下载掌阅APP，畅读海量书库
立即打开
Linkage Analysis in Man by Somatic Cell Genetics
F. H. Ruddle
Editor’s Note
Today, linkage analysis, where the location of a gene is worked out relative to a known sequence, is a sophisticated affair involving high-throughput machinery and computers. But in the 1970s researchers had to resort to cell culture methods, mapping genes from somatic cell hybrids. Here geneticist Frank Ruddle describes the state of play. Hybrids combine different genomes within a single cell, and the discovery that chromosomes from parental genomes can be lost or segregated from the hybrid cell, enabled researchers to effectively isolate and identify chromosome fragments and assign genes to chromosomes. The technique correctly assigned many genes to their chromosomes, but was “unacceptably slow”. Somatic cell recombination and gene transfer, Ruddle prophetically muses, could provide the way forwards. 中文
Techniques for the study of somatic cell genetics, and particularly those involving the expression of enzyme markers in hybrid cells, have already made possible a large number of gene-chromosome assignments. Genetic and family studies, as well as cellular studies on recombination and gene transfer, promise more and quicker results in the future. 中文
SOMATIC cell hybridization, first demonstrated by Barski et al. 1 , was an important early step in the formulation of somatic cell genetic systems, allowing as it does the combination of genetically different genomes within a single cell. In a series of investigations Ephrussi and his colleagues showed that hybrid combinations could be obtained between the cells of different species 2 , and that chromosomes of one or both parental genomes could be lost or segregated from the hybrid cell 3 . Weiss and Green 4 first demonstrated the practical application of somatic fusion-segregation systems for the purpose of gene mapping in man (see below). Other investigators have contributed useful procedures which enhance the formation of hybrid cells and their enrichment. These developments now allow a completely new cell culture approach to gene mapping in man. 中文
For the formation of hybrid cells the parental cells are mixed together and co-cultivated. Membrane fusion can be enhanced by treatment with inactivated Sendai virus 5,6 or with lysolecithin 7 . Fusion between two parental cells of different origins gives rise to a binucleate heterokaryon. Heterokaryons have a short life expectancy, and following their first mitosis, generally form mononucleated or hybrid daughter cells which contain chromosomes from both parental genomes. In many parental cell combinations, the hybrids have an infinite life expectancy and can be grown into large clonal cell populations. In man–mouse and man–Chinese hamster hybrids there is a unilateral loss or segregation of human chromosomes. The segregation of human chromosomes is variable in extent in different clones, and in many instances clones can be obtained which maintain for many generations a partial human chromosome constitution. Thus, it is possible in effect to sample different numbers and combinations of human chromosomes in a series of man–rodent hybrids of independent origin. Each clone represents a partial human karyotype superimposed on an intact mouse or Chinese hamster genome. The experimental isolation of partial human chromosome complements forms the basis of somatic cell linkage analysis. 中文
Enzyme complementation has been used to enrich for hybrids in mixed populations of parental cells. Littlefield has shown that drug resistance mutant cell lines can be useful in this regard 8 . Mutant cell lines can be selected which are deficient in the enzymes hypoxanthine-guanine phosphoribosyltransferase (HGPRT) and thymidine kinase (TK) by exposing cells to the antimetabolites thioguanine and BUdR respectively. HGPRT deficient cells cannot incorporate hypoxanthine, whereas TK deficient cells cannot metabolize thymidine. If de novo synthesis of purines and pyrimidines is blocked by the antimetabolite aminopterin, cells become dependent for survival on exogenous hypoxanthine and thymidine. HGPRT and TK deficient cells are thus conditional lethal mutants which are killed by aminopterin irrespective of the availability of hypoxanthine and thymidine. The fusion of HGPRT deficient with TK deficient parental cells yields hybrid cells whose enzyme deficiencies are complemented and which can grow in nonpermissive selection medium containing hypoxanthine, aminopterin, and thymidine (HAT medium). Kusano et al. have shown that adenine phosphoribosyltransferase (APRT) deficiency mutations can be used similarly for hybrid cell selection 9 . Conditional lethal mutants other than those based on drug resistance can also be used for hybrid selection. Puck et al. have used nutritional auxotrophs with good results 10,11 , and it is likely that temperature sensitive mutations can be used in the same way 12 . Moreover, it is possible to make use of conditional mutant established rodent cell lines in combination with diploid human fibroblasts or leucocytes which have low growth potentials in vitro 12,13 . One can select against the rodent parent using nonpermissive medium and against the human diploid parent by virtue of its inherently poor growth characteristics. 中文
Conditional lethal cell mutants in rodent cell populations are extremely useful for genetic analysis. In nonpermissive conditions only hybrids which retain the complementing human gene will survive. Thus, if one forms hybrids between HGPRT deficient rodent cells and wild type human cells (HGPRT + ), and cultivates them in HAT medium, only those cells which retain the human HGPRT gene will survive. Generally, the intact human X chromosome which carries the HGPRT gene is retained in the complemented hybrid. It is possible to conceive of a series of rodent cell lines each of which carry different conditional lethal mutations which are complemented by genes on each of the human autosomes and sex chromosomes. Such a panel of rodent cell lines would be extremely useful in mapping studies because each would produce hybrids in which the segregation of a specific human chromosome would be fixed. Conditional lethal mutants of this type are tabulated in Table 1. It should be pointed out that the drug resistance complementation systems also lend themselves to counter selection. Cells which retain TK, APRT, and HGPRT activity are susceptible to the antimetabolites BUdR, fluoroadenine, and thioguanine, respectively 8-10 . Thus it is possible to use these agents in permissive medium to select against hybrid cells which have retained human chromosomes 17, 16 and X . 中文
Table 1. Conditional Drug Resistance and Nutritional Auxotrophic Genetic Markers
In rodent–human hybrids, the human chromosomes are unilaterally segregated, both homologous human and rodent enzymes are expressed and can be identified, and human and mouse chromosomes can be discriminated and accurately identified on an individual basis. These hybrids are therefore particularly suitable for human gene linkage analysis. 中文
The loss of human chromosomes from mouse–human and Chinese hamster–human hybrids is well documented, but the mechanism of loss is poorly understood. Preferential loss of human chromosomes in rodent X human hybrids, irrespective of the origins of the parental cell populations, is the rule, and only one possible exception has been reported 15 . A mechanism of loss suggested by Handmaker (personal communication) is that human chromosomes cannot attach efficiently to the hybrid spindle apparatus and thus have higher incidence of loss. Another possibility, which is not necessarily incompatible with this, is a mechanism of segregation based on random non-disjunction of mouse and human chromosomes in combination with the preferential selection of hybrids which possess partial human karyotypes 16 . Nabholz et al. 17 have suggested that chromosomes are lost by two temporarily distinct processes. Early loss, possibly during the first several mitotic divisions after fusion, can result in the abrupt loss of a few or many human chromosomes. Late loss is characterized by slow progressive loss in some instances over many cell generations. Nabholz et al. 17 have also presented evidence that human chromosomes are segregated non-randomly into hybrid clones. Preliminary results in our laboratory, based on twenty-eight independent hybrid clones, indicate a very low frequency of retention of human chromosome 9 (7%) compared with the overall frequency of human chromosome retention (29%). It has been reported that hybrids with two rodent genomes (2 s hybrids) retain more human chromosomes than 1 s hybrids 15 . The relationship between rodent and human chromosome number is significant and should be resolved, because it is fundamental to the problem of chromosome segregation. 中文
The amino-acid constitution of homologous enzymes between man and rodents generally differs to some degree as a result of evolutionary divergence, and it is generally possible to detect these differences by electrophoretic procedures. There is thus a very large potential catalogue of genetic markers in the rodent–human cell hybrid system, limited only by the development of adequate test procedures. A compilation of isozyme procedures has been reported by Ruddle and Nichols 18 . 中文
It is important for genetic testing that the enzyme markers be constitutive—that is, they must invariably be expressed if the corresponding structural gene is retained in the hybrid. Facultative markers are defined as those markers which are subject to modulation and which may not be expressed even if the corresponding cistron is present. It is very difficult to define phenotypes as being absolutely constitutive or facultative. Generally speaking, enzymes which are expressed in all cell types in vivo and which contribute to vital metabolic pathways are termed constitutive, whereas enzymes which are restricted to one or a few specialized cell types and which do not participate in vital metabolic activities at a cellular level are termed facultative. Good evidence exists for the modulation of certain facultative functions in hybrid combinations between parental cells of different epigenetic types 19 . This necessarily complicates the linkage analysis of such phenotypes. Phenotypes classified as constitutive may under certain conditions be modulated. For example, hybrid clones have been recovered by Ricciuti 20 which possess normal C-7 human chromosomes, but which do not express detectable levels of mannose-phosphate isomerase (MPI) activity. Linkage analysis in other cell hybrids shows a strong correlation between C-7 and MPI. I have concluded that MPI may represent a partially constitutive phenotype. Such phenotypes pose problems for linkage analysis, but they also provide useful material for studies on phenotype modulation. 中文
It is now possible to identify all of the chromosomes of the Chinese hamster, laboratory mouse, and man by cytological procedures. Caspersson and co-workers 21 have shown that quinacrine binds differentially to specific regions of the human chromosomes, and that each chromosome possesses a unique banding pattern. Mouse chromosomes are similarly unique and the banding patterns have now been correlated with known murine linkage groups 22 . Giemsa banding procedures provide results comparable with those of quinacrine 23 . Pardue and Gall have introduced an in situ annealing technique which makes possible the localization of highly redundant DNA in mouse and human chromosomes 24 . Purified isotopically labelled redundant DNA is annealed to intact chromosomes which have been pretreated with DNA denaturation agents. The labelled redundant DNA is hybridized to complementary DNA in the chromosome, and its location revealed by autoradiography. Pardue and Gall have shown that the murine redundant DNA (satellite DNA) is restricted to the centromere regions and that denaturation followed by Giemsa staining reveals positively staining, constitutive heterochromatin regions in the chromosomes. Arrighi and Hsu 25 have adapted this method to the analysis of human chromosomes and subsequent studies have shown a correspondence between constitutive heterochromatin and redundant DNA 26 . Human and mouse satellite DNA are specific and do not cross react. It is thus possible by hybridization in situ to distinguish human and mouse centromeric regions, which has proved useful in the detection of human–mouse chromosome translocations 27 (see below). Several laboratories have reported evidence indicating that the centromeric constitutive heterochromatin in man has different physical properties unique for several of the human chromosomes 28 . 中文
Linkage of enzyme phenotypes can be inferred from their concordant segregation. The human chromosomes maintain their integrity for the most part, seldom undergoing rearrangement or deletion, and the concordant segregation of markers thus provides evidence for their location on the same chromosome irrespective of map distance. It is therefore appropriate to employ the term “synteny” coined by Renwick to signify merely localization on the same chromosome. Synteny testing is performed by comparing the segregation pattern of all markers in all pairwise combinations. The synteny test is less biased if performed on clones of independent origin and the detection of valid syntenic relationships is enhanced by using clones derived from separate hybridization experiments, using different hybrid combinations. This generally entails computer analysis because of the number of clones and markers involved. 中文
Individual genes or syntenic genes can be assigned to specific chromosomes by tabulating the human chromosomes in each of the clones and correlating them with the enzyme markers. The concordant presence or absence between a chromosome and a phenotype provides evidence for the assignment of the gene governing a particular phenotype to a specific chromosome. Twenty to thirty metaphases are analysed per clone by means of quinacrine banding, Giemsa banding, or constitutive heterochromatin staining techniques. Identification is enhanced if cells are first photographed by quinacrine fluorescence and then by constitutive heterochromatin staining procedures. 中文
It is frequently possible to strengthen the assignment of a gene to a particular chromosome by correlating the frequency of a particular chromosome within a clone with the intensity of expression of an assigned phenotype(s). Discrepant clones of two classes can occur, however. In the first, presuming a valid assignment, the chromosome cannot be detected but the phenotype is present. We have demonstrated cryptic, rearranged chromosomes in a number of such instances which explain an apparently discordant chromosome/gene relationship 29 . A second type of discrepant clone involves the presence of a specific chromosome, but the absence of its corresponding phenotype(s). Clones of this type are difficult to explain, but could involve subtle rearrangements in chromosome structure, gene mutation, modulation of gene expression, or technical failure in phenotype detection. 中文
It is possible to assign genes to particular regions of chromosomes such as chromosome arms, or band regions as defined by particular staining reactions. This can be accomplished by making use of chromosome rearrangements such as translocations and deletions in the human parental cell population or by making use of spontaneous chromosome rearrangements which are generated in the hybrids. Translocations of chromosomes to or between chromosomes X , 17, and 16 are useful because these chromosomes possess selectable loci. A number of translocations affecting the same chromosome but with different breakpoints can be used to restrict the localization of genes. For this purpose it will be particularly useful to characterize and store in central repositories all detected human translocations, to serve as a library of rearrangement products for future somatic cell gene mapping. Programs of human mutant cell banking are now being formulated in several countries. An example of regional linkage assignments based on translocations in parental cells is cited below for the X chromosome. 中文
It is also possible to make use of spontaneous, sporadic chromosome rearrangements which occur in hybrid cells to fix the location of genes within subregions of particular chromosome. Human chromosomes within hybrids may undergo rearrangement and even translocation to mouse chromosomes 29 and it has been possible to make use of such rearrangements to restrict the localization of the thymidine kinase gene to the long arm of human chromosome 17 29 . The translocation of human chromosome segments to the mouse chromosome set is significant from a genetic point of view because it may serve to restrict the further segregation of human genes involved in the translocation. It is conceivable that treatment of parental cells or hybrids with physical or chemical chromosome breaking agents could be used to induce chromosome rearrangements. Enrichment procedures could be devised to select particular classes of rearrangement products. Such systems could be used to increase the resolution of subregional chromosome gene assignments. 中文
A significant number of syntenic relationships and chromosome assignments has been established by somatic cell genetics. A survey of the current linkage information is presented below for each of the human chromosomes in turn. The results are also summarized in Table 2. For chromosome 1, Van Cong et al. 30 have reported a syntenic relationship between phosphoglucomutase-1 (PGM 1 ) and peptidase C (Pep C) using mouse–human hybrids. This synteny has been confirmed by Ruddle et al. 31 . Using Chinese hamster–human hybrids, Westerveld et al. 32 have reported a synteny between 6-phosphogluconate dehydrogenase (PGD) and Pep C. Taken together, these findings imply that PGD, Pep C, and PGM 1 are all syntenic. Pep C has been assigned to chromosome 1 using mouse–human cell hybrids 31 . This assignment has been confirmed by the assignment of PGD to chromosome 1 independently by Bootsma et al. (personal communication) and Hamerton et al. 34 using Chinese hamster–human hybrids. If the findings based on cell hybrids are combined with linkages known from human pedigree analysis, the following additional gene markers can be assigned to chromosome 1: zonular pulverulent cataract, Duffy blood group, auriculo-osteodysplasia, salivary amylase, pancreatic amylase, elliptocytosis and rhesus blood group. For complete literature citations see Ruddle et al. 31 中文
Table 2. Assignments of Genes to Chromosomes
These genes were assigned or confirmed by cell hybrid analysis. Each trait is identified by McKusick’s human gene catalogue number 59 . IPO-A and B used here agree with the original designation of Brewer 33 .
Preliminary results in our laboratory (R. P. Creagan and F. H. R.) suggest that isocitrate dehydrogenase (IDH) and NADP-malate dehydrogenase (MOD) which have been shown to be syntenic 44 can be assigned to chromosome 2. No loci have been assigned to chromosome 3. For chromosomes 4 and 5, Kao and Puck 35 using Chinese hamster–human hybrids have demonstrated a positive association between hamster adenine B auxotrophy and a human B group chromosome when hybrids are propagated on minimal medium. The specific enzyme involved is unknown. Chen et al. 36 using mouse–human hybrids have shown that cytoplasmic malate dehydrogenase (MOD) is assignable to chromosome 6 and there is evidence to support a syntenic relationship between indolephenol oxidase, tetrameric form B (IPO-B) and MOD (J. A. Tischfield, R. P. Creagan and F. H. R., unpublished). 中文
McMorris et al. 37 using mouse X human hybrids have assigned mannose phosphate isomerase (MPI) to chromosome 7. Shows 38 has reported a syntenic relationship between MPI and the leucocytic form of pyruvate kinase 3 (PK 3 ). 中文
There are no assignments to chromosomes 8 or 9. There is evidence from mouse–human hybrids for the assignment of the cytoplasmic form of glutamate oxaloacetate transaminase (GOT) to chromosome 10 (unpublished work of R. P. Creagan, J. A. Tischfield, F. A. McMorris, M. Hirschi, T. R. Chen and F. H. R. 中文
Boone et al. 29 using mouse–human hybrids have assigned lactate dehydrogenase A (LDH-A) to chromosome 11. Shows 39 using mouse–human hybrids has reported a syntenic association between LDH-A and human esterase A 4 (EsA 4 ). Van Someren et al. 40 have reported a syntenic association between glutamic-pyruvic transaminase (GPT-C) and LDH-A. The possibility exists, however, that their enzyme detection system is recording LDH-A activity. This may also apply to LDH-B and GPT-B (see below). Nabholz et al. 17 using mouse–human cells have reported a positive correlation between the segregation of LDH-A or B activity and sensitivity of hybrid cells to anti-human cytotoxic antisera. Puck et al . 41 have reported on the segregation of a possibly similar human antigen(s) in Chinese hamster–human hybrid cells, which they have found to be syntenic with LDH-A. 中文
Ruddle and Chen 42 and Chen et al. 36 using mouse–human hybrids have demonstrated positive correlation between lactate dehydrogenase B (LDH-B) and chromosome 12. Hamerton et al. 34 have confirmed this assignment using Chinese hamster–human hybrids. In mouse–human hybrids a syntenic relationship has been demonstrated between LDH-B and peptidase-B (Pep-B) 14,43 . The Pep-B/LDH-B synteny has been confirmed by Shows 39,44 , Van Cong et al . 30 and van Someren et al. , who have also reported a syntenic association between glutamic-pyruvic transaminase-B (GPT-B) and LDH-B 40 . Jones et al. 45 using Chinese hamster–human hybrids have reported a syntenic relationship between LDH-B and the complement to the Chinese hamster glycine auxotrophic mutant A. Serine hydroxymethylase has been implicated as the specific deficiency in glycine A auxotrophy. 中文
No assignments have been made to chromosome 13. For chromosome 14, Ricciuti and Ruddle (ref. 46 and F. Ricciuti and F. H. R., unpublished) using a 14/ X translocation in a human diploid fibroblastic cell strain (KOP) hybridized to a mouse cell line have demonstrated a segregation of nucleoside phosphorylase (NP) with the X linked markers, HGPRT, GPD, and PGK. Somatic cell genetic 46 and family studies 47 have provided evidence for the autosomal linkage of NP. The studies of Ricciuti and Ruddle thereby support the assignment of NP to chromosome 14. Unreported experiments from our laboratory using a 14/22 translocation also confirm the assignment of NP to 14. Hamerton et al . 34 have reported results based on Chinese hamster–mouse hybrids which are consistent with the above findings of Ricciuti and Ruddle. No assignments have been made to chromosome 15. 中文
Tischfield and Ruddle (unpublished) using an adenine phosphoribosyltransferase (APRT) deficient mouse cell line hybridized to normal human diploid cells have obtained evidence for the assignment of APRT to chromosome 16. On evidence from family studies, Robson et al. 48 have assigned α-haptoglobin to chromosome 16. APRT activity variants have been reported in man, and it would be reasonable to identify kindreds in which α-haptoglobin and APRT variants are jointly expressed to test for linkage between these two markers. 中文
For chromosome 17, Green 49 and Migeon and Miller 50 using mouse–human cell hybrids assigned thymidine kinase (TK) to an E group chromosome. This assignment was based on the earlier findings of Weiss and Green 4 and has since been verified by Boone and Ruddle 51 . Miller et al . 52 , Ruddle and Chen 53 , and Boone et al. 29 have now assigned TK specifically to chromosome 17. Boone et al. 29 , making use of a spontaneously occurring 17 translocation to a mouse chromosome, have provided evidence for the assignment of TK to the long arm of chromosome 17. Kit et al. 54 and McDougall et al. 55 have demonstrated that adenovirus 12 infection induces host TK activity, and concurrently a secondary constriction in the proximal segment of the long arm of 17. These findings suggest that the TK gene may be located near the adeno-12 induced gap region. 中文
Creagan et al. (unpublished) using mouse–human cell hybrids have provided evidence for the assignment of peptidase-A (Pep-A) to chromosome 18. 中文
Glucosephosphate isomerase (GPI) has been assigned to chromosome 19 on the basis of evidence from mouse–human cell hybrids 37 . Hamerton et al. 34 have confirmed this assignment using Chinese hamster–human cell hybrids. Linkage studies in the mouse have revealed a loose linkage between GPI and β haemoglobin: it will be of interest to test for a similar linkage relationship in human kindreds. 中文
Boone et al. 29 using mouse–human hybrids reported a weak association between cytoplasmic isocitrate dehydrogenase (IDH), cytoplasmic maleate oxidoreductase (MOR) and chromosome 20. More extensive data from mouse–human hybrids have now shown that IDH and MOR cannot be assigned to 20 and using mouse–human hybrids we have obtained evidence of the assignment of tissue-specific adenosine deaminase (ADA) to chromosome 20 (J. A. Tischfield, R. P. Creagan and F. H. R., unpublished). ADA is asyntenic with both IDH and MOR. Family studies have demonstrated linkage between HL-A phosphoglucomutase 3, P blood group, and ADA 56 . 中文
On chromosome 21, Tan et al. 57 , using somatic cell hybrids, have provided evidence for the syntenic association between indolephenoloxidase-A, dimeric (IPO-A) and a genetic factor (AVP) which controls an antiviral response specifically induced by human interferon. The genetic factor may regulate the interferon receptor and/or the antiviral protein. We have also shown that the interferon and AVP loci are asyntenic 57 . These results confirm earlier studies by Cassingena et al. 58 on their asyntenic association based on monkey–rat somatic cell hybrids. Tan et al. 57 have assigned AVP/IPO-A to chromosome 21. No assignments have yet been made to chromosome 22. 中文
Glucose-6-phosphate dehydrogenase (GPD), hypoxanthine-guanine phosphoribosyltransferase (HGPRT), and phosphoglycerate kinase (PGK) have all been assigned to the X chromosome by segregation analysis in families 59 . Nabholz et al. 17 using mouse–human hybrids confirmed the X linkage of HGPRT. Meera Kahn et al. 60 have demonstrated the X linkage of PGK by cell hybrid analysis. Ruddle et al. 61 using mouse–human hybrids confirmed the X linkage of HGPRT, glucose-6-phosphate dehydrogenase (GPD), and Phosphoglycerate kinase (PGK). Grzeschik et al. 62 have recently provided evidence based on Chinese hamster–human hybrids for the assignment of α-galactosidase (α-Gal) to the X chromosome. In an earlier report, Grzeschik et al. 63 analysed cell hybrids between human KOP cells which possess a 14/ X translocation (KOP) and mouse and Syrian hamster cells. They observed an infrequent segregation of HGPRT and GPD from PGK, which led them to postulate the assignment of PGK to the long arm, and the possible assignment of HGPRT and GPD to the short arm, although assignment to the long arm was not altogether discounted. Ricciuti and Ruddle (ref. 46 and unpublished results) using the same KOP material hybridized to mouse cells have obtained data which indicate that all three markers are located on the X long arm. This suggests that PGK is proximal to the centromere and distant from the other two markers. HGPRT and GPD seem to be close together and distal to the centromere with respect to PGK; preliminary evidence indicates that HGPRT is proximal to GPD. P. Gerald and co-workers (personal communication) have recently studied a human cell with a 19/ X translocation hybridized to mouse cells. A translocation product composed of the 19 long arm, the proximal half of 19 short arm, and the distal half of the X long arm was correlated with GPI, HGPRT, and GPD, but not PGK. This result is consistent with the human X linkage map proposed by Ricciuti and Ruddle 46 . It also confirms the assignment of GPI to 19. 中文
No assignments have been made to the Y chromosome. 中文
The development of a detailed human genetic map is certain to provide insight into the evolutionary origins of man and the primates. LDH-A and LDH-B are located on chromosomes 11 and 12 respectively. These chromosomes are similar in size, centromere position, and banding pattern, which is consistent with the occurrence of a primordial polyploid event in the early primate genome as discussed by Comings 64 . Somatic cell genetic analysis should be feasible for representative members of the order primates using rodent–primate hybrids. Linkage data from such hybrids should provide information on the relatedness of these forms, and yield estimates for rates of evolutionary divergence, especially when combined with comparative studies on the chromosome constitutions and amino-acid sequences of proteins in the representative specimens. 中文
It is already obvious that somatic cell genetics has contributed and will in the future contribute important data to human genetics. Moreover, these developments enhance the significance and future role of family and population genetic studies. Already map distances between genes have been refined or established by the certain knowledge from somatic cell genetics that certain gene pairs are syntenic. We should expect a fruitful interaction and collaboration between practitioners of somatic cell genetics and classical genetics. We must, however, keep clearly in mind that somatic genetic procedures as they now exist are still unacceptably slow and linkage estimates cannot yet be made. If we are soon to develop genetic maps of man comparable to those available for the lower eukaryotes, it will be necessary to develop new procedures. Possibilities for this are to be found in somatic cell recombination and gene transfer. 中文
( 242 , 165-169; 1973)
Frank H. Ruddle
Kline Biology Tower, Yale University, New Haven, Connecticut 06520
References: