Inhibition of cellular proliferation by antisense oligodeoxynucleotides to PCNA cyclin.

The proliferating cell nuclear antigen (PCNA or cyclin) is a nuclear protein recently identified as a cofactor of DNA polymerase delta. When exponentially growing Balb/c3T3 cells are exposed to antisense oligodeoxynucleotides to PCNA, both DNA synthesis and mitosis are completely suppressed. A corresponding sense oligodeoxynucleotide has no inhibitory effects. These experiments indicate that PCNA (cyclin) is important in cellular DNA synthesis and in cell cycle progression.

Tumorigenic conversion of p53-deficient colon epithelial cells by an activated Ki-ras gene.

Distinct genetic abnormalities (loss-of-function mutations of APC and p53 and oncogenic activation of Ki-ras) are associated with specific stages of the sporadic, most common types of colorectal tumors. However, the inability to maintain primary colon epithelial cells in culture has hindered the analysis of the pathogenetic role of these abnormalities in colorectal tumorigenesis. We have now established primary cultures of epithelial cells from the colon crypts of p53-deficient mice; these cells are nontumorigenic as indicated by their failure to form colonies in soft agar and to grow as tumors in immunodeficient SCID mice and in immunocompetent syngeneic hosts. Upon ectopic expression of an activated Ki-ras gene, p53-deficient colon epithelial cells form colonies in soft agar and highly invasive subcutaneous tumors in both immunodeficient and immunocompetent mice. Ectopic expression of wild-type p53, but not of a DNA-binding-deficient mutant, markedly suppressed the colony-forming ability of the Ki-ras-transformed p53-deficient epithelial cells. Together, these findings establish a functional synergism in colorectal tumorigenesis dependent on the effects of an oncogenic Ki-ras in a p53-deficient background. This model of tumorigenic conversion of colon epithelial cells might be useful to identify genetic changes associated with disease progression and to evaluate the therapeutic response to conventional and novel anticancer drugs.

Role of the p53 protein in cell proliferation as studied by microinjection of monoclonal antibodies.

Two monoclonal antibodies against the p53 protein, PAb 122 and 200-47, were microinjected into mammalian cells as a probe to determine the role of the p53 protein in cell proliferation. PAb 122 recognizes the p53 proteins of mouse and human cells but not of hamster cells, whereas 200-47 recognizes the p53 proteins of mouse and hamster cells but not of human cells. The ability of these antibodies to inhibit serum-stimulated DNA synthesis of cells in culture correlates with their ability to recognize the species-specific antigenic determinants. More important, however, is the observation that microinjected PAb 122 inhibits the transition of Swiss 3T3 cells from G0 to S phase, but has no effect on the progression of these cells from mitosis to the S phase.

Adenovirus type 2 activates cell cycle-dependent genes that are a subset of those activated by serum.

We have studied a panel of 10 genes and cDNA sequences that are expressed in a cell cycle-dependent manner in different types of cells from different species and that are inducible by different mitogens. These include five sequences (c-myc, 4F1, 2F1, 2A9, and KC-1) that are preferentially expressed in the early part of the G1 phase, three genes (ornithine decarboxylase, p53, and c-rasHa) preferentially expressed in middle or late G1, and two genes (thymidine kinase and histone H3) preferentially expressed in the S phase of the cell cycle. We have studied the expression of these genes in nonpermissive (tsAF8) and semipermissive (Swiss 3T3) cells infected with adenovirus type 2. Under the conditions of these experiments, adenovirus type 2 infection stimulates cellular DNA synthesis in both tsAF8 and 3T3 cells. However, four of the five early G1 genes (c-myc, 4F1, KC-1, and 2A9) and one of the late G1 genes (c-ras) are not induced by adenovirus infection, although they are strongly induced by serum. The other sequences (2F1, ornithine decarboxylase, p53, thymidine kinase, and histone H3) are activated by both adenovirus and serum. We conclude that the cell cycle-dependent genes activated by adenovirus 2 are a subset of the cell cycle-dependent genes activated by serum. The data suggest that the mechanisms by which serum and adenovirus induce cellular DNA synthesis are not identical.

Constitutively expressed c-myb abrogates the requirement for insulinlike growth factor 1 in 3T3 fibroblasts.

The proto-oncogene c-myb, whose expression is usually limited to cells of the hematopoietic lineages, can be expressed in fibroblasts if placed under the control of a constitutive promoter, such as the simian virus SV40 early promoter. 3T3 cells carrying a constitutively expressed human c-myb were found to grow in 1% serum or in a serum-free, platelet-derived growth factor-supplemented medium, whereas the parent cell line, BALB/c 3T3, needed insulinlike growth factor 1 (IGF-1) in addition to platelet-derived growth factor for growth. myb-carrying cells, however, could not grow in platelet-poor plasma. In fibroblasts, therefore, a constitutively expressed c-myb can abrogate the requirement for platelet-poor plasma or IGF-1. When 3T3 cells constitutively expressed both c-myc and c-myb, they could grow in serum-free medium without added growth factors. The ability of c-myb to abrogate in fibroblasts the IGF-1 requirement seems to be due to its ability to induce overexpression of IGF-1, as indicated by an increase in steady-state levels of IGF-1 mRNA. These results have some important implications; for instance, they suggest a commonality of pathways for entry into S phase in different cell types and the possibility of a myb-like or myb-equivalent gene product of critical importance for entry of fibroblasts into S phase.

Gene transfer: DNA microinjection compared with DNA transfection with a very high efficiency.

We have developed a procedure that gives a very high efficiency of transfection in mammalian cells with low-molecular-weight DNA (approximately 10(4) base pairs). The procedure uses cells in suspension that are shocked with polyethylene glycol 4 h after replating. We compared this transfection technique to the standard technique involving manual microinjection of DNA into the nuclei of mammalian cells, using recombinant plasmids containing the simian virus 40 A gene or the herpes simplex virus thymidine kinase gene or both. The efficiency of transfection depends on a number of variables, the most important of which is the difference in transfectability of different cell lines. In our laboratory, the cell line that had the highest efficiency of transfection was tk-ts13, which is derived from baby hamster kidney cells that are deficient in thymidine kinase and temperature sensitive for growth. Under the appropriate conditions, as many as 70% of these cells can be transfected so that transient gene expression can be detected. With the manual microinjection technique, gene expression is independent of the cell line used and occurs faster than after transfection. The results suggest that the critical stage in transfection is the delivery of DNA molecules to the nucleus. Our experiments also indicate that an enzymatic function, in our case, thymidine kinase activity, gives a higher percentage of positive transfectants than when proteins are visualized only by indirect immunofluorescence. The transfection procedure described in this paper is simple and reproducible and, although less efficient than microinjection, ought to be useful in phenotypic and genotypic studies in which transfer of genes to a large number of cells is desirable.

Induction of WAF1/CIP1 by a p53-independent pathway.

The p53-inducible gene WAF1/CIP1 encodes a M(r) 21,000 protein (p21) that has been shown to arrest cell growth by inhibition of cyclin-dependent kinases. Induction of WAF1/CIP1 in cells undergoing p53-dependent G1 arrest or apoptosis supports the idea that WAF1/CIP1 is a critical downstream effector of p53. In the present study, we used embryonic fibroblasts from p53 "knock-out" mice to demonstrate p53-independent induction of WAF1/CIP1. We show that serum or individual growth factors such as platelet-derived growth factor, fibroblast growth factor, and epidermal growth factor but not insulin are able to induce WAF1/CIP1 in quiescent p53-deficient cells as well as in normal cells. The kinetics of this transient induction, which is enhanced by cycloheximide, demonstrates that WAF1/CIP1 is an immediate-early gene the transcript of which reaches a peak at approximately 2 h following serum or growth factor stimulation. On the other hand, DNA damage elicited by gamma-irradiation induces WAF1/CIP1 in normal human and mouse fibroblasts but does not affect WAF1/CIP1 expression in p53-deficient cells. These results suggest the existence of two separate pathways for the induction of WAF1/CIP1, a p53-dependent one activated by DNA damage and a p53-independent one activated by mitogens at the entry into the cell cycle. The possible function of p21 at this early stage is discussed.

Growth-dependent expression of human Mr 53,000 tumor antigen messenger RNA in normal and neoplastic cells.

We have investigated the expression of Mr 53,000 protein (p53) in total RNA isolated from human peripheral blood mononuclear cells stimulated by phytohemagglutinin, in serum-stimulated human diploid fibroblasts, and in normal and tumor cells of human epithelial colon tissue. We have found that the expression of p53 messenger RNA is growth regulated in human cells following kinetics similar to that previously shown in mouse 3T3 cells, and is increased in the large majority of colon adenocarcinomas in comparison to adjacent normal mucosa and adenoma. This increased expression of p53 is accompanied by a nearly proportional increase in the expression of histone H3. As the expression of histone H3 is restricted to the S phase of the cell cycle and therefore measures the growth fraction of a given population, we suggest that the increased expression of p53 observed in the large majority of colon tumors simply reflects the increased number of cycling cells frequently found in a neoplastic tissue. At variance with these findings a true overexpression of p53 was detected in one SV40-transformed human fibroblasts cell line.

Expression of a deletion mutant of the E2F1 transcription factor in fibroblasts lengthens S phase and increases sensitivity to S phase-specific toxins.

To better understand how the E2F1 transcription factor contributes to the process of cell proliferation, NIH-3T3 cell lines were generated that constitutively express either the wild-type E2F1 protein or an amino terminal deletion mutant, termed E2F1d87. Proliferating E2F1d87-expressing cells exhibit a significant lengthening of S phase relative to control and E2F1 cell lines and are hypersensitive to the cytotoxic effects of the S phase-specific antitumor drug camptothecin. This sensitivity is associated with an increase in drug-induced p53 and WAF1 levels. The E2F1 and E2F1d87 cell lines are both able to initiate, but not complete, S phase under conditions of serum starvation. However, quantitation of DNA synthesis, during culture in serum-deprived media, indicates that the E2F1d87 cell line synthesizes more DNA/cell as compared to the E2F1 cell line. Consistent with this relative increase in DNA synthesis, the E2F1d87 cell line undergoes camptothecin-induced apoptosis when cultured under conditions of serum starvation, while the control and E2F1 cell lines are unaffected by drug treatment under the same conditions. Thus, the sensitivity of the E2F1d87 cell line to camptothecin is not dependent on cell proliferation. The data presented here suggest that cell cycle parameters can be manipulated in order to enhance sensitivity of a cell to the toxic effects of specific chemotherapeutic agents.

p53-independent induction of WAF1/CIP1 in human leukemia cells is correlated with growth arrest accompanying monocyte/macrophage differentiation.

The p53 tumor suppressor gene plays a role in controlling a G1 phase checkpoint. The WAF1/CIP1 gene with encodes p21WAF1/CIP1 protein, an inhibitor of cyclin-dependent kinases, is a downstream mediator of p53 function. We examined expression of the WAF1/CIP1 gene and its relationship to growth arrest and differentiation in p53-null human leukemic cell lines. We show that p53-independent induction of WAF1/CIP1 occurs in human leukemia cells treated with 12-O-tetradecanoylphorbol-13-acetate, okadaic acid, or IFN-gamma but not with retinoic acid, vitamin D3, or DMSO. Furthermore, WAF1/CIP1 induction correlates with growth arrest associated with monocyte-macrophage differentiation. The present studies support the idea that WAF1/CIP1 gene expression can be regulated through multiple mechanisms, suggesting that strategies may be designed to restore the G1 checkpoint controls in p53-null cells by targeting these p53-independent mechanisms of WAF1/CIP1 induction.

Human wild-type p53 adopts a unique conformational and phosphorylation state in vivo during growth arrest of glioblastoma cells.

The wild-type (wt) human tumor-suppressor gene product, p53, and its mutant form have been analysed in an in vivo system in which the inducible expression of wt p53 results in growth arrest in the G1 phase of the cell cycle. Two major pools of p53 are detected after wt p53 expression by their differential reactivity with the p53 monoclonal antibodies PAb 421 and 1801 as well as the mutant and wt-specific monoclonal antibodies PAb 240 and 1620; one pool contains wt and mutant p53 and is characterized as having a mutant conformation, whereas the other pool contains only wt p53 with a wt conformation. As G1 arrest is entered, the amount of wt p53 associated with the mutant pool decreases, such that by 12 h free wt and mutant p53 are the major pools. Two-dimensional gel analysis of the p53 pools revealed that free wt p53 is phosphorylated to a greater degree than mutant p53, which correlated with the loss of the PAb 421 epitope on wt p53. In summary, the ability of wt p53 to exert an antiproliferative effect correlates with the presence of a unique conformational state of wt p53 characterized by increased phosphorylation and the loss of both the PAb 421 epitope and association with mutant p53 pool, whereas mutant p53 is unable to assume this conformational state.

Wild type human p53 is antiproliferative in SV40-transformed hamster cells.

The transformation related protein p53 has been implicated in the process of normal cell proliferation and neoplastic transformation. In this study, the influence of wild type human p53 on cell proliferation was examined. Plasmid constructs encoding the wild type human p53 and various mutant p53 cDNAs, driven by the mouse mammary tumor virus (MMTV) promoter linked to the dominant biochemical selection marker gpt, were used in a colony forming assay employing SV40 transformed HR8 hamster cells. Plasmids encoding wild type p53 drastically reduced the number of gpt+ colonies obtained after transfection, whereas the mutant forms of p53 had no effect. Stable clonal hamster cell lines that constitutively express wild type p53 were isolated and found to have altered growth characteristics (i.e. lower saturation densities, increased doubling times). These findings are consistent with the notion that wild type p53 protein could function as a growth suppressor. The potential role of p53 in the normal cell cycle and in the transformation process is discussed.

The carboxy-terminal serine 392 phosphorylation site of human p53 is not required for wild-type activities.

Wild-type p53 functions in the G1 DNA damage checkpoint pathway by activating gene transcription and preventing cell cycle progression. Others reported that mutation of the serine 386 codon in mouse p53 abolished its ability to suppress growth. Serine 386 of murine p53 and the homologous residue of human p53, serine 392, are phosphorylated in vivo and can be phosphorylated in vitro by casein kinase II (CKII). We constructed mutants that changed serine 392 of human p53 to alanine (p53-S392A) or aspartic acid (p53-S392D); cotransfection of both these mutants with a reporter gene carrying a p53-responsive element into the p53-null Saos-2 cell line activated transcription as well as did wild-type p53. Furthermore, both mutants blocked cell cycle progression after transient transfection in these cells. A stable derivative of the T98G human glioblastoma cell line was established that expressed p53-S392A in response to dexamethasone. Overexpression of this mutant activated transcription of the endogenous waf1 (also called cip1) and mdm2 genes to the same extent as wild-type p53 and also produced growth arrest. Finally, p53-S392A and p53-S392D suppressed foci formation by activated ras and adenovirus E1A oncogenes as efficiently as did wild-type p53. Thus, unlike mutants that altered the serine 15 phosphorylation site, elimination of the serine 392 phosphorylation site had no discernible effect on p53 function. We conclude that neither phosphorylation nor RNA attachment to serine 392 are required for human p53's ability to suppress cell growth or to activate transcription in vivo.

Mutation of the serine 15 phosphorylation site of human p53 reduces the ability of p53 to inhibit cell cycle progression.

Overexpression of wild-type p53 prevents cells from entering the S phase of the cell cycle. The amino-terminal transactivation region of p53 is phosphorylated by several protein kinases, including DNA-PK, a nuclear serine/threonine protein kinase that in vitro requires DNA for activity. DNA-PK was recently shown to phosphorylate serines 15 and 37 of human p53 (Lees-Miller et al., 1992. Mol. Cell. Biol., 12, 5041-5049). To prevent phosphorylation at these sites, mutants were constructed that changed the codons for serine 15 or serine 37 to alanine codons. Expression of p53-Ala-37 in stably transformed T98G cells blocked progression of the cells into S phase as well as did the expression of wild-type p53. In contrast, p53-Ala-15 was partially defective in blocking cell cycle progression. Several cell clones transformed with the mutant p53-Ala-15 gene expressed normal levels of p53 mRNA but accumulated little or no detectable p53 protein. However, by using a transient expression system driven by a strong cytomegalovirus promoter, we showed that the inability of p53-Ala-15 to fully block cell cycle progression was not due to inadequate levels of expression or to a failure of the mutant protein to accumulate in the nucleus. These results suggest that phosphorylation of Ser-15 may affect p53 function.

Cell cycle regulation and the p53 tumor suppressor protein.

Somatic mutations of the p53 gene have been implicated as causal events in the formation of a large number of common human tumors. Several lines of evidence suggest that the nuclear phosphoprotein encoded for by the wild-type gene (wt-p53) plays a role in regulating cell proliferation. Wt-p53 protein encodes a potent negative growth regulatory function that is lacking in mutant forms of the protein found in human tumors. In this review, the relationship between the expression of wt-p53 protein and cell proliferation is examined with emphasis on recent studies that provide clues as to the possible role that p53 plays in cell cycle regulation. A model for the action of p53 in regulating cell proliferation is proposed in which wt-p53 acts as a "checkpoint" protein to control the transit of cells through the restriction point in late G1-phase. After cells pass this "checkpoint" they become committed to enter S-phase and initiate DNA replication. This checkpoint function may be defective in cells that lack p53, express mutant p53, or in which the antiproliferative form of the protein is functionally inactive. Under these conditions stringent control of the initiation of DNA replication may no longer be possible, providing an environment conducive to the emergence oncogenic clones.

Immunohistochemical localization of p21(WAF1/CIP1) in normal, hyperplastic, and neoplastic uterine tissues.

p21WAF1/CIP1 is a nuclear protein that binds to cyclin-dependent kinase complexes (CDKs) and inhibits the activity of multiple kinases. These CDKs are involved in the regulation of cell cycle progression at several checkpoints. In this study, the authors have analyzed by immunohistochemistry the expression of p21WAF1/CIP1 in normal uterine tissues, 12 endometrial hyperplasias, 17 endocervical adenocarcinomas, and 31 endometrial adenocarcinomas. In addition, a group of 10 leiomyomas and 10 uterine leiomyosarcomas were also stained. To evaluate cell proliferation, the monoclonal antibody Ki-67 was used in all of the available cases. Terminally differentiated epithelial endocervical and endometrial cells showed variable expression of p21WAF1/CIP1, whereas the endometrial hyperplasias, and endocervical and endometrial adenocarcinomas showed decreased expression or were negative. All of the cases of cervical squamous dysplasia were positive. Normal smooth muscle cells and 50% of leiomyomas were negative, whereas all leiomyosarcomas showed expression of p21WAF1/CIP1. These results indicate that p21WAF1/CIP1 contributes to differentiation in normal endometrial and endocervical glands. The decreased expression of p21WAF1/CIP1 in endometrial hyperplasias and carcinomas may be important in the process of neoplastic transformation. The role of certain CDK inhibitors, such as p21WAF1/CIP1, is different in epithelial and mesenchymal tumorigenesis in the uterus.

The role of p53, p21WAF1/C1PI, and bcl-2 in radioresistant colorectal carcinoma.

Genetic alterations in the p53 tumor suppressor gene are common in human colorectal cancers, occurring in approximately 70% of tumors. In vitro studies have shown that wild-type p53 is involved in controlling cell cycle checkpoint functions and apoptosis involved in the cytotoxic response induced by ionizing radiation and several anticancer chemotherapeutic agents. Wild-type p53 protein can transcriptionally activate the WAF gene, which encodes a cyclin-dependent kinase inhibitory protein, p21WAF1/C1PI protein, and transcriptionally repress the bcl-2 gene, which encodes an inhibitor of apoptosis. To learn more about the in vivo relationship between p53 protein and the expression of p21WAF1/C1PI and bcl-2 proteins in human colorectal cancers treated with radiation therapy, we examined the expression of these proteins by immunohistochemistry in pre-irradiated biopsy specimens and surgical specimens with residual tumor of 27 patients with colorectal carcinoma. Cell proliferation was measured using Ki-67 expression in the tumor cells. The p53 protein was not detected in normal colorectal mucosa, but it was expressed in 21 of 27 (78%) of pre-irradiated tumor samples and in 19 of 27 (70%) of post-irradiated tumors. Expression of the bcl-2 protein in normal colorectal mucosa was confined to the basal epithelial cells of the crypts. Diffuse bcl-2 staining was detected in tumor cells in 13 of 27 (48%) of pre-irradiated samples and in 14 of 27 (52%) of post-irradiated samples. p21WAF1/C1PI expression was detected in 14 of 27 (52%) of pre-irradiated samples but only in 7 of 27 (26%) of post-irradiated samples. No inverse relationship between expression of p53 protein and abnormal bcl-2 expression was apparent. p21WAF1/C1PI was expressed in most nonproliferating Ki-67-negative epithelial cells at the apical tips of the crypts in normal colorectal mucosa, but not in proliferating Ki-67-positive cells of adjacent adenomatous mucosa. An inverse relationship between Ki-67 and p21WAF1/C1PI expression was observed in normal colorectal mucosa and adjacent adenomatous mucosa. After radiation therapy, p53 protein accumulation did not change among residual tumors in 18 cases (three of which were initially negative and remained negative); in four cases there was a significant increase, and five cases had a substantial decrease of p53 expression. Aberrant bcl-2 expression is not correlated with expression of p53 and does not increase significantly in post-irradiated tumor cells. p21WAF1/C1PI expression is markedly reduced in tumor cells that survive radiation therapy.

Cell cycle effects of microinjected antisense oligodeoxynucleotides to p34cdc2 kinase.

In this study the effect of antisense oligomers targeted against the mRNA transcripts of p34cdc2 kinase on G1 progression into S-phase was examined. For this purpose, antisense, sense, or nonsense oligomers were introduced directly into the cytoplasm of T98G cells grown in monolayer cultures by glass-capillary microinjection. The microinjection of antisense oligomers (but not sense or nonsense oligomers) into growth-arrested cells before serum stimulation inhibited G1 progression into S-phase. This inhibition was correlated with a reduction in the steady-state levels of nuclear p34cdc2 protein. Microinjection of antisense oligomers into cells at 2 and 6 hours after serum stimulation also resulted in a marked inhibition in the ability of cells to enter S-phase. The inhibitory effect decreased when cells were microinjected at 12 hours after serum stimulation. When cells were microinjected at 18 and 24 hours after serum stimulation, only a slight inhibition was observed. As the antisense oligomers were introduced directly into the cytoplasm of cells at each of the time points examined, the observed differences in the inhibitory effects of the antisense oligomers at later times after serum stimulation cannot be explained by differences in uptake. An alternative explanation is that after a certain threshold level of nuclear p34cdc2 protein is reached in late G1 phase; no further increase is necessary, because the cells become committed to enter S-phase. In yeast, p34cdc2 appears to play an important role in the G1/S-phase transition at a control point in late G1 phase called START (reviewed by Lewin). In mammalian cells a control point that could be equivalent to START is the "restriction point" which is defined as the time after which inhibition of protein synthesis fails to block entry into S-phase (reviewed by Pardee). The effects observed with antisense oligomers to p34cdc2 kinase are strikingly similar to what is observed when low concentrations of the drug cycloheximide are added to these cells at different times after serum stimulation; entry into S-phase is significantly inhibited when cycloheximide is added up to 12 hours postimulation. Thus, the results reported in this study are in agreement with the idea that p34cdc2 kinase plays a role in the G1/S phase transition in mammalian cells. Finally, introduction of antisense oligomers directly into the cytoplasm of cells grown in monolayer cultures by glass-capillary microinjection appears to be a viable alternative to simply adding the oligomers to the culture medium.(ABSTRACT TRUNCATED AT 400 WORDS)

Computer-assisted microfluorometric detection of individual malignant bladder cells.

A prospective study was done comparing the use of computer-assisted fluorescence photomicroscopy (cytophotometry) with routine cytology for the detection of human bladder cancers. A total of 129 specimens were analyzed from 89 patients. The patients were divided into two groups: Group I, with cancer demonstrated cystoscopically (55 specimens) and Group II, with no demonstrable cancer (74 specimens). The false-negative rate for cytology in the Group I patients was 64 percent but only 7 percent for microfluorometry. The false-positive rate for cytology in Group II was 5 percent but 59 percent for microfluorometry because of the presence of "reactive" cells. The sequential measurement of both routine cytology and deoxyribonucleic acid (DNA) content can be performed on the same cells. When both modalities were employed, the false-negative and false-positive rates were only 4 percent and 5 percent, respectively. The technique and advantages of cytophotometry are discussed in detail, especially in relation to flow cytometry.

Production of transgenic mice following deoxyribonucleic acid microinjection and embryo freezing.

Experiments with mouse embryos were designed to assess the feasibility of freezing embryos after DNA microinjection. One-cell pronuclear stage mouse embryos were microinjected with cloned deoxyribonucleic acid (DNA) and cultured in vitro to the late eight-cell stage. Microinjected and matched control embryos were frozen and stored in liquid nitrogen. Following thawing, embryos were cultured for 8 h and transferred to recipient females. In a separate set of experiments, embryos were transferred to recipients immediately following DNA microinjection. Control (uninjected) embryos developed to the late eight-cell stage significantly better than surviving microinjected embryos. Of the embryos thawed, 76% of the microinjected and 60% of the control embryos survived to be transferred to recipients. Progeny were obtained with similar survival rates from both groups following embryo transfer with transgenic mice identified among the progeny from microinjected embryos. Mouse embryos can be microinjected with DNA, cultured in vitro, frozen, thawed, transferred to recipients and transgenic progeny can be obtained.