decribed in a)-e) so as to measure any enhancement in the expression levels of the DNA molecule in response to treatment of the host cell with that substance.
- j) An expression vector or plasmid containing any of the nucleic acid molecules described in a)-e) above which enables the DNA molecules to be expressed in mammalian cells.
- k) A method for the determination of the gene or genes responsible for short stature in a biological sample of body tissues or body fluids.

In the method k) above, preferably nucleotide amplification techniques, e.g. PCR, are used for detecting specific nucleotide sequences known to persons skilled in the art, and described, for example, by Mullis et al. 1986, Cold Spring Harbor Symposium Quant. Biol. 51, 263-273, and Saiki et al., 1988, Science 239, 487-491, which are incorporated herein by reference. The short stature nucleotide sequences to be determined are mainly those represented by sequences SEQ ID No. 2 to SEQ ID No. 7.
In principle, all oligonucleotide primers and probes for amplifying and detecting a genetic defect responsible for deminished human growth in a biological sample are suitable for amplifying a target short stature associated sequence. Especially, suitable exon specific primer pairs according to the invention are provided by table 1. Subsequently, a suitable detection, e.g. a radioactive or non-radioactive label is carried out.

TABLE 1

Exon	Sense primer	Antisense primer	Product (bp)	Ta (° C.)

5′-I (G310)	SP 1	ASP 1	194	58
3′-I (G310)	SP 2	ASP 2	295	58
II (ET93)	SP 3	ASP 3	262	76/72/68
III (ET45)	SP 4	ASP 4	120	65
IV (G108)	SP 5	ASP 5	154	62
Va (SHOXa)	SP 6	ASP 6	265	61

explanation of the abbreviations for the primers:

	SP1	ATTTCCAATGGAAAGGCGTAAATAAC

	SP2	ACGGCTTTTGTATCCAAGTCTTTTG

	SP3	GCCCTGTGCCCTCCGCTCCC

	SP4	GGCTCTTCACATCTCTCTCTGCTTC

	SP5	CCACACTGACACCTGCTCCCTTTG

	S5P6	CCCGCAGGTCCAGGCTCAGCTG

	ASP1	CGCCTCCGCCGTTACCGTCCTTG

	ASP2	CCCTGGAGCCGGCGCGCAAAG

	ASP3	CCCCGCCCCCGCCCCCGG

	ASP4	CTTCAGGTCCCCCCAGTCCCG

	ASP5	CTAGGGATCTTCAGAGGAAGAAAAAG

	ASP6	GCTGCGCGGCGGGTCAGAGCCCCAG

Also, a single stranded RNA can be used as target. Methods for reversed transcribing RNA into cDNA are also well known and described in Sambrook et al., Molecular Cloning: A Laboratory Manual, New York, Cold Spring Harbor Laboratory 1989. Alternatively, preferred methods for reversed transcription utilize thermostable DNA polymerases having RT activity.
Further, the technique described before can be used for selecting those person from a group of persons being of short stature characterized by a genetic defect and which allows as a consequence a more specific medical treatment.
In another subject of the present invention, the transcription factors A, B and C can be used as pharmaceutical agents. These transcription factors initiate a still unknown cascade of biological effects on a molecular level involved with human growth. These proteins or functional fragments thereof have a mitogenic effect on various cells. Especially, they have an osteogenic effect. They can be used in the treatment of bone diseases, such as e.g. osteoporosis, and especially all those diseases involved with disturbance in the bone calcium regulation.
As used herein, the term “isolated” refers to the original derivation of the DNA molecule by cloning. It is to be understood however, that this term is not intended to be so limiting and, in fact, the present invention relates to both naturally occurring and synthetically prepared seqences, as will be understood by the skilled person in the art.
The DNA molecules of this invention may be used in forms of gene therapy involving the use of an expression plasmid prepared by incorporating an appropriate DNA sequence of this invention downstream from an expression promotor that effects expression in a mammalian host cell. Suitable host cells are procaryotic or eucaryotic cells. Procaryotic host cells are, for example, E. coli, Bacillus subtilis, and the like. By transfecting host cells with replicons originating from species adaptable to the host, that is, plasmid vectors containing replication starting point and regulator sequences, these host cells can be transfected with the desired gene or cDNA. Such vectors are preferably those having a sequence that provides the transfected cells with a property (phenotype) by which they can be selected. For example, for E. coli hosts the strain E. coli K12 is typically used, and for the vector either pBR322 or pUC plasmids can be generally employed. Examples for suitable promoters for E. coli hosts are trp promotor, lac promotor or Ipp promotor. If desired, secretion of the expression product through the cell membrane can be effected by connecting a DNA sequence coding for a signal peptide sequence at the 5′ upstream side of the gene. Eucaryotic host cells include cells derived from vertebrates or yeast etc.. As a vertebrate host cell, COS cells can be used (Cell, 1981, 23: 175-182), or CHO cells. Preferably, promotors can be used which are positioned 5′ upstream of the gene to be expressed and having RNA splicing positions, polyadenylation and transcription termination seqences.
The transcription factors A, B and C of the present invention can be used to treat disorders caused by mutations in the human growth genes and can be used as growth promoting agents. Due to the polymorphism known in the case of eukaryotic genes, one or more amino acids may be substituted. Also, one or more amino acids in the polypeptides can be deleted or inserted at one or more sites in the amino acid sequence of the polypeptides of SEQ ID NO: 11, 13 or 16. Such polypeptides are generally referred to equivalent polypeptides as long as the underlying biological acitivity of the unmodified polypeptide remains essentially unchanged.
The present invention is illustrated by the following examples.

EXAMPLE 1

Patients

All six patients studied had de novo sex chromosome aberrations.
CC is a girl with a karyotype 45,X/46,X psu dic (X) (Xqter→Xp22.3::Xp22.3→Xqter). At the last examination at 6½ years of age, her height was 114 cm (25-50 the % percentile). Her mother's height was 155 cm, the father was not available for analysis. For details, see Henke et al., 1991.
GA is a girl with a karyotype 46,X der X (3pter→3p23::Xp22.3→Xqter). At the last examination at 17 years, normal stature (159 cm) was observed. Her mother's height is 160 cm and her father's height 182 cm. For details, see Kulharya et al, 1995.
SS is a girl with a karyotype 46,X rea (X) (Xqter→Xq26::Xp22.3→Xq26:). At 11 years her height remained below the 3rd percentile growth curve for Japanese girls; her predicted adult height (148.5 cm) was below her target height (163 cm) and target range (155 to 191 em). For details, see Ogata et alt, 1992.
AK is a girl with a karyotype 46,X rea (X) (Xqter→Xp22.3::Xp22.3→Xp21.3:). At 13 years her height remained below the 2nd percentile growth curve for Japanese girls; her predicted adult height (142.8 cm) was below her target height (155.5 cm) and target range (147.5-163.5 em). For details, see Ogata et alt, 1995.
RY: the karyotype of the ring Y patient is 46,X,r(Y)/46,Xdic r(Y)/45,X[95:3:2], as examined on 100 lymphocytes; at 16 years of age his final height was 148; the heights of his three brothers are all in the normal range with 170 cm (16 years, brother 1), 164 cm (14 years, brother 2) and 128 cm (9 years, brother 3), respectively. Growth retardation of this patient is so severe that it would also be compatible with an additional deletion of the GCY locus on Yq.
AT: boy with ataxia and inv(X); normal height of 116 cm at age 7, parents' heights are 156 cm and 190 cm, respectively.

Patients for Mutation Analysis:

250 individuals with idiopathic short stature were tested for mutations in SHOXa. The patients were selected on the following criteria: height for chronological age was below the 3rd centile of national height standards, minus 2 standard deviations (SDS); no causative disease was known, in particular: normal weight (length) for gestational age, normal body proportions, no chronic organic disorder, normal food intake, no psychiatric disorder, no skeletal dysplasia disorder, no thyroid or growth hormone deficiency.

Family A:

Cases 1 and 2 are short statured children of a German non-consanguineous family. The boy (case 1) was born at the 38th week of gestation by cesarian section. Birth weight was 2660 g, birth length 47 cm. He developed normally except for subnormal growth. On examination at the age of 6.4 years, he was proportionate small (106.8 cm, −2.6 SDS) and obese (22.7 kg), but otherwise normal. His bone age was not retarded (6 yrs) and bone dysplasia was excluded by X-ray analysis. IGF-I and IGFBP-3 levels as well as thyroid parameters in serum rendered GH or thyroid hormone deficiency unlikely. The girl (case 2) was born at term by cesarian section. Birth weight was 2920 g, birth length 47 cm. Her developmental milestones were normal, but by the age of 12 months poor growth was apparent (length: 67 cm, −3.0 SDS). At 4 years she was 89.6 cm of height (−3.6 SDS). No dysmorphic features or dysproportions were apparent. She was not obese (13 kg). Her bone age was 3.5 years and bone dysplasia was excluded. Hormone parameters were normal. It is interesting to note that both the girl and the boy grow on the 50 percentile growth curve for females with Turner syndrome. The mother is the smallest of the family and has a mild rhizomelic dysproportion (142.3 cm, −3.8 SDS). One of her two sisters (150 cm, −2.5 SDS) and the maternal grandmother (153 cm, −2.0 SDS) are all short without any dysproportion. One sister has normal stature (167 cm, +0.4 SDS). The father's height is 166 cm (−1.8 SDS) and the maternal grandfather' height is 165 cm (−1.9 SDS). The other patient was of Japanese origin and showed the identical mutation.

EXAMPLE 2

Identification of the Short Stature Gene

A. In situ Hybridization
a) Florescence in situ Hybridization (FISH)
Florescence in situ hybridization (FISH) using cosmids residing in the Xp/Yp pseudoautosomal region (PAR1) was carried out. FISH studies using cosmids 64/75cos (LLNLc 110H032), E22cos (2e2), F1/14cos (110A7), M1/70cos (110E3), P99F2cos (43C11), P99cos (LLNLc110P2410), B6cosb (1CRFc104H0425), F20cos (34F5), F21cos (ICRFc104G0411), F3cos2 (9E3), F3cos1 (11E6), P117cos (29B11), P6cos1 (ICRFc104P0117), P6cos2 (LLNLc110E0625) and E4cos (15G7) was carried out according to published methods (Lichter and Cremer, 1992). In short, one microgram of the respective cosmid clone was labeled with biotin and hybridized to human metaphase chromosomes under conditions that suppress signals from repetitive DNA sequences. Detection of the hybridization signal was via FITC-conjugated avidin. Images of FITC were taken by using a cooled charge coupled device camera system (Photometrics, Tucson, Ariz.).

b) Physical Mapping

Cosmids were derived from Lawrence Livermore National Laboratory X- and Y-chromosome libraries and the Imperial Cancer Research Fund London (now Max Planck Institute for Molecular Genetics Berlin) X chromosome library. Using cosmids distal to DXYS15, namely E4cos, P6cos2, P6cos1, P1117cos and F3cos1 one can determine that two copies are still present of E4cos, P6cos2, P6cos1 and one copy of P117cos and F3cos1. Breakpoints of both patients AK and SS map on cosmid P6cos1, with a maximum physical distance of 10 kb from each other. It was concluded that the abnormal X chromosomes of AK and SS have deleted about 630 kb of DNA. Further cosmids were derived from the ICRF X chromosome specific cosmid library (ICRFc104), the Lawrence Livermore X chromosome specific cosmid library (LLNLc110) and the Y chromosome specific library (LLCO3′M′), as well as from a self-made cosmid library covering the entire genome. Cosmids were identified by hybridisation with all known probes mapping to this region and by using entire YACs as probes. To verify overlaps, end probes from several cosmids were used in cases in which overlaps could not be proven using known probes.

c) Southern Blot Hybridisation

Southern blot analysis using different pseudoautosomal markers has provided evidence that the breakpoint on the X chromosome of patient CC resides between DXYS20 (3cosPP) and DXYS60 (U7A) (Henke et al, 1991). In order to confirm this finding and to refine the breakpoint location, cosmids 64/75cos, E22cos, F1/14cos, M1/70cos, F2cos, P99F2cos and P99cos were used as FISH probes. The breakpoint location on the abnormal X of patient CC between cosmids 64/75cos (one copy) and F1/14cos (two copies) on the E22PAC could be determined. Patient CC with normal stature consequently has lost approximately 260-290 kb of DNA.
Southern blot hybridisations were carried out at high stringency conditions in Church buffer (0.5 M NaPi pH 7.2, 7% SDS, 1 mM EDTA) at 65° C. and washed in 40 mM NaPi, 1% SDS at 65° C.

d) FISH Analysis

Biotinylated cosmid DNA (insert size 32-45 kb) or cosmid fragments (10-16 kb) were hybridised to metaphase chromosomes from stimulated lymphocytes of patients under conditions as described previously (Lichter and Cremer, 1992). The hybridised probe was detected via avidin-conjugated FITC.

e) PCR Amplification

All PCRs were performed in 50 μl volumes containing 100 pg-200 ng template, 20 pmol of each primer, 200 μM dNTP's (Pharmacia), 1.5 mM MgCl₂, 75 mMTris/HCl pH9, 20 mM (NH₄)₂SO₄0.01% (w/v) Tween20 and 2 U of Goldstar DNA Polymerase (Eurogentec). Thermal cycling was carried out in a Thermocycler GeneE (Techne).

f) Exon Amplification

Four cosmid pools consisting of each four to five clones from the cosmid contigs were used for exon amplification experiments. The cosmids in each cosmid pool were partially digested with Sau3A. Gel purified fractions in the size range of 4-10 kb were cloned in the BamHI digested pSPL3B vector (Burn et al, 1995) and used for the exon amplification experiments as previously described (Church et al., 1994).

g) Genomic Sequencing

Sonificated fragments of the two cosmids LLOYNCO3′M′15D10 and LLOYNCO3′M′34F5 were subcloned separately into M13mp18 vectors. From each cosmid library at least 1000 plaques were picked, M13 DNA prepared and sequenced using dye-terminators, ThermoSequenase (Amersham) and universal M13-primer (MWG-BioTech). The gels were run on ABI-377 sequencers and data were assembled and edited with the GAP4 program (Staden).
Of all six patients, GA had the least well characterized chromosomal breakpoint. The most distal markers previously tested for their presence or absence on the X were DXS1060 and DXS996, which map approximately 6 Mb from the telomere (Nelson et al., 1995). Several cosmids containing different gene sequences from within PAR1 (MIC2, ANT3, CSF2RA, and XE7) were tested and all were present on the translocation chromosome. Cosmids from within the short stature critical region e.g., chromosome, thereby placing the translocation breakpoint on cosmid M1/70cos. A quantitative comparison of the signal intensities of M1/70cos between the normal and the rearranged X indicates that approximately 70% of this cosmid is deleted.

TABLE 2

Table 2: This table summarizes the FISH data
for the 16 cosmids tested on four patients.

	CC	GA	AK	SS

64/75cos	−	−
E22cos	−	−
F1/14cos	+	−
M1/70cos	+	(+)
F2cos	+	+
P99F2cos	+	+
P99cos	+	+
B6cos		+
F20cos
F21cos
F3cos2
F3cos1			−	−
P117cos			−	−
P6cos1			+	+
P6cos2			+	+
E4cos			+	+

[−] one copy; indicates that the respective cosmid was deleted on the rearranged X, but present on the normal X chromosome
[+] two copies; indicates that the respective cosmid is present on the rearranged and on the normal X chromosome
[(+)] breakpoint region; indicates that the breakpoint occurs within the cosmid as shown by FISH

In summary, the molecular analysis on six patients with X chromosomal rearrangements using florescence-labeled cosmid probes and in situ hybridization indicates that the short stature critical region can be narrowed down to a 270 kb interval, bounded by the breakpoint of patient GA from its centromere distal side and by patients AK and SS on its centromere proximal side.
Genotype-phenotype correlations may be informative and have been chosen to delineate the short stature critical interval on the human X and Y chromosome. In the present study FISH analysis was used to study metaphase spreads and interphase nuclei of lymphocytes from patients carrying deletions and translocations on the X chromosome and breakpoints within Xp22.3. These breakpoints appear to be clustered in two of the four patients (AK and SS) presumably due to the presence of sequences predisposing to chromosome rearrangements. One additional patient Ring Y has been found with an interruption in the 270 kb critical region, thereby reducing the critical interval to a 170 kb region.
By correlating the height of all six individuals with their deletion breakpoint, an interval of 170 kb was mapped to within the pseudoautosomal region, presence or absence of which has a significant effect on stature. This interval is bounded by the X chromosomal breakpoint of patient GA at 340 kb from the telomere (Xptel) distally and by the breakpoints of patients AT and RY at 510/520 kb Xptel proximally. This assignment constitutes a considerable reduction of the critical interval to almost one fourth of its previous size (Ogata et al., 1992; Ogata et al., 1995). A small set of six to eight cosmids are now available for FISH experiments to test for the prevalence and significance of this genomic locus on a large series of patients with idiopathic short stature.

B. Identification of the Candidate Short Stature Gene

To search for transcription units within the smallest 170 kb critical region, exon trapping and cDNA selection on six cosmids (110E3, F2cos, 43C11, P2410, 15D10, 34F5) was carried out. Three different positive clones (ET93, ET45 and G108) were isolated by exon trapping, all of which mapped back to cosmid 34F5. Previous studies using cDNA selection protocols and an excess of 25 different cDNA libraries had proven unsuccessful, suggesting that genes in this interval are expressed at very low abundancy.
To find out whether any gene in this interval was missed, the nucleotide sequence of about 140 kb from this region of the PAR1 was determined, using the random M13 method and dye terminator chemistry. The cosmids for sequence analysis were chosen to minimally overlap with each other and to collectively span the critical interval. DNA sequence analysis and subsequent protein prediction by the “X Grail” program, version 1.3c as well as by the exon-trapping program FEXHB were carried out and confirmed all 3 previously cloned exons. No protein-coding genes other than the previously isolated one could be detected.

C. Isolation of the Short Stature Candidate Gene SHOX

Assuming that all three exon clones ET93, ET45 and G108 are part of the same gene, they were used collectively as probes to screen 14 different cDNA libraries from 12 different fetal (lung, liver, brain 1 and 2) and adult tissues (ovary, placenta 1 and 2, fibroblast, skeletal muscle, bone marrow, brain, brain stem, hypothalamus, pituitary). Not a single clone among approximately 14 million plated clones was detected. To isolate the full-length transcript, 3′ and 5′RACE were carried out. For 3′RACE, primers from exon G108 were used on RNA from placenta, skeletal muscle and bone marrow fibroblasts, tissues where G108 was shown to be expressed in. Two different 3′RACE clones of 1173 and 652 bp were derived from all three tissues, suggesting that two different 3′exons a and b exist. The two different forms were termed SHOXa and SHOXb.
To increase chances to isolate the complete 5′portion of a gene known to be expressed at low abundancy, a Hela cell line was treated with retinoic acid and phorbol ester PMA. RNA from such an induced cell line and RNA from placenta and skeletal muscle were used for the construction of a ‘Marathon cDNA library’. Identical 5′RACE cDNA clones were isolated from all three tissues.

Experimental Procedure:

RT-PCR and cDNA Library Construction
Human polyA⁺RNA of heart, pancreas, placenta, skeletal muscle, fetal kidney and liver was purchased from Clontech. Total RNA was isolated from a bone marrow fibroblast cell line with TRIZOL reagent (Gibco-BRL) as described by the manufacturer. First strand cDNA synthesis was performed with the Superscript first strand cDNA synthesis kit (Gibco-BRL) starting with 100 ng polyA⁺RNA or 10 μg total RNA using oligo(dt)-adapter primer (GGCCACGCGTCGACTAGTAC[dT]₂₀N. After first strand cDNA synthesis the reaction mix was diluted 1/10. For further PCR experiments 5 μl of this dilutions were used.
A ‘Marathon CDNA library’ was constructed from skeletal muscle and placenta polyA⁺RNA with the marathon cDNA amplification kit (Clontech) as described by the manufacturer.
Fetal brain (catalog # HL5015b), fetal lung (HL3022a), ovary (HL1098a), pituitary gland (HL1097v) and hypothalamus (HL1172b) cDNA libraries were purchased from Clontech. Brain, kidney, liver and lung cDNA libraries were part of the quick screen human cDNA library panel (Clontech). Fetal muscle cDNA library was obtained from the UK Human Genome Mapping Project Resource Center.

D. Sequence Analysis and Structure of SHOX Gene

A consensus sequence of SHOXa and SHOXb (1349 and 1870 bp) was assembled by analysis of sequences from the 5′ and 3′RACE derived clones. A single open reading frame of 1870 bp (SHOXa) and 1349 bp (SHOXb) was identified, resulting in two proteins of 292 (SHOXa) and 225 amino acids (SHOXb). Both transcripts a and b share a common 5′end, but have a different last 3′exon, a finding suggestive of the use of alternative splicing signals. A complete alignment between the two cDNAs and the sequenced genomic DNA from cosmids LL0YNCO3″M″15D10 and LL0YNC3″M″34F5 was achieved, allowing establishment of the exon-intron structure (FIG. 4). The gene is composed of 6 exons ranging in size from 58 bp (exon III) to 1146 bp (exon Va). Exon I contains a CpG-island, the start codon and the 5′ region. A stop codon as well as the 3′-noncoding region is located in each of the alternatively spliced exons Va and Vb.

EXAMPLE 3

Two cDNAs have been identified which map to the 160 kb region identified as critical for short stature. These cDNAs correspond to the genes SHOX and pET92. The cDNAs were identified by the hybridization of subclones of the cosmids to cDNA libraries.
Employing the set of cosmid clones with complete coverage of the critical region has now provided the genetic material to identify the causative gene. Positional cloning projects aimed at the isolation of the genes from this region are done by exon trapping and cDNA selection techniques. By virtue of their location within the pseudoautosomal region, these genes can be assumed to escape X-inactivation and to exert a dosage effect.
The cloning of the gene leading to short stature when absent (haploid) or deficient, represents a further step forward in diagnostic accuracy, providing the basis for mutational analysis within the gene by e.g. single strand conformation polymorphism (SSCP). In addition, cloning of this gene and its subsequent biochemical characterization has opened the way to a deeper understanding of biological processes involved in growth control.
The DNA sequences of the present invention provide a first molecular test to identify individuals with a specific genetic disorder within the complex heterogeneous group of patients with idiopathic short stature.

EXAMPLE 4

Expression Pattern of SHOXa and SHOXb

Northern blot analysis using single exons as hybridisation probes reveiled a different expression profile for every exon, strongly suggesting that the bands of different size and intensities represent cross-hybridisation products to other G,C rich gene sequences. To achieve a more realistic expression profile of both genes SHOXa and b, RT-PCR experiments on RNA from different tissues were carried out. Whereas expression of SHOXa was observed in skeletal muscle, placenta, pancreas, heart and bone marrow fibroblasts, expression of SHOXb was restricted to fetal kidney, skeletal muscle and bone marrow fibroblasts, with the far highest expression in bone marrow fibroblasts.
The expression of SHOXa in several cDNA libraries made of fetal brain, lung and muscle, of adult brain, lung and pituitary and of SHOXb in none of the tested libraries gives additional evidence that one spliced form (SHOXa) is more broadly expressed and the other (SHOXb) expressed in a predominantly tissue-specific manner.
To assess the transcriptional activity of SHOXa and SHOXb on the X and Y chromosome we used RT-PCR of RNA extracted from various cell lines containing the active X, the inactive X or the Y chromosome as the only human chromosomes. All cell lines revealed an amplification product of the expected length of 119 bp (SHOXa) and 541 bp (SHOXb), providing clear evidence that both SHOXa and b escape X-inactivation.
SHOXa and SHOXb encode novel homeodomain proteins. SHOX is highly conserved across species from mammalian to fish and flies. The very 5′ end and the very 3′ end—besides the homeodomain—are likely conserved regions between man and mouse, indicating a functional significance. Differences in those amino acid regions have not been allowed to accumulate during evolution between man and mouse.

Experimental Procedures:

a) 5′ and 3′RACE

To clone the 5′ end of the SHOXa and b transcripts, 5′RACE was performed using the constructed ‘Marathon cDNA libraries’. The following oligonucleotide primers were used: SHOX B rev, GAAAGGCATCCGTAAGGCTCCC (position 697-718, reverse strand [r]) and the adaptor primer AP1. PCR was carried out using touchdown parameters: 94° C. for 2 min, 94° C. for 30 sec, 70° C. for 30 sec, 72° C. for 2 min for 5 cycles. 94° C. for 30 sec, 66° C. for 30 sec, 72° C. for 2 min for 5 cycles. 94° C. for 30 sec, 62° C. for 30 sec, 72° C. for 2 min for 25 cycles. A second round of amplification was performed using 1/100 of the PCR product and the following nested oligonucleotide primers: SHOX A rev, GACGCCTTTATGCATCTGATTCTC (position 617-640 r) and the adaptor primer AP2. PCR was carried out for 35 cycles with an annealing temperature of 60° C.
To clone the 3′ end of the SHOXa and b transcripts, 3′RACE was performed as previously described (Frohman et al., 1988) using oligo(dT)adaptor primed first strand cDNA. The following oligonucleotide primers were used: SHOX A for, GAATCAGATGCATAAAGGCGTC (position 619-640) and the oligo(dT)adaptor. PCR was carried out using following parameters: 94° C. for 2 min, 94° C. for 30 sec, 62° C. for 30 sec, 72° C. for 2 min for 35 cycles. A second round of amplification was performed using 1/100 of the PCR product and the following nested oligonucleotide primers: SHOX B for, GGGAGCCTTACGGATGCCTTTC (position 697-718) and the oligo(dT)adaptor. PCR was carried out for 35 cycles with annealing temperature of 62° C.
To validate the sequences of SHOXa and SHOXb transcripts, PCR was performed with a 5′ oligonucleotide primer and a 3′ oligonucleotide primer. For SHOXa the following primers were used: G310 for, AGCCCCGGCTGCTCGCCAGC (position 59-78) and SHOX D rev, CTGCGCGGCGGGTCAGAGCCCCAG (position 959-982 r). For SHOXb the following primers were used: G310 for, AGCCCCGGCTGCTCGCCAGC and SHOX2A rev, GCCTCAGCAGCAAAGCAAGATCCC (position 1215-1238 r). Both PCRs were carried out using touchdown parameters: 94° C. for 2 min, 94° C. for 30 sec, 70° C. for 30 sec, 72° C. for 2 min for 5 cycles. 94° C. for 30 sec, 68° C. for 30 sec, 72° C. for 2 min for 5 cycles. 94° C. for 30 sec, 65° C. for 30 sec, 72° C. for 2 min for 35 cycles. Products were gel-purified and cloned for sequencing analysis.

b) SSCP Analysis

SSCP analysis was performed on genomic amplified DNA from patients according to a previously described method (Orita et al., 1989). One to five μl of the PCR products were mixed with 5 μl of denaturation solution containing 95% Formamid and 10 mM EDTA pH8 and denaturated at 95° C. for 10 min. Samples were immediately chilled on ice and loaded on a 10% Polyacryamidgel (Acrylamide:Bisacryamide=37.5:1 and 29:1; Multislotgel, TGGE base, Qiagen) containing 2% glycerol and 1×TBE. Gels were run at 15° C. with 500V for 3 to 5 hours and silver stained as described in TGGE handbook (Qiagen, 1993).

c) Cloning and Sequencing of PCR Products

PCR products were cloned into pMOSBlue using the pMOSBlueT- Vector Kit from Amersham. Overnight cultures of single colonies were lysed in 100 μl H₂O by boiling for 10 min. The lysates were used as templates for PCRs with specific primers for the cloned PCR product. SSCP of PCR products allowed the identification of clones containing different alleles. The clones were sequenced with CY5 labelled vector primers Uni and T7 by the cycle sequencing method described by the manufacturer (ThermoSequenase Kit (Amersham)) on an ALF express automated sequencer (Pharmacia).
d) PCR Screening of cDNA Libraries
To detect expression of SHOXa and b, a PCR screening of several cDNA libraries and first strand cDNAs was carried out with SHOXa and b specific primers. For the cDNA libraries a DNA equivalent of 5×10⁸pfu was used. For SHOXa, primers SHOX E rev, GCTGAGCCTGGACCTGTTGGAAAGG (position 713-737 r) and SHOX a for were used. For SHOXb, the following primers were used: SHOX B for and SHOX2A rev. Both PCRs were carried out using touchdown parameters: 94° C. for 2 min; 94° C. for 30 sec, 68° C. for 30 sec, 72° C. for 40 sec for 5 cycles. 94° C. for 30 sec, 65° C. for 30 sec, 72° C. for 40 sec for 5 cycles. 94° C. for 30 sec, 62° C. for 30 sec, 72° C. for 40 sec for 35 cycles.
e) PCR Screening of cDNA Libraries
To detect expression of SHOXa and b, a PCR screening of several cDNA libraries and first strand cDNAs was carried out with SHOXa and b specific primers. For the cDNA libraries a DNA equivalent of 5×10⁸pfu was used. For SHOXa, primers SHOX E rev, GCTGAGCCTGGACCTGTTGGAAAGG (position 713-737 r) and SHOX a for were used. For SHOXb, the following primers were used: SHOX B for and SHOX2A rev. Both PCRs were carried out using touchdown parameters: 94° C. for 2 min; 94° C. for 30 sec, 68° C. for 30 sec, 72° C. for 40 sec for 5 cycles. 94° C. for 30 sec, 65° C. for 30 sec, 72° C. for 40 sec for 5 cycles. 94° C. for 30 sec, 62° C. for 30 sec, 72° C. for 40 sec for 35 cycles.

EXAMPLE 5

Expression Pattern of OG12, the Putative Mouse Homolog of Both SHOX and SHOT

In situ hybridisation on mouse embryos ranging from day 5 p.c. and day 18.5 p.c., as well as on fetal and newborn animals was carried out to establish the expression pattern. Expression was seen in the developing limb buds, in the mesoderm of nasal processes which contribute to the formation of the nose and palate, in the eyelid, in the aorta, in the developing female gonads, in the developing spinal cord (restricted to differentiating motor neurons) and brain. Based on this expression pattern and on the mapping position of its human homolog SHOT, SHOT represents a likely candidate for the Cornelia de Lange syndrome which includes short stature.

EXAMPLE 6

Isolation of a Novel SHOX-Like Homeobox Gene on Chromosome Three, SHOT, Being Related to Human Growth/Short Stature

A new gene called SHOT (for SHOX-homolog on chromosome three) was isolated in human, sharing the most homology with the murine OG12 gene and the human SHOX gene. The human SHOT gene and the murine OG12 genes are highly homologous, with 99% identity at the protein level. Although not yet proven, due to the striking homology between SHOT and SHOX ( identity within the homeodomain only), it is likely that SHOT is also a gene likely involved in short stature or human growth.
SHOT was isolated using primers from two new human ESTs (HS 1224703 and HS 126759) from the EMBL database, to amplify a reverse-transcribed RNA from a bone marrow fibroblast line (Rao et al, 1997). The 5′ and 3′ ends of SHOT were generated by RACE-PCR from a bone marrow fibroblast library that was constructed according to Rao et al., 1997. SHOT was mapped by FISH analysis to chromosome 3q25/q26 and the murine homolog to the syntenic region on mouse chromosome 3. Based on the expression pattern of OG12, its mouse homolog, SHOT represents a candidate for the Cornelia Lange syndrome (which shows short stature and other features, including craniofacial abnormalities) mapped to this chromosomal interval on 3q25/26.

EXAMPLE 7

Searching for Mutations in Patients with Idiopathic Short Stature

The DNA sequences of the present invention are used in PCR, LCR, and other known technologies to determine if such individuals with short stature have small deletions or point mutations in the short stature gene.
A total of initially 91 (in total 250 individuals) unrelated male and female patients with idiopathic short stature (idiopathic short stature has an estimated incidence of 2-2.5% in the general population) were tested for small rearrangements or point mutations in the SHOXa gene. Six sets of PCR primers were designed not only to amplify single exons but also sequences flanking the exon and a small part of the 5′UTR. For the largest exon, exon one, two additional internal-exon primers were generated. Primers used for PCR are shown in table 2.
Single strand conformation polymorphism (SSCP) of all amplified exons ranging from 120 to 295 bp in size was carried out. Band mobility shifts were identified in only 2 individuals with short stature (Y91 and A1). Fragments that gave altered SSCP patterns (unique SSCP conformers) were cloned and sequenced. To avoid PCR and sequencing artifacts, sequencing was performed on two strands using two independent PCR reactions. The mutation in patient Y91 resides 28bp 5′ of the start codon in the 5′UTR and involves a cytidine-to-guanine substitution. To find out if this mutation represents a rare polymorphism or is responsible for the phenotype by regulating gene expression e.g. though a weaker binding of translation initiation factors, his parents and a sister were tested. As both the sister and father with normal height also show the same SSCP variant (data not shown), this base substitution represents a rare polymorphism unrelated to the phenotype.
Cloning and sequencing of a unique SSCP conformer for patient A1 revealed a cytidine-to-thymidine base transition (nucleotide 674) which introduces a termination codon at amino-acid position 195 of the predicted 225 and 292 amino-acid sequences, respectively. To determine whether this nonsense mutation is genetically associated with the short stature in the family, pedigree analysis was carried out. It was found that all six short individuals (defined as height below 2 standard deviations) showed an aberrant SSCP shift and the cytidine-to-thymidine transition. Neither the father, nor one aunt and maternal grandfather with normal height showed this mutation, indicating that the grandmother has transferred the mutated allele onto two of her daughters and her two grandchildren. Thus, there is concordance between the presence of the mutant allele and the short stature phenotype in this family.
The identical situation as indicated above was found in another short stature patient of Japanese origin.

EXAMPLE 8

The DNA sequences of the present invention are used to characterize the function of the gene or genes. The DNA sequences can be used as search queries for data base searching of nucleic acid or amino acid databases to identify related genes or gene products. The partial amino acid sequence of SHOX93 has been used as a search query of amino acid databases. The search showed very high homology to many known homeobox proteins. The cDNA sequences of the present invention can be used to recombinantly produce the peptide. Various expression systems known to those skilled in the art can be used for recombinant protein production.
By conventional peptide synthesis (protein synthesis according to the Merrifield method), a peptide having the sequence CSKSFDQKSKDGNGG (SEQ ID NO: 42) was synthesized and polyclonal antibodies were derived in both rabbits and chicken according to standard protocols.

REFERENCES

The following references are herein incorporated by reference.
Ashworth A, Rastan S, Lovell-Badge R, Kay G (1991): X-chromosome inactivation may explain the difference in viability of X0 humans and mice. Nature 351: 406-408.
Ballabio A, Bardoni A, Carrozzo R, Andria G, Bick D, Campbell L, Hamel B, Ferguson-Smith MA, Gimelli G, Fraccaro M, Maraschio P, Zuffardi O, Guilo S, Camerino G (1989): Contiguous gene syndromes due to deletions in the distal short arm of the human X chromosome. Proc Natl Acad Sci USA 86:10001-10005.
Blagowidow N, Page D C, Huff D, Mennuti M T (1989): Ullrich-Tumer syndrome in an XY female fetus with deletion of the sex-determining portion of the Y chromosome. Am. J. med. Genet. 34:159-162.
Cantrell M A, Bicknell J N, Pagon R A et al. (1989): Molecular analysis of 46,XY females and regional assignment of a new Y-chromosome-specific probe. Hum. Genet. 83: 88-92.
Connor J M, Loughlin S A R (1989): Molecular genetics of Turner's syndrome. Acta Pediatr. Scand. (Suppl.) 356: 77-80.
Disteche C M, Casanova M, Saal H, Friedmen C, Sybert V, Graham J, Thuline H, Page D C, Fellous M (1986): Small deletions of the short arm of the Y-chromosome in 46,XY females. Proc Natl Acad Sci USA 83:7841-7844.
Ferguson-Smith M A (1965): Karyotype-phenotype correlations in gonadal dysgenesis and their bearing on the pathogenesis of malformations. J. med. Genet. 2: 142-155.
Ferrari D, Kosher R A, Dealy C N (1994): Limb mesenchymal cells inhibited from undergoing cartilage differentiation by a tumor promoting phorbol ester maintain expression of the homeobox-containing gene MSX1 and fail to exhibit gap junctional communication. Biochemical and Biophysical Research Communications. 205(1): 429-434.
Fischer M, Bur-Romero P, Brown L G et al. (1990): Homologous ribosomal protein genes in the human X- and Y-chromosomes escape from X-inactivation and possible implementation for Turner syndrome. Cell 63: 1205-1218.
Freund C, Horsford D J, McInnes R R (1996): Transcription factor genes and the developing eye: a genetic perspective. Hum Mol Genet 5: 1471-1488.
Gehring W J, Qian Y Q, Billeter M, Furukubo-Tokunaga K, Schier A F, Resendez-Perez D, Affolter M, Otting G, Wüthrich K (1994): Homeodomain-DNA recognition. Cell 78: 211-223.
Gough N M, Gearing D P, Nicola N A, Baker E, Pritchard M, Callen D F, Sutherland G R (1990). Localization of the human GM-CSF receptor gene to the X-Y pseudoautosomal region. Nature 345: 734736
Grumbach M M, Conte F A (1992): Disorders of sexual differentiation. In: Williams textbook of endocrinology, 8th edn., edited by Wilson J D, Foster D W, pp. 853-952, Philadelphia, W B Saunders.
Hall J G, Gilchrist D M (1990): Turner syndrome and its variants. Pedriatr. Clin. North Am. 37: 1421-1436.
Henke A, Wapenaar M, van Ommen G-J, Maraschio P, Camerino O, Rappold G A (1991): Deletions within the pseudoautosomal region help map three new markers and indicate a possible role of this region in linear growth. Am J Hum Genet 49:811-819.
Hernandez D, Fisher E M C (1996): Down syndrome genetics: unravelling a multifactorial disorder. Hum Mol Genet 5:1411-1416.
Kenyon C (1994): If birds can fly, why can't we? Homeotic genes and evolution. Cell 78: 175-180.
Krumlauf R (1994): Hox genes in vertebrate development. Cell 78: 191-201.
Kulharya A S, Roop H, Kukolich M K, Nachtman R G, Belmont J W, Garcia-Heras J (1995): Mild phenotypic effects of a de novo deletion Xpter→Xp22.3 and duplication 3pter→3p23. Am J Med Genet 56:16-21.
Lawrence P A, Morata G (1994): Homeobox genes: their function in Drosophila segmentation and pattern formation. Cell 78: 181-189.
Lehrach H, Drmnac R, Hoheisel J D, Larin Z, Lemon G, Monaco A P, Nizetic D, et a,. Hybridization finger printing in genome mapping and sequencing. In Davies K E, Tilghman S, Eds. Genome Analysis 1990: 39-81 Cold Spring Harbor, N.Y.
Levilliers J, Quack B, Weissenbach J, Petit C (1989): Exchange of terminal portions of X- and Y-chromosomal short arms in human XY females. Proc Natl Acad Sci USA 86:2296-2300.
Lichter P, Cremer T, Human Cytogenetics: A practical Approach, IRL Press 1992, Oxford, New York, Tokyo
Lippe B M (1991): Turner Syndrome. Endocrinol Metab Clin North Am 20: 121-152. Magenis R E, Tochen M L Holahan K P, Carey T, Allen L, Brown M G (1984): Turner syndrome resulting from partial deletion of Y-chromosome short arm: localization of male determinants. J Pediatr 105: 916-919.
Nelson D L, Ballabio A, Cremers F, Monaco A P, Schlessinger D (1995).—Report of the sixth international workshop on the X chromosome mapping. Cytogenet. Cell Genet. 71: 308-342
Ogata T, Goodfellow P, Petit C, Aya M, Matsuo N (1992): Short stature in a girl with a terminal Xp deletion distal to DXYS15: localization of a growth gene(s) in the pseudoautosomal region. J Med Genet 29:455-459.
Ogata T, Tyler-Smith C, Purvis-Smith S, Turner G (1993): Chromosomal localisation of a gene(s) for Turner stigmata on Yp. J. Med. Genet. 30: 918-922.
Ogata T, Yoshizawa. A, Muroya K, Matsuo N, Fukushima Y, Rappold GA, Yokoya S (1995): Short stature in a girl with partial monosomy of the pseudoautosomal region distal to DXYS15: further evidence for the assignment of the critical region for a pseudoautosomal growth gene(s). J Med Genet 32:831-834.
Ogata T, Matsuo N (1995): Turner syndrome and female sex chromosome aberrations: deduction of the principle factors involved in the development of clinical features. Hum. Genet. 95: 607-629.
Orita M, Suzuki Y, Sekiya T and Hayashi K (1989): Rapid and sensitive detection of point mutations and polymorphisms using the polymerase chain reaction. Genomics 5:874-879.
Pohlschmidt M, Rappold G A, Krause M, Ahlert D, Hosenfeld D, Weissenbach J, Gal A (1991): Ring Y chromosome: Molecular characterization by DNA probes. Cytogenet Cell Genet 56:65-68.
Qiagen (1993) TGGE Handbook, Diagen GmbH, TGMA 4112 3/93.
Rao E, Weiss B, Mertz A et al. (1995): Construction of a cosmid contig spanning the short stature candidate region in the pseudoautosomal region PAR 1. in: Turner syndrome in a life span perspective: Research and clinical aspects. Proceedings of the 4th International Symposium on Turner Syndrome, Gothenburg, Sweden, 18-21 May, 1995., edited by Albertsson-Wikland K, Ranke M B, pp. 19-24, Elsevier.
Rao E, Weiss B, Fukami M, Rump A, Niesler B, Mertz A, Muroya K, Binder G, Kirsch S, Winkelmann M, Nordsiek G, Heinrich U, Breuning M H, Ranke M B, Rosenthal A, Ogata T, Rappold G A (1997): Pseudoautosomal deletions encompassing a novel homeobox gene cause growth failure in idiopathic short stature and Turner syndrome. Nature Genet 15:54-62
Rappold G A (1993): The pseudoautosomal region of the human sex chromosomes. Hum Genet 92:315-324.
Rappold G A, Willson T A, Henke A, Gough N M (1992): Arrangement and localization of the human GM-CSF receptor α chain gene CSF2RA within the X-Y pseudoautosomal region. Genomics 14:455-461.
Ried K, Mertz A, Nagaraja R, Trusnich M, Riley J, Anand R, Page D, Lehrach H., Elliso J, Rappold G A (1995): Characterization of a yeast artificial chromosome contig spanning the pseudoautosomal region. Genomics 29:787-792.
Robinson A (1990): Demography and prevalence of Turner syndrome. In: Turner Syndrome., edited by Rosenfeld R G, Grumbach M M, pp. 93-100, New York, Marcel Dekker.
Rosenfeld R G (1992): Turner syndrome: a guide for physicians. Second edition. The Turner's Syndrome Society.
Rosenfeld R G, Tesch L-G, Rodriguez-Rigau L J, McCauley E, Albertsson-Wikland K, Asch R, Cara J, Conte F, Hall J G, Lippe B, Nagel T C, Neely E K, Page D C, Ranke M, Saenger P, Watkins J M, Wilson D M (1994): Recommendations for diagnosis,treatment, and management of individuals with Turner syndrome. The Endocrinologist 4(5): 351-358.
Rovescalli A C, Asoh S, Nirenberg M (1996): Cloning and characterization of four murine homeobox genes. Proc Natl Acad Sci USA 93:10691-10696.
Schaefer L, Ferrero G B, Grillo A, Bassi M T, Roth E J, Wapenaar M C, van Ommen G-J B, Mohandas T K, Rocchi M, Zoghbi H Y, Ballabio A (1993): A high resolution deletion map of human chromosome Xp22. Nature genetics 4: 272-279.
Shalet S M (1993): Leukemia in children treated with growth hormone. Journal of Pediatric Endocrinology 6: 109-11
Vimpani G V, Vimpani A F, Lidgard G P, Cameron E H D, Farquhar J W (1977) Prevalence of severe growth hormone deficiency. Br Med J. 2: 427-430
Zinn A R, Page D C, Fisher E M C (1993): Turner syndrome: the case of the missing sex chromosome. TIG 9 (3): 90-93.

Claims

1. A method for the treatment of short stature in patients identified as having a genetic: defect in the human growth gene SHOX, comprising

a) identifying a genetic defect in a human subject suspected of having a genetic mutation in the SHOX gene having the nucleotide sequence according to SEQ. ID NO. 14, and

b) administering to said patient a therapeutically active amount of human growth hormone.

2. The method according to claim 1, wherein said genetic defect is identified by obtaining a biological sample molecule to be examined, and amplifying said biological sample molecule in the presence of two nucleotide probes completely or in part complementary to any of the DNA sequences of SEQ. ID. NO: 2 to SEQ ID NO. 7.

3. The method according to claim 1, wherein the genetic mutation is caused by a hot spot of mutation in the nucleic acid sequence encoding a protein truncation at amino acid position 195 in the SHOX gene.

4. The method according to claim 1, wherein said patient is not suffering from Turner's Syndrome.

5. A method for the treatment of short stature in patients identified as having a genetic defect in the human growth gene SHOX, said SHOX gene having the nucleotide sequence according to SEQ. ID NO. 14, comprising

a) identifying a genetic defect in a human subject suspected of having a genetic mutation in the SHOX gene, comprising

i) obtaining a biological sample containing a polynucleotide from a human subject,

ii) amplifying the polynucleotide of i) in the presence of a primer, wherein said primer is an exon flanking primer or a primer to an exon nucleotide sequence of the SHOX gene according to SEQ ID NO: 14 and wherein the oligonucleotide primer has a length of 18-26 nucleotides and the oligonucleotide sequence of said primer is identical to a partial sequence of an exon nucleotide sequence of the SHOX gene according to SEQ ID NO: 14, and

iii) sequencing any amplification product of the polynucleotide of

i) to determine the presence of a genetic mutation in the SHOX gene of said human subject, and

b) administering to said human subject a therapeutically active amount of human growth hormone.

6. The method according to claim 5, wherein the exon nucleotide sequence in step a) ii) is a polynucleotide sequence selected from the group consisting of SHOX ET93 (SEQ ID NO: 2), SHOX G310 (SEQ ID NO: 3), SHOX ET45 (SEQ ID NO: 4), SHOX G108 (SEQ ID NO: 5), SHOX Va (SEQ ID NO: 6) and SHOX Vb (SEQ ID NO: 7).