Remodelling of cerebrospinal fluid lipoproteins after subarachnoid hemorrhage.

Lipoprotein particles (Lps) in normal human cerebrospinal fluid (CSF) are distinct from those found in plasma and include unique apolipoprotein E (apoE indicates protein; APOE, gene) containing lipoproteins rarely seen in human plasma. Less favourable neurological recovery after subarachnoid hemorrhage (SAH) has been observed in patients who possess the APOE epsilon4 allele raising the possibility that apoE influences neuronal survival after brain injury. We analysed Lps from control and SAH CSF testing the hypotheses that following brain injury CSF Lps undergo remodelling and apoE containing Lps are selectively depleted from brain injury CSF. Lipoproteins were fractionated using CSF from six control pools and six patients with SAH on a sepharose 6HR 10/30 size exclusion column. Fractions were assayed for total cholesterol (TC), free cholesterol (FC), phospholipid, triglyceride (TG), apoE, apolipoprotein B (apoB), and apolipoprotein AI (apoAI). Compared to control CSF there were significant (P<0.05) increases in TC, FC, TG, and apoAI in SAH CSF. Plasma sized apoB-containing lipoproteins and a very small apoAI-containing Lps were identified in the SAH CSF, which were not present in controls. However, despite the release of plasma lipoproteins into the subarachnoid space, there was no significant increase in CSF apoE. These data provide novel indirect evidence suggesting that after SAH CSF Lps undergo remodelling and apoE containing Lps are selectively reduced in brain injury CSF. The remodelling of CSF Lps and selective reduction of apoE containing lipoproteins may reflect an important response of the human brain to injury.

Platelet aggregatory responses to low-dose collagen are maintained in hirudin-anticoagulated whole blood for 24 h when stored at room temperature.

Whole blood from 15 volunteers was anticoagulated with hirudin (200U/l) and the response to a known submaximal concentration of collagen (0.6 microg/ml) was tested by impedance aggregometry. In 8 volunteers platelet counts were also taken before and after the maximum aggregatory response. These tests were repeated when the samples had rested for 24 h at room temperature. The median [interquartile range] aggregatory response immediately after sampling was 17.3 [16.7-18.4] ohms. At 24 h it was 17.7 [15.8-19.3] ohms (p = 0.88) although variance was increased (p = 0.006). The immediate platelet count before collagen exposure was 438 [381-510] x 10(9)/l and 258 [227-297] x 10(9)/l post-collagen. At 24 h the platelet count was 448 [443-473] x 10(9)/l (p = 0.224 versus immediate count) but variance was not increased (p = 0.215). After full aggregation the count fell to 284 [234-304] x 10(9)/l (p = 0.592 versus early post-collagen). Variances were similar (p = 0.558). Aggregate response ratios increased non-significantly after 24 h from 0.59 [0.53-0.62] to 0.64 [0.51-0.68] although variance was increased (p = 0.021). Full macroaggregatory responses by impedance aggregometry were seen after 24h storage of whole blood with hirudin at room temperature. This suggests both that distant assessment of platelet function using a standardized method is possible and a potential role of thrombin inhibition for platelet storage.

Quantitative detection of platelet aggregates in whole blood without fixation.

Platelet counting detects lesser degrees of platelet aggregation than conventional aggregometry. In order to prevent progressive platelet aggregation or disaggregation after sampling it is customary to fix blood samples. However fixation may introduce other artefacts. We first compared stability of platelet counts in EDTA-, citrate- and r-hirudin-anticoagulated blood from healthy volunteers. Second, the stability of platelet counts in unfixed EDTA- and hirudin-anticoagulated blood was compared with glutaraldehyde-fixed blood in the same anticoagulants. Third, the effect of in vivo heparin administration on platelet counts in EDTA- and hirudin-anticoagulated blood was studied. Platelet counts within 2 h of collection were significantly higher in EDTA- than in hirudin- or citrate-anticoagulated blood (P = 0.002 vs. hirudin and P = 0.001 vs. citrate). Twenty-four hour counts in hirudin and EDTA were unchanged (P = 0.3 and P = 0.2, respectively, vs. earlier counts). Counts in citrate increased significantly (P = 0.007; n = 10). Platelet counts in fixed blood did not differ significantly from those in unfixed blood. Heparin administered for cardiopulmonary bypass reduced platelet counts in hirudin-anticoagulated blood from (mean +/- 1 standard deviation) 180 +/- 45 to 162 +/- 30 x 10(9) l-1 (P = 0.01; n = 14), without significantly lowering counts with EDTA-anticoagulation, consistent with increased platelet aggregation. Hirudin and EDTA provided stable platelet counts, suggesting that fixation is unnecessary.

Heparin-induced platelet dysfunction and cardiopulmonary bypass.

BACKGROUND: Cardiopulmonary bypass is associated with impaired platelet macroaggregation. Heparin contributes to platelet dysfunction before extracorporeal circulation. In vitro heparinization of whole blood does not impair macroaggregation. Heparin releases several endothelial proteins; thus heparin may inhibit macroaggregation indirectly. METHODS: Patients undergoing operations using cardiopulmonary bypass and ABO blood group compatible volunteers were studied. Whole blood impedance aggregometry assessed macroaggregation in response to collagen (0.6 microg ml(-1)) in blood diluted either with normal saline or with platelet poor plasma, obtained from patients at different stages of cardiopulmonary bypass. RESULTS: Before heparinization, blood diluted with its own platelet poor plasma recorded an impedance change of 13.0 (4.7 to 15.6) Ohms. Platelet poor plasma obtained after heparinization or during extracorporeal circulation reduced this response to 3.7 (1.1 to 8.4) and 2.0 (1.1 to 3.3) Ohms, respectively (both p < 0.0001 versus pre-heparin; n = 13). Macroaggregation in blood from volunteers was similarly inhibited by patients' platelet poor plasma (n = 30). The macroaggregatory response in blood sampled after heparinization for cardiopulmonary bypass, decreased gradually from 11.4 (8.2 to 15.9) Ohms immediately after sampling to 1.7 (1.4 to 4.1) Ohms 2 hours later (p < 0.0001; n = 11). CONCLUSIONS: In vivo heparinization induces plasma changes that inhibit platelet macroaggregation. This is an indirect, delayed inhibition that is transferable in vitro to normal platelets.

Role of genotypic analysis of the thymidine kinase gene of herpes simplex virus for determination of neurovirulence and resistance to acyclovir.

Mutations in the thymidine kinase (TK) gene of herpes simplex virus (HSV) have been associated with resistance to acyclovir (ACY) and possible recognition of neurotropic strains. We sequenced a 335-bp segment of the TK gene to determine the frequency of mutations in HSV strains recovered from dermal, genital, and cerebrospinal fluid (CSF) specimens (n = 200; 102 HSV type 1 [HSV-1] 98 HSV-2 strains). Four polymorphic sites were detected in HSV-1 strains; C513T, A528G, C575T, and C672T. Among the polymorphisms, only C575T resulted in a change of amino acid sequence (residue 192, Ala-->Val). For HSV-2 strains, only one polymorphism (G420T) which resulted in an amino acid substitution (residue 139, Leu-->Phe) was detected. Phenotypic determination of resistance to ACY by a plaque reduction assay of 48 HSV isolates was not correlated with the sequence results of 11 strains in that 7 of these with genotypic polymorphisms were susceptible to the drug in vitro. In addition, of 32 ACY-resistant HSV strains, 28 (87.5%) had no polymorphisms detected in the 335-bp amplicon of the TK gene. There was no statistical difference in the frequency of polymorphisms according to the source of the specimens. We conclude that the detection of nucleic acid polymorphisms in a previously implicated 335-bp segment of the TK gene cannot be interpreted as indicative of either ACY resistance or neurotropism of HSV strains from dermal, genital, and CSF sources.

Diethylcarbamazine-related antimicrobial activity in Mycobacterium tuberculosis-infected blood.

Previous studies have suggested that diethylcarbamazine (DEC), a drug used for filariasis control, may be useful in the treatment of mycobacterial infections. In this experiment, Mycobacterium tuberculosis was added to blood samples from two groups (healthy and diabetic) of adult non-smoking donors. Portions of each sample were tested with and without DEC at clinically achievable levels. Statistically significant DEC-related percentage decreases in BACTEC growth index counts were noted for each group (Wilcoxon one-sample signed-rank tests, alpha = 0.05, two-tailed). These results suggest that administration of DEC for filariasis control could have a positive impact on tuberculosis control.

Enterotoxigenic Escherichia coli detected in foods by PCR and an enzyme-linked oligonucleotide probe.

A polymerase chain reaction (PCR) and an enzyme-linked oligonucleotide probe hybridization assay were developed for the detection of enterotoxigenic Escherichia coli (ETEC) in ground beef, chicken, pork and raw milk. Two synthetic primers, one of which was biotinylated, were used in the PCR to amplify a fragment of the E. coli heat-labile enterotoxin (LT) gene. The identity of the amplified products was confirmed by liquid hybridization using a horseradish peroxidase-linked internal oligonucleotide probe in a 96-well microplate coated with streptavidin. The final quantitation of the PCR products was performed by a colorimetric reaction. Under established conditions (including 1 min at 60 degrees C for primer annealing and extension in PCR cycles), this method detected all 7 LT-producing E. coli pathogenic for humans, but did not detect all 7 LT-positive E. coli of animal origin 3 E. coli strains that do not produce LT, and 9 other bacteria. Under less stringent PCR conditions (55 degrees C for annealing and extension), 2 strains of LT-producing E. coli of porcine origin were detected while the results of other bacterial strains remained unchanged. In pure cultures, the detection limit of the method was 1.4 colony forming units (CFU). Prior to PCR amplification, all food samples inoculated with an LT-producing ETEC, were subjected to enrichment in brain heart infusion broth for 8 h at 37 degrees C. From these cultures, 10 microliters was heated at 95 degrees C for 10 min and directly used in the PCR. An initial inoculum of as few as 1.2 to 12 CFU of the LT-producing ETEC per 25 g (or ml) of food sample gave a positive reaction.

Abilities of OspA proteins from different seroprotective groups of Borrelia burgdorferi to protect hamsters from infection.

The ability of vaccination with recombinant OspA from six seroprotective groups of Borrelia burgdorferi sensu lato to induce protection against infection with homologous and other Lyme spirochetes was examined in hamsters. Antisera generated against the OspA proteins of B. burgdorferi sensu stricto S-1-10 and C-1-11 (seroprotective groups 1 and 2, respectively), Borrelia afzelii BV1 (seroprotective group 4), and Borrelia garinii LV4 (seroprotective group 5) were able to kill the homologous spirochete in vitro but not other isolates. Surprisingly, antisera against B. afzelii PKo (seroprotective group 6) and B. burgdorferi sensu lato LV5 (seroprotective group 3) OspA proteins were unable to kill the homologous organism, although LV5 OspA antisera killed the heterologous isolates S-1-10 and LV4. In vivo vaccination studies supported the in vitro findings, confirming that vaccination with a single OspA protein does not provide complete protection against challenge with all Lyme disease spirochetes. In addition, OspA antibodies from some isolates may not protect against the homologous isolate. The induction of protective antibodies against other B. burgdorferi proteins may be necessary to insure a comprehensive Lyme disease vaccine.

Low efficiency of the 5' nontranslated region of hepatitis A virus RNA in directing cap-independent translation in permissive monkey kidney cells.

To characterize in vivo the translational control elements present in the 5' nontranslated region (5'NTR) of hepatitis A virus (HAV) RNA, we created an HAV-permissive monkey kidney cell line (BT7-H) that stably expresses T7 RNA polymerase and carries out cytoplasmic transcription of uncapped RNA from transfected DNA containing the T7 promoter. The presence of an internal ribosomal entry site (IRES) within the 5'NTR of HAV was confirmed by using BT7-H cells transcribing bicistronic RNAs in which the 5'NTR was placed within the intercistronic space, controlling translation of a downstream reporter protein (bacterial chloramphenicol acetyltransferase). However, translation directed by the 5'NTR in these bicistronic transcripts and in monocistronic T7 transcripts in which the HAV 5'NTR was placed upstream of the chloramphenicol acetyltransferase coding sequence was very inefficient compared with the translation of monocistronic transcripts containing either the IRES of encephalomyocarditis (EMC) virus or a short nonpicornavirus 5' nontranslated leader sequence. A large deletion within the HAV IRES (delta 355-532) eliminated IRES activity in bicistronic transcripts. In contrast, larger deletions within the IRES in monocistronic transcripts (delta 1-354, delta 1-532, delta 1-633, and delta 158-633) resulted in 4- to 14-fold increases in translation. In the latter case, this was most probably due to a shift from IRES-directed translation to translation initiation by 5'-end-dependent scanning. Translation of RNAs containing either the EMC virus IRES or the nonpicornavirus leader was significantly enhanced by cotransfection of the reporter constructs with pEP2A, which directs transcription of RNA containing the EMC virus IRES fused to the poliovirus 2Apro coding region. This 2Apro enhancement of cap-independent translation suggests a greater availability of limiting cellular translation factors following 2Apro-mediated cleavage of the p220 subunit of the eukaryotic initiation factor eIF-4F and subsequent shutdown of 5' cap-dependent translation. In contrast, pEP2A cotransfection resulted in severe inhibition of translation directed by the HAV IRES in either monocistronic or bicistronic transcripts. This inhibition was due to competition from the EMC virus IRES present in pEP-2A transcripts, as well as the expression of proteolytically active 2Apro. 2Apro-mediated suppression of HAV translation was not seen with transcripts containing large deletions in the HAV IRES (delta 158-633, delta 1-532, or delta 1-633). These data suggest that the HAV IRES may have a unique requirement for intact p220 or that it may be dependent on active expression of another cellular translation factor which is normally present in severely limiting quantities.(ABSTRACT TRUNCATED AT 400 WORDS)

Detection of hepatitis A virus in environmental samples by antigen-capture PCR.

The efficacy of the antigen-capture PCR (AC-PCR) method for the detection of hepatitis A virus (HAV) in environmental samples was demonstrated. HAV was captured from a seeded liquid waste or a shellfish sample with homologous antibody and then heat denatured and subjected to reverse transcription and the PCR, all in the same tube. Subsequently, the AC-PCR products were analyzed by oligonucleotide probe hybridization in solution, agarose gel electrophoresis, and autoradiography. The AC-PCR detected as little as 0.053 PFU of cell culture-adapted HAV strain HM175/18f. The results of cDNA-RNA hybridization indicated that the particle/PFU ratio of this virus strain is approximately 79:1. Therefore, the detection limit of the AC-PCR was estimated to be four virus particles. No amplified products were observed when poliovirus 1, coxsackievirus A9, coxsackievirus B3, echovirus 6, reovirus 1, adenovirus type 40, human rotavirus type 1, and bovine enterovirus type 2 were tested, confirming the specificity of the assay. There were no differences between the nucleotide sequences of AC-PCR products of HAV strain HM175/18f and the sequences of wild-type HAV strain HM175 derived from molecularly cloned cDNA. Of 121 waste and shellfish samples tested by both plaque assays (PA) in cell cultures and the AC-PCR, 81 (67%) were positive and 31 (26%) were negative as determined by both methods, whereas 9 (7%) were positive as determined by the AC-PCR and negative as determined by the PA, and none were positive as determined by the PA and negative as determined by the AC-PCR.

Analysis of hepatitis A virus translation in a T7 polymerase-expressing cell line.

Hepatitis A virus (HAV) exhibits several characteristics which distinguish it from other picornaviruses, including slow growth in cell culture even after adaptation, and lack of host-cell protein synthesis shut-down. Like other picornaviruses, HAV contains a long 5' nontranslated region (NTR) incorporating an internal ribosomal entry site (IRES), which directs cap-independent translation. We compared HAV IRES-initiated translation with translation initiated by the structurally similar encephalomyocarditis virus (EMCV) IRES, using plasmids in which each of the 5'NTRs is linked in-frame with the chloramphenicol acetyltransferase (CAT) gene. Translation was assessed in an HAV-permissive cell line which constitutively expresses T7 RNA polymerase and transcribes high levels of uncapped RNA from these plasmids following transfection. RNAs containing the EMCV IRES were efficiently translated in these cells, while those containing the HAV IRES were translated very poorly. Analysis of translation of these RNAs in the presence of poliovirus protein 2A, which shuts down cap-dependent translation, demonstrated that their translation was cap independent. Our results suggest that the HAV IRES may function poorly in these cells, and that inefficient translation may contribute to the exceptionally slow replication cycle characteristic of cell culture-adapted HAV.

Continuous-time temporal back-propagation with adaptable time delays.

Backpropagation is extended to continuous-time feedforward networks with internal, adaptable time delays. The new technique is suitable for parallel hardware implementation, with continuous multidimensional training signals. The resulting networks can be used for signal prediction, signal production, and spatiotemporal pattern recognition tasks. Unlike conventional backpropagation networks, they can easily adapt while performing true signal prediction. Simulation results are presented for networks trained to predict future values of the Mackey-Glass chaotic signal, using its present value as an input. For this application, networks with adaptable delays had less than half the prediction error of networks with fixed delays, and about one-quarter the error of conventional networks. After training, the network can be operated in a signal production configuration, where it autonomously generates a close approximation to the Mackey-Glass signal.

Characterization of the borreliacidal antibody response to Borrelia burgdorferi in humans: a serodiagnostic test.

An in vitro assay was used to characterize the borreliacidal activity of sera from Lyme disease patients. The mean percentage of killing was 23% with sera from patients with a single erythema migrans lesion, 42% from patients with multiple lesions, 58% from patients with Lyme arthritis of short duration, and 83% from patients with Lyme arthritis of long duration. Borreliacidal activity was abrogated when Lyme disease serum was treated with anti-human IgM or IgG1. In addition, human sera from Lyme arthritis patients containing borreliacidal antibody prevented the induction of Lyme arthritis in irradiated hamsters challenged with the Lyme spirochete. Removal of outer surface protein A antibodies from late Lyme disease sera caused reductions in the borreliacidal antibody titer. The results demonstrate an important role for borreliacidal antibody against infection with B. burgdorferi in humans and confirm that detection of borreliacidal antibody in human sera can be a specific serodiagnostic test for Lyme disease.

Mutations within the 5' nontranslated region of hepatitis A virus RNA which enhance replication in BS-C-1 cells.

Passage of human hepatitis A virus (HAV) in cell culture results in attenuation of the virus as well as progressive increases in the efficiency of virus replication in cell culture. Because the presence of identical mutations within the 5' nontranslated regions (5'NTRs) of several independently isolated cell culture-adapted HAV variants suggests that the 5'NTR may play a role in determining this change in virus host range, we constructed chimeric infectious cDNA clones in which portions of the 5'NTR of cell culture-adapted HM175/p35 virus were replaced with cDNA from either wild-type virus (HM175/wt) or a second independently isolated, but closely related cell culture-adapted virus (HM175/p16). Substitution of the complete 5'NTR of HM175/p35 with the 5'NTR of HM175/wt resulted in virus with very small replication foci in continuous African green monkey kidney (BS-C-1) cells, indicating that 5'NTR mutations in HM175/p35 virus are required for optimal growth in these cells. A chimera with the 5'NTR sequence of HM175/p16 retained the large foci of HM175/p35 virus, while the growth properties of other viruses having chimeric 5'NTR sequences indicated that mutations at bases 152 and/or 203 to 207 enhance replication in BS-C-1 cells. These findings were confirmed in one-step growth experiments, which also indicated that radioimmunofocus size is a valid measure of virus replication competence in cell culture. An additional mutation at base 687 of HM175/p16 had only a minor role in enhancing growth. In contrast to their effect in BS-C-1 cells, these 5'NTR mutations did not enhance replication in continuous fetal rhesus monkey kidney (FRhK-4) cells. Thus, mutations at bases 152 and/or 203 to 207 enhance the replication of HAV in a highly host cell-specific fashion.

Immunoglobulin G2 confers protection against Borrelia burgdorferi infection in LSH hamsters.

We showed that immune serum and its immunoglobulin fractions, specifically immunoglobulin G2 (IgG2), could confer complete protection to irradiated hamsters challenged with the Lyme disease spirochete. Immune serum and its immunoglobulin fractions also killed Borrelia burgdorferi in vitro. Depletion of complement in vivo abrogated the ability of IgG2 to confer complete protection against B. burgdorferi. Furthermore, the majority of antibody reactivity directed against B. burgdorferi was found in the IgG2 fraction. These findings demonstrate that IgG2 plays an important role in acquired resistance against infection with B. burgdorferi. Additional studies are needed to determine the mechanism(s) by which B. burgdorferi evades host defenses despite the development of an effective borreliacidal antibody response.

The 5' nontranslated region of hepatitis A virus RNA: secondary structure and elements required for translation in vitro.

Although the lengthy 5' nontranslated regions (5'NTRs) of other picornaviral RNAs form highly ordered structures with important functions in viral translation, little is known about the 5'NTR of hepatitis A virus (HAV). We determined the nearly complete 5'NTR nucleotide sequences of two genetically divergent HAV strains (PA21 and CF53) and included these data in a comparative phylogenetic analysis of the HAV 5'NTR. We identified covariant nucleotide substitutions predictive of conserved secondary structures and used this information to develop a model of the 5'NTR secondary structure, which was further refined by thermodynamic predictions and nuclease digestion experiments. According to this model, the 5'NTR comprises six major structural domains. Domains I and II (bases 1 to 95) contain a 5'-terminal hairpin and two stem-loops followed by a single-stranded and highly variable pyrimidine-rich tract (bases 96 to 154). The remainder of the 5'NTR (domains III to VI, bases 155 to 734) contains several complex stem-loops, one of which may form a pseudoknot, and terminates in a highly conserved region containing an oligopyrimidine tract preceding the putative start codon by 13 bases. To determine which structural elements might function as an internal ribosome entry site, RNA transcripts representing the HAV 5'NTR with progressive 5' deletions were translated in rabbit reticulocyte lysates. The translation product was truncated, unprocessed P1 polyprotein. Removal of the 5'-terminal 354 bases of the 5'NTR had little effect on translation. However, deletion to base 447 slightly decreased translation, while deletion to base 533 almost completely abolished it. These data indicate that sequences 3' of base 355 play an important role in the translation mechanism utilized by genomic-length HAV RNA. Significantly, this region shares several conserved structural features with the internal ribosome entry site element of murine encephalomyocarditis virus.

Genetic variability within the 5' nontranslated region of hepatitis A virus RNA. Implications for secondary structure and function.

The RNA genome of hepatitis A virus (HAV) contains a lengthy and relatively well conserved 5' nontranslated region (5'NTR). In other picornaviruses, the 5'NTR has been shown to have important functions related to the initiation of viral translation and replication of viral RNA, functions which are critically dependent on both primary and secondary RNA structure. We have utilized a phylogenetic approach to construct a model of the secondary structure of the HAV 5'NTR. By comparing the nucleotide sequences of genetically divergent simian and human HAV strains, we identified a series of covariant nucleotide substitutions which are predictive of conserved, double-stranded helical structures within the 5'NTR, and which thus permitted improved thermodynamic modeling of the secondary structure. The model was further refined based on the observed sites of cleavage of synthetic RNA by single- and double-strand specific RNAses. The results of these studies suggest that the 5'NTR of HAV has a general organization similar to that of other picornaviruses, and shares certain structural features and perhaps specific functions with the 5'NTRs of the cardioviruses and aphthoviruses.

Evidence that the gene for herpes simplex virus type 1 DNA polymerase accounts for the capacity of an intertypic recombinant to spread from eye to central nervous system.

HSV-1(17) replicates 100-fold more efficiently than HSV-2(186) within trigeminal ganglia following ocular infection. In order to identify the nucleotide sequences responsible for the differences in the capacity of the two HSV strains to grow within the peripheral nervous system, an intertypic recombinant was generated by infecting neuroblastoma cells with HSV-2(186) and a HSV strain possessing nucleotide sequences from HSV-1(17). The genome of the intertypic recombinant was composed entirely of HSV-2(186) DNA except for 2.0 kb of HSV-1(17) DNA positioned between m.u. 0.413 and 0.426. Following corneal infection of mice, the intertypic recombinant grew to higher titers in both ocular tissues and trigeminal ganglia than did the HSV-2 parent. Most significantly, the intertypic recombinant could spread into the brain from the trigeminal ganglion and kill the host whereas mice inoculated with the HSV-2(186) parent survived infection. The 2.0 kb of HSV-1(17) DNA inserted into the genome of the intertypic recombinant encodes the 5' terminus of the HSV-1 gene for DNA polymerase. Thus, the results suggest that the difference in the capacity of two HSV strains to replicate within the trigeminal ganglion of its host and to spread into the brain is determined by nucleotide sequences within the gene for DNA polymerase.

Nucleotide sequences important in DNA replication are responsible for differences in the capacity of two herpes simplex virus strains to spread from cornea to central nervous system.

Two herpes simplex virus (HSV) intertypic recombinants were isolated with genomes composed entirely of HSV-2(186) nucleotide sequences except for a 6.0 kb segment of HSV-1(17) DNA positioned between 0.40 and 0.44 map units. Following corneal infection of mice, HSV-1(17) and the two intertypic recombinants spread from infected eyes into the central nervous system and induced a fatal encephalitis. Ocular infection with the HSV-2(186) parent did not lead to detectable amounts of virus in the brain, and none of the mice developed encephalitis. The 6.0 kb HSV-1(17) DNA inserted within the genome of the two intertypic recombinants contained nucleotide sequences involved in DNA replication. These include the HSV-1(17) oriL, the HSV-1(17) gene for DNA polymerase and portions of the HSV-1(17) gene coding for DNA-binding protein ICP8. Thus, our results indicate that the difference in the capacity of HSV-1(17) and HSV-2(186) to spread from the cornea into the CNS is determined solely by nucleotide sequences associated with DNA replication.