Search

To determine the expression of metallothionein (MT) in prostatic carcinoma by immunohistochemical staining. Several lines of evidence have indicated that MT may play a role in carcinogenesis and in drug resistance of tumours.Retrospective pathologic study.Formalin-fixed, paraffin-embedded archival tissues from 39 radical prostatectomies were analysed. All tumour foci were stained by ABC technique using a primary polyclonal rabbit antibody against rat liver MT. The staining intensity for MT was graded on a scale of 0 to 2+, and the histologic grading was done by the scheme of Gleason.Correlation of MT expression with Gleason grade, preoperative serum prostate-specific antigen (PSA) levels, pathologic stage and DNA content, including S-phase fraction (SPF) and proliferative index (PI).Most of the epithelium of normal prostate tissue had patchy, intense MT staining. All the grade II tumours foci showed intense (2+) staining for MT, while all grade IV and V foci were persistently negative. The grade III tumours foci were heterogeneous. The MT-positive foci showed both nuclear and cytoplasmic staining of variable extent. There were 9, 15, 13 and 2 tumours with pathologic stage B, C1, C2 and D1, respectively. The serum PSA levels ranged from 1 to 16 ng/mL. No apparent correlation existed between the MT staining pattern and the pathologic stage or preoperative PSA level. Thirty-four of the tumours were diploid and 5 were tetraploid. There were significantly higher SPF and PI mean values in the MT-stained tumour cells (p0.05), suggesting that MT preferentially stains an epithelial subpopulation, possibly the proliferating cell compartment.The positive correlation of MT expression with Gleason grade in prostatic adenocarcinoma suggests a possible role for MT in oncogenesis in prostate cancer.

To assess nuclear activity by DNA flow cytometry (FCM), Gleason score and serum prostate-specific antigen (PSA) levels in predicting extracapsular tumour involvement in patients with prostate cancer.Retrospective pathologic study.Forty patients with clinical stage-B prostate cancer who underwent radical prostatectomy.The relationship between the pathologic state and each of the proliferative index (PI). Gleason score and PSA level was analysed retrospectively with the use of archival specimens. Preoperative serum PSA levels, were measured by the Hybritech assay. Gleason score was determined by 2 of the authors. Tumours were classified as stage B (confined to prostate), C1 (focal capsular penetration) or C2 (involvement of seminal vesicles or capsular perforation). FCM PI measurements were performed on deparaffinized tumour specimens.All 40 specimens were diploid. There were 9 pathologic stage-B, 16 stage-C1 and 15 stage-C2 tumours. The serum PSA level was 20 ng/mL or less for all patients except 2, for whom the levels were 27.8 ng/mL and 45.9 ng/mL, respectively. A Gleason score lower than 7 had a 76.0% sensitivity and 53.5% specificity in predicting organ confinement. In contrast, a PI of 21 or lower had a 84.0% sensitivity and a 73.0% specificity in predicting organ confinement, with a positive predictive value of 84.0%. Of the 17 tumours with both "favourable" features (Gleason score lower than 7 and PI of 21 or lower), only 1 (5.9%) had extracapsular involvement (stage C2). Of the 6 tumours with both "unfavourable" features (Gleason score higher than 7 and PI of 21 or higher) 5 of 6 were stage C2 and 1 was stage C1.The single most consistent predictor of organ confinement was PI alone.

Proliferative rates of 73 node-negative adenocarcinomas of breast with 5-year or greater follow-up were studied using cytokeratin staining in two-parameter DNA flow cytometry of archival specimens. Quality control data of accuracy of the measurements were determined and all analyses were compared with single-parameter results of the same specimens, using demarcated tumor areas, quadruple analyses, and computerized nonspecific staining subtraction. Mitotic rates of the same samples correlated highly significantly with the S-phase fractions and proliferative index (S + G2 + M phases), especially for the cytokeratin data. The predictive value of mitotic rates was found significant, but that of the DNA flow-cytometry-obtained indices was not, probably because of low numbers of deaths in this study. The cytokeratin method identified heteroploid tumors containing a diploid cell population not identifiable by single-parameter analysis. In conclusion, cytokeratin staining can be reliably applied to DNA flow cytometry of archival specimens giving accurate ploidy, S-phase fractions, and proliferative index data limited almost exclusively to neoplastic cell populations. This will permit large-scale retrospective studies aimed at establishing the usefulness of DNA flow cytometry for clinical decisions on therapy of surgically removed node-negative adenocarcinomas of breast.

Hepatic iron and zinc concentrations were determined in 26 consecutive nonalcoholic cirrhotic patients who had previously undergone portacaval shunting and then liver transplantation and 37 control patients. Stainable iron was graded on a scale of 0 to 4; 11 shunt patients and 3 control patients had grade 2 to 4 iron staining. Mean hepatic iron concentration was significantly increased (20.1 +/- 4.1 mumol/gm) in the shunt group compared with the control group (9.1 +/- 2.3 mumol/gm) (normal value,35 mumol/gm) (p = .015, Student's t test). Mean hepatic iron index (hepatic iron/age) was 0.5 +/- 0.10 in the shunt group and 0.19 +/- 0.04 in the control group (p = 0.002). One shunt patient had a hepatic iron index of greater than 2. Hepatic zinc was not significantly different between the shunt group (2.42 +/- 0.18 mumol/gm) and the control group (2.56 +/- 0.14 mumol/gm). Serial biopsy specimens were analyzed in 17 cases (6 shunt, 11 controls), and the mean annual rate of iron accumulation was significantly greater in shunt patients (4.75 mumol/gm/yr) than in control cases (0.93 mumol/gm/yr, p = 0.037). Although increased stainable iron was common after portacaval shunting, quantitative iron analysis demonstrated increased hepatic iron concentrations in only 6 of 26 shunt patients and 3 of 37 control patients (p = 0.14, chi 2 test). Quantitative hepatic iron analysis demonstrated in this study that portacaval shunt patients have far less accumulated iron than hemochromatosis patients and are unlikely to have tissue injury resulting from iron overload.

The antibody to the metal-binding low molecular weight protein metallothionein (MT) stains preferentially the proliferative edge of epithelial tumors in paraffin sections. The present report demonstrates its usefulness as an epithelial cell marker in DNA flow cytometry of archival specimens. Nine control breast (mammoplasty) specimens, 10 control colonic specimens (resection edges), 12 adenocarcinomas of breast, and 13 adenocarcinomas of colon were analyzed by DNA flow cytometry after MT and DNA staining. The average percentage of cells stained by MT ranged from 12% to 27% in these groups of specimens, which contain epithelial as well as stromal and inflammatory cells. Comparing cell turnover, measured as S-phase fraction (SPF) in unstained and MT-stained preparations, it was 10% and 20%, respectively, in control tissues and 10% and 30%, respectively, in adenocarcinomas. The SPF is lower in unstained preparations because of dilution by noncycling inflammatory and stromal cells. Immunohistochemical staining of various tissues for MT showed specific staining of epithelial cells. Evaluation of aneuploid malignant epithelial cells detected in six breast and eight colonic adenocarcinomas showed that on average, 47% of cells were stained with MT and that their SPF increased by about 50% when MT staining was compared with the unstained preparations. The results suggest that MT stains epithelial cells adequately for ploidy and cell cycle evaluation and that it may stain preferentially the proliferating cell compartment, which is considered to be an index of malignancy.

DNA flow cytometry measurements were performed using cytokeratin as a second parameter to identify epithelial cells selectively in fresh and in archival paraffin samples of normal and adenocarcinoma tissues from breast and colon. Fresh specimens consisted of 22 adenocarcinomas of breast, 20 adenocarcinomas of colon, 16 control breast samples, and 13 control colon samples. Paraffin block specimens consisted of 22 adenocarcinomas of breast (the same as fresh samples), 20 adenocarcinomas of colon (the same as fresh samples), 37 control breast samples and 34 control colon samples. The average proportion of cytokeratin-positive cells per group ranged from 31 to 55% for fresh samples and from 14 to 34% for paraffin samples. For aneuploid cells populations of adenocarcinomas, which consist only of epithelial cells, the average percentage of cytokeratin-positive cells ranged from 60 to 72%. The technique gave satisfactory measurements of ploidy and of cell cycle data in both types of samples. Cell cycle measurements were less accurate than ploidy measurements in both types of samples, and multiple sampling will be required for adequate accuracy. The average S-phase fraction of cytokeratin-positive cells ranged from 6 to 15% for fresh specimens and from 11 to 20% for paraffin samples. Similar data were obtained for the proliferative index (G1 + S + G2 + M phases). The coefficients of variation were smaller for proliferative index than for S-phase fraction data, indicating greater accuracy. Paraffin data give higher cycling cell measurements than corresponding fresh data, so separate standardization of measurements may be required for fresh and for paraffin data.

A total of ten of 60 nonneoplastic lesions and eight of 76 benign neoplasms of skin and breast showed an aneuploid peak in DNA flow cytometry profiles obtained from archival paraffin blocks. This confirms previous similar scattered reports and emphasizes that caution needs to be exercised in interpreting aneuploidy in DNA flow cytometry to mean preneoplasia, neoplasia, or malignancy.

DNA flow cytometry was performed on 44 paraffin blocks from 16 specimens of 11 lesions in five patients from two families identified as probably having the dominant gene of the Lynch Type II syndrome. All but one specimen showed diploidy, although two such specimens were metastases and one a recurrence. The one aneuploid gastric biopsy was followed by a resection with diploidy in all 13 blocks from the malignancy and may represent a superficial change without clinical significance. The prolonged survival after discovery of malignancy in such families may be explained in part by diploidy of the lesions.

An iron loaded liver from a 40 year old man with occult haemochromatosis was transplanted into a 19 year old woman with acute liver failure secondary to a paracetamol overdose. Increased parenchymal hepatic iron was found in a liver specimen at biopsy undertaken because of mild rejection 30 days after transplantation. After transplantation the patient had two episodes of liver rejection confirmed by biopsy. The hepatic iron concentration fell from 161 mumol/g on day 30 after transplant to 26.5 mumol/g (normal less than 40) on day 210. Iron absorption, measured 45 days after transplant, was in the normal range at 12.4%. The rapid fall in hepatic iron and the normal iron absorption study result suggest that the genetic defect of haemochromatosis is not exclusively an intrahepatic defect.

Ploidy of neoplastic large-bowel cells in lesions from which biopsies showed only adenoma tissue by histopathologic examination was investigated retrospectively by DNA flow cytometry. Biopsy and resected-specimen blocks gave interpretable ploidy profiles from 83 lesions. The expectation that adenomas will show diploidy in biopsies and that adenocarcinomas will show aneuploidy in biopsies was fulfilled in 80% of the lesions. In addition, four lesions with diploid biopsies were polyps with minimal invasion. Thus, diploidy would have correctly predicted that local resection was adequate treatment for 62 lesions and conversely that cancer resection was the appropriate treatment for 8 lesions, for an overall predictive success rate of 87%. In conjunction with endoscopic, radiologic and other data, DNA flow cytometry of biopsies from apparently benign large-bowel lesions could become a useful addition to therapeutic decision-making.

Since an intestinal absorptive interaction between iron and zinc has been described in animals and humans, the possibility of increased accumulation of zinc as well as iron in the liver was studied in patients with hereditary hemochromatosis. Hepatic zinc was determined by atomic absorption spectrophotometry in liver biopsy specimens from 21 homozygotes for hemochromatosis, 21 normal liver samples from autopsies, and 15 cases of cirrhosis unrelated to iron overload. Mean hepatic zinc concentrations in the three groups were compared by one-way analysis of variance. Hemochromatosis patients had hepatic iron determinations by atomic absorption spectrophotometry, and iron absorption studies using 59Fe and total body counting had been previously documented in 18 of the 21 hemochromatosis patients. The mean hepatic zinc was significantly increased at 25.9 +/- 26.7 mumol/g (dry weight) in the hemochromatosis patients, as compared to 4.99 +/- 1.51 mumol/g in the control patients (p less than 0.05), and 2.13 +/- 1.13 mumol/g in the cirrhosis patients without iron overload (p less than 0.05). Hepatic zinc concentration was elevated in hemochromatosis patients who had either normal histology, fibrosis, and cirrhosis. Hepatic zinc concentration was not directly related to patient age, hepatic iron concentration, or iron absorption. In conclusion, hepatic zinc was increased approximately fivefold in patients with hemochromatosis. This finding suggests the concomitant hepatic accumulation of zinc as well as iron in this disorder, possibly by means of increased intestinal absorption of zinc and hepatic sequestration.

Granulomatous hepatitis is a common cause of fever of unknown origin in up to 13% of patients with prolonged fever. Attempts to define an exact etiology of the granulomatous hepatitis frequently does not yield a precise diagnosis, so that the physician must consider empiric treatment. In this paper we retrospectively review 23 patients in whom granulomatous hepatitis was found as part of the initial assessment of fever of unknown origin, and we report on their outcomes after an overall prospective follow-up of 37 months. In 26% a precise diagnosis was established at the time of assessment: Q-fever in three, mycobacterial disease in two, and histoplasmosis in one. In the remaining 74% no etiology was established after 44 months follow-up. Forty-one percent of the idiopathic group resolved spontaneously without therapy, and 18% received short-term prednisone or indomethacin with a favourable outcome. The remaining 41% required long-term prednisone therapy for a mean of 33.1 months, but all have remained afebrile and otherwise healthy after 59.6 months follow-up. We conclude that patients with fever of unknown origin who are diagnosed as having idiopathic granulomatous hepatitis have an excellent prognosis, even the minority who require long-term corticosteroids.

The relationship between iron absorption and hepatic iron was studied in 21 patients with hemochromatosis. Iron absorption was studied using 59Fe and total body counting and hepatic iron was measured by atomic absorption spectrophotometry. Iron absorption was inversely related to hepatic iron concentration (r = -0.51, p = 0.009) in this patient population. This observation suggests that iron absorption is regulated by body iron stores even in hemochromatosis, and does not support the hypothesis that the primary metabolic defect in hemochromatosis is a deregulation of iron absorption in relation to iron stores.

Eighteen large bowel adenomas with invasion of the head or the stalk were removed by endoscopic polypectomy or by segmental resection at University Hospital, London, Ontario between 1973 and 1982. All were completely removed by histological criteria. All the patients were traced for an average follow-up period of 4.6 years. None had developed disseminated large bowel cancer. Adding these results to those in the literature, it appears that, provided there is not a high degree of anaplasia of the tumor or lymphatic or venous invasion, endoscopic polypectomy is adequate therapy for such adenomas. An endoscopic recheck of the site of removal after approximately 2 months may be worthwhile as a few local recurrences of benign tumor were reported, although not in the present series. The patients should be followed for life.

The records of the rectal biopsy diagnoses of ulcerative colitis and Crohn's disease in the Department of Pathology, St Mark's Hospital, London, were reviewed. The biopsy diagnoses were compared to subsequent resection diagnoses on the same patients, and annual and seasonal variations in the frequency of these and related diagnoses were studied. The accuracy rate for the biopsy diagnosis of ulcerative colitis was about 70% and for Crohn's disease about 40% each time a biopsy was read. The low figure for the accuracy rate for Crohn's disease could be attributed to sampling error inherent in the diagnosis of a disease which is essentially patchy, showing discontinuous pathology. Also, many patients with Crohn's disease have a normal rectum which is biopsied to demonstrate the distinction from ulcerative colitis. In practical terms therefore a 40% accuracy rate in Crohn's disease is probably adequate. The rate of "false-positive" diagnoses was about 5%. There was a seasonal variation in the frequency of these two diagnoses, but no variation attributable to changes in observers, as pathology trainees in the Department change regularly. The frequency of diagnoses of non-specific inflammation and of normal colon did show such non-random variations.

A single treatment of adult animals with the potent carcinogen NMU was known to induce tumours in a wide variety of organs, with the notable exception of liver. Administration of NMU after partial hepatectomy gave rise to the first liver cell adenomata ever observed in rats due to this carcinogen. The tumours were induced when NMU was given during the period of increased DNA synthesis but not when given early in the pre-replicative period. Although tumours were induced in other organs, the incidence of these did not correlate with the timing of NMU administration. It is suggested that replication of damaged DNA may be a relevant event in carcinogenesis. ImagesFig.p507-a

A temperature-sensitive mutant of Moloney murine leukemia virus defective in an early function and injected into newborn mice produced lower limb paralysis. Susceptible mice were inbred strains CFW/D, CBA/H, C3H/Bi/Ka, and outbred NIH Swiss stock. Inbred W/Fu rats and C57BL/Ka mice did not develop the paralysis, though the latter were infected with virus; the sera from these mice produced paralysis in susceptible CFW mice.

Primary inbred Swiss mouse embryo cells treated with doses of methylnitrosourea which prevent their proliferation for up to 4 days show no change in their G1 and G2 cell cycle phases. They are released from the developing monolayer to a lesser extent than untreated cells. Their DNA synthesis is depressed soon after treatment and for up to 3 days. The results indicate that methylnitrosourea treatment delays the progression of mouse embryo cells in tissue culture through the cell cycle by prolonging the S phase without killing the cells.

Epidermal tumors were induced in Swiss female mice by a topical application of 9,10-dimethyl-1,2-benzanthracene solution followed by repeated applications of croton oil solutions. Fourteen benign and malignant tumors were sampled 25 weeks after the treatment had begun, fixed in osmium tetroxide or potassium permanganate, and embedded in Epon. Sections stained with lead hydroxide were examined. In three tumors defects of the epidermal basement membrane were seen. These defects accompanied local invasion of the tumors. The possible mechanisms of the development of this unusual anatomical situation are discussed.

Induction of tumours in intact and partially hepatectomized rats with ethyl methanesulphonate

Specific-pathogen-free inbred C57BL/Cbi mice (adult virgin females) were given single sublethal doses of N-methyl-N-nitrosourea or N-ethyl-N-nitrosourea and were studied for a lifetime for the development of thymomas. A fatal lymphocytic lymphoma with a "starry sky" pattern due to the presence of macrophages was induced in the thymuses of treated mice within 250 days of treatment. Control and low-dose treatment groups had up to 70% incidence of a histiocytic lymphoma that was usually primary in mesenteric lymph nodes and nearly always occurred later than 250 days after treatment. A "one-hit" linear relationship existed between the time of appearance of induced thymic lymphomas and the log fraction of non-tumor-bearing mice. The absolute latency period of these tumors was constant and independent of dose. The effect of dose was an exponential increase of the total incidence of induced thymic lymphomas. By mathematical analysis, the best estimate of the exponent from the results was 2 or 3, indicating that the development of these induced tumors may be produced by 2 or 3 "events" in the target cell. Possible candidates for these events are premutagenic alkylation of DNA, inactivation of DNA repair, oncovirus activation, regenerative hyperplasia, development of trisomy No 15, and inhibited immunosurveillance.

Wilson's disease is a rare inherited disorder of copper (Cu) metabolism characterized by the deposition of Cu in the liver, brain, and cornea. The levels of metallothionein (MT), Cu, and zinc (Zn) in the livers of two Wilson's disease patients were analyzed in this study. About 50-fold increase in the Cu levels above normal controls was observed in both patients (160 and 298 micrograms/g of wet tissue). About 73% of subcellular Cu was present in the cytoplasmic fraction and most of it was in association with MT. Analysis of hepatic MT levels showed a 3-fold increase (863 micrograms/g of wet tissue) over control human levels (321 micrograms/g of wet tissue). The two forms of MT (MT-I and MT-II) were isolated from one liver sample. Both forms contained high amounts of Cu (11 to 12 g atoms/mole), indicating saturation of MT which had only 2 to 3 g atoms of zinc. The distribution of MT in the hepatocytes was investigated using an immunohistochemical method. In tissue sections with minimal tissue damage, there was intense cytoplasmic staining for MT in hepatocytes whereas both nuclear and cytoplasmic staining was found in tissue sections with extensive necrosis and fibrosis. These results suggest that MT is the major hepatic Cu-binding protein in Wilson's disease, that it is present in a form saturated with Cu, and that only in degenerating hepatocytes is it found in the nucleus as well as the cytoplasm.

The value of tests for the detection of body iron overload was investigated in 8 patients with clinically manifest primary hemochromatosis, 12 patients with cirrhosis and iron overload and 20 patients with liver disease and low or normal iron stores. Iron overload was defined as the presence of stainable iron in more than 50% of hepatocytes in a liver biopsy specimen. The percentages of patients with a true-positive (abnormal) or true-negative (normal) result were: serum iron concentration 65%, transferin saturation 85%, serum ferritin concentration 78%, serum ferritin:serum glutamic oxaloacetic transaminase (SGOT) index 78%, percent iron absorption 58%, percent iron absorption in relation to serum ferritin concentration 80% and percent iron absorption in relation to serum ferritin:SGOT index 93%. The calculated predictive value of a normal test result for the exclusion of iron overload in patients with liver disease, a group with an assumed prevalence of iron overload of 10%, was 98% to 99% for transferrin saturation and serum ferritin concentration used alone and 100% for these measures used together; the predictive value of an abnormal result for the diagnosis of iron overload was less than 50% for all of the above measures used alone or in combination. Hence, in patients with an increased serum ferritin concentration or transferrin saturation, or both, determination of the hepatocellular iron content of a specimen from a percutaneous liver biopsy is required for the diagnosis of iron overload.

Eighteen cases of adenoid cystic carcinoma of the minor salivary glands are reviewed. Noteworthy in the history is the report of pain at the site of the lesion which radiates elsewhere, or of numbers or tingling in its area. Radiation therapy is as able to control the primary as local surgery. Involvement of a much wider field than is required to treat the primary may control the perineural spread common to this tumor and avoid the massive procedures necessary to cure it by surgical means. Metastases to the lung, bone, and brain by venous spread can probably be avoided only by early diagnosis.

1-(2-Hydroxyethyl)-1-nitrosourea (HNU) was prepared by the action of nitrosyl chloride on (2-hydroxyethyl)urea. Attempts to synthesize HNU by an earlier described method were unsuccessful and led to the formation of the cyclized derivative 1-nitroso-2-oxazolidone. In addition, the spectral data that we obtained for HNU differed from those reported earlier. Female C57BL/Cbl mice were treated with single ip doses of HNU to determine its median lethal dose (LD50) and its ability to induce lymphocytic thymic lymphomas in these mice. The results showed that the LD50 was the same as that for 1-ethyl-1-nitrosourea (ENU) and that its was slightly more potent than ENU as a carcinogen in this system.

Concern for the late formation of strictures at the site of an endoscopic sphincterotomy has delayed the acceptance of this procedure as treatment for choledocholithiasis in otherwise healthy patients. The authors addressed this issue by comparing the biochemical and histologic sequelae of sphincterotomy in 23 dogs with those in 10 sham-operated controls. Twenty-four hours after sphincterotomy, hemorrhagic, edematous mucosa surrounded the incision. Microscopically, there was an acute inflammatory exudate bridging the mucosal surfaces. Mucosal regeneration was sufficient after 1 week to cover the defect caused by the cautery, although some inflammatory changes were still evident. A widely patent sphincterotomy orifice was seen in 15 dogs followed up for 10 weeks. In three dogs, the fibres of the papilla had reunited below the incision, resulting in a choledochoduodenal fistula. Histologically, complete healing of the mucosal surface had occurred with no evidence of scar formation or chronic inflammation. Serum bilirubin and liver enzyme measurements did not show evidence of biliary obstruction due to the sphincterotomy. From the results of our study, there is no evidence to suggest that an endoscopic sphincterotomy is predisposed to late stenosis.

A single dose (0.8 mmole/kg) of N-methyl-N-nitrosourea (MNUA) causes significantly more chromosome damage in the bone marrow of mice than a dose of equal toxicity to the animals, (1.1 mmole/kg) of methyl methanesulphonate (MMS) 6, 24 and 48 h after treatment. At these doses both agents alkylate bone-marrow DNA to similar extents, but only MNUA induces thymic lymphomata. The greater chromosome-damaging effects of MNUA are ascribed to the known differences in the pattern of DNA alkylation by each agent, in particular the much higher levels of O-6 methylguanine and phosphotriesters produced by MNUA. The greater chromosome-damaging effect of MNUA may account for its higher toxicity to the bone marrow which in turn may be a significant factor in the induction of thymomata. The enhancement by caffeine of chromosome damage seen particularly 48 h after MMS-treatment suggests that post-replication repair protects cells from the effects of DNA-methylation in vivo.

The methylating agents methyl methanesulphonate (MMS) and N-methyl N'-nitro-N-nitrosoguanidine (MNNG), administered by single i.p. injection in mice failed to yield thymic lymphoma at doses around 60% of the LD50 values, in contrast to MNUA which gives a high yield of tumours by this route. Comparison of the tissue distribution and mode of DNA methylation by these agents showed a positive correlation with ability to methylate the 0-6 atom of guanine in DNA of the target tissues thymus and bone marrow and tumorigeneis. MMS gave a low yield of this product due to its relatively low Sn1 reactivity but was able to methylate DNA extensively at other sites in the target tissues and other organs examined. MNNG despite its ability to methylate 0-6 of guanine in DNA in vitro to the same relative extent as the potent carcinogen MNUA, methylated DNA of thymus and bone marrow to a very small extent in vivo but was able to methylate DNA in certain other tissues nearer the site of i.p. injection. These findings contrast with the general relatively extensive methylation of 0-6 of guanine in DNA of the target tissues and other organs by N-methyl-N-nitrosourea (MNUA).